Saud M. Al-Jufaili for the degree of Doctor of Philosophy... presented on June 12, 2002. Title: Qualitative Analysis of Sardine...

advertisement
AN ABSTRACT OF THE THESIS OF
Saud M. Al-Jufaili for the degree of Doctor of Philosophy in Fisheries Science
presented on June 12, 2002. Title: Qualitative Analysis of Sardine and Anchovy
Oscillations and Implications for the Management of Sardine and Anchovy
Fisheries in Oman.
Abstract approved:
Redacted for privacy
David B. Sampson
Sardines and anchovies are small pelagic fishes that support important commercial
fisheries around the world. This project reviews the inverse cyclic behavior in the
abundance of these two stocks, which is a striking feature in many regions. In
addition, the project used qualitative loop analysis techniques to analyze the
feasible sardine-anchovy model configurations that result in the inverse relationship
between sardines and anchovies. A simple community model was examined that
considers fishing, sardine and anchovy biomass, and food resources for the sardines
and anchovies. First the stability of these model configurations was investigated to
determine the conditions that should be met to stabilize the unstable configurations.
Second, the behavior of the feasible sardine-anchovy model configurations was
examined when fishing was removed from the models. Finally, model
configurations were identified that best represent the sardine-anchovy system in
terms of predicting qualitative changes in the system variables. These best models
define the crucial interactions between sardines and anchovies that require further
studies. Based on the results of the literature review and the ioop analysis a set of
questions was developed and used in interviews with fishers in Oman to investigate
whether the sardines and anchovies in Oman are inversely related. Based on the
survey results and lessons learned from the literature review and loop analysis,
recommendations were developed for further research and management of the
fisheries in Oman for sardines and anchovies. In systems where the sardines and
anchovies vary inversely in abundance refuge areas for the sardines and anchovies
are very important for maintaining the two fish stocks and their cyclic behavior.
The results from the ioop analysis suggested that interactions between sardines and
anchovies (e.g., competition and amensalism) are not important provided the two
fish populations can regulate themselves by means of their refuge areas. The
expansion and contraction of sardines and anchovies is a function of environment
suitability and long-term shifts in the environmental regime. The study found no
evidence that sardines and anchovies in Oman are inversely related.
©Copyright by Saud M. Al-Jufaili
June 12, 2002
All Rights Reserved
Qualitative Analysis of Sardine and Anchovy Oscillations and Implications for the
Management of Sardine and Anchovy Fisheries in Oman
Saud M. Al-Jufaili
A THESIS
Submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Doctor of Philosophy
Presented June 12, 2002
Commencement June 2003
Doctor of Philosophy thesis of Saud M. Al-Jufaili presented on June 12. 2002
APPROVED:
Redacted for privacy
Major Professor, representing Fisheries Science
Redacted for privacy
Head of the Department of Fisheries and Wildlife
Redacted for privacy
Dean of the Graduate School
I understand that my thesis will become part of the management collection of
Oregon State University libraries. My signature below authorizes release of my
thesis to any reader upon request.
Redacted for privacy
Saud M. Al-Jufaili, Author
ACKNOWLEDGMENTS
Alhamdulilah. Thanks to Dr. David B. Sampson, my major advisor, for his help and
patience. I thank Dr. Philippe Rossignol for his help in understanding ioop analysis.
I am thankful to Janet Webster for her assistance in editing this thesis. Thanks to
Coastal Oregon Marine Experiment station for funding me to attend a conference
related to my subject. I would like to thank the Department of Fisheries Science at
Sultan Qaboos University for providing transportation while interviewing the
Omani fishers.
My sincere thanks to my father and my wife for their support and prayers. I
thank all the family and friends that prayed to successfully finish the work.
Alhamdulilah.
TABLE OF CONTENTS
Page
Chapter One: Introduction
1
Introduction ....................................................................................... 1
Problem statement ............................................................................. 3
Research objectives ........................................................................... 3
Chapter Two: Inverse Cyclic Behavior between Sardine
and Anchovy Stocks around the World......................................................... 5
Introduction .......................................................................................
5
The California Sardine and Anchovy Case ....................................... 10
The Japanese Sardine and Anchovy Case ......................................... 26
The Peruvian Sardine and Anchovy Case ......................................... 38
The South African Sardine and Anchovy Case ................................. 48
Synthesis and Discussion .................................................................. 56
Conclusion........................................................................................ 62
Chapter Three: Qualitative Analysis of the Inverse-Cyclic Behavior
between Sardine and Anchovy ...................................................................... 64
Introduction ....................................................................................... 64
Methodology ..................................................................................... 66
Results ............................................................................................... 81
Basic Set 1 and Basic Set 5 ................................................... 82
Pooled Table of Predictions (PTP) ............................ 87
Conditionally stable models ...................................... 92
BasicSet 2 ............................................................................. 94
Pooled Tables of Predictions (PTP) .......................... 95
Conditionally stable models ...................................... 106
Discussion ......................................................................................... 107
Conclusion........................................................................................ 110
TABLE OF CONTENTS (continued)
Chapter Four: Sardine and Anchovy Abundance Fisheries in Oman ........... 112
Introduction ....................................................................................... 112
Oceanographic Conditions in the Waters of Oman ............... 113
Omani Sardines and Anchovies ............................................ 116
Methodology ..................................................................................... 118
Results ............................................................................................... 121
Discussion ......................................................................................... 128
Future of the Omani Sardine and Anchovy
Fishery ................................................................................... 132
Recommendations ............................................................................. 136
Chapter Five: Summary and Conclusions ..................................................... 139
Bibliography .................................................................................................. 143
Appendices .................................................................................................... 160
Appendix A: Summary of the Sardines and Anchovies Cases Studied ........ 161
Appendix B: Terminology ............................................................................ 164
Appendix C: Different calculations involved in analysis of a loop system.. 166
Appendix D: Tables of predictions and their polynomial coefficients
for different Basic Sets ............................................................ 170
Appendix E: Omani Fishers Responses to the Interview Questions
byRegion ................................................................................. 190
LIST OF FIGURES
Figure
Page
2.1
A conceptual model for the inverse cyclical relationship
between sardine (solid) and anchovy abundance .............................. 9
2.2
Ekrnan transport areas off the California coast (shaded areas,
magnitude is proportional to the symbol dimension). Solid
arrows indicate the California Current, no magnitude implied.
An eddy within the Southern California Bight is indicated by the
dottedline .......................................................................................... 12
2.3
Pacific sardine and northern anchovy landings ................................. 13
2.4
The sardine (solid) and anchovy cycle in the California system ...... 16
2.5a
Distribution of California sardine during a low abundance period... 17
2.6a
Distribution of northern anchovy during a high abundance period.. 18
2.5b
Distribution of California sardine during a high abundance period.. 20
2.6b
Distribution of northern anchovy during a low abundance period ... 21
2.7
Sardines and anchovies landings in the Japanese system ................. 28
2.8
Important areas for the Japanese sardine and anchovy ..................... 31
2.9
The sardine (solid) and anchovy cycle in the Japanese system ........ 32
2.10
Japanese sardine expansion areas (light) and its spawning
areas (dark) when the population is at high levels ............................ 34
2.11
Japanese anchovy expansion areas (light) and its spawning
areas (dark) when the population is at high levels ........................... 37
2.12
The Peruvian Current System ........................................................... 40
2.13
The Peruvian sardine and anchovy landings ..................................... 41
2.14
The sardine (solid) and anchovy cycle in the Peruvian Current
System ............................................................................................... 44
2.1 5a Distribution of the Peruvian sardine during low abundance ............. 45
2.15b Distribution of the Peruvian sardine during high abundance ............ 46
2.16
The Benguela upwelling and current system and some localities
mentioned in the text ......................................................................... 50
2.17
Southern African sardine and anchovy landings ............................... 52
LIST OF FIGURES (continued)
Elgure
Page
2.1 8a Distribution of the South African sardine and anchovy
during a period of high abundance .................................................... 54
2.18b Distribution of the South African sardine and anchovy
during a period of low abundance ..................................................... 55
2.19
Conceptual model of sardine (solid) and anchovy (dotted)
cyclic changes in abundance ............................................................. 61
3.1
Example Model Y, where sardines are denoted by S, anchovies
by A, and their shared food resource by FR...................................... 72
3.2
The Basic Sets of sardine-anchovy systems ...................................... 73
3.3
Different possible groups derived from Basic Set 1 ......................... 75
3.4
Different sardine and anchovy model configurations
considered under Basic Set 1, Subset a ............................................. 76
3.5
Different Subsets (A, B, and C) from Basic Set 2 ............................. 78
3.6
Basic Sets 5 and 6 were derived by removing fishing
from Basic Sets 1 and 2 ..................................................................... 79
3.7
Different Subsets (A, B, and C) under Basic Set 6 ........................... 79
3.8
Model configurations from Basic Set 1 that had 100%
agreement with the pooled table of predictions ................................ 90
3.9
Model configurations from Basic Set 5 that still held the inverse
relationship between sardines and anchovies even after the
Fishing effect was removed............................................................... 90
3.10
The model from Basic Set 1 that most reliably predicts
changes in the different variables ...................................................... 91
3.11
The model configurations from Basic Set 2, Subset A;
that had 100% agreement with the pooled table of
predictions(Table 3.6a) ..................................................................... 98
LIST OF FIGURES (continued)
Figure
Page
3.12
Model configurations from Basic Set 6, Subset A that
maintained the inverse relationship between sardines and
anchovies when the Fishing effect was removed .............................. 99
3.13
The model from Basic Set 1, Subset A that most plausibly
predicts changes in the different variables ........................................ 100
3.14
The models from Basic Set 2, Subset B that had 100%
agreement with the pooled table of predictions (Table 3.6b) ........... 101
3.15
Model configurations from Basic Set 6, Subset B that
maintained the inverse relationship between sardines and
anchovies when the fishing effect was removed ............................... 102
3.16
The most reliable model from Basic Set 2, Subset B ........................ 103
3.17
The models from Basic Set 2, Subset C that had 100%
agreement with the pooled table of predictions (Table 3.6c) ............ 104
3.18
The simplest model from Basic Set 2, Subset C ............................... 105
3.19
Conditionally stable models from Basic Set 2 .................................. 107
4.1
Map of the different regions in Oman and other localities
mentioned in the text ......................................................................... 115
4.2
Mean monthly air temperature in different regions in
Oman calculated from 1977-2000 ..................................................... 117
4.3
Western India total sardines and anchovies landings ........................ 129
LIST OF TABLES
Table
3.1
Different possible interactions between sardine (5) and
anchovy (A) populations and their loop analysis representations ..... 68
3.2
Illustrative table of predictions .......................................................... 72
3.3a
Summary table of results for all the models analyzed
with fishing included ......................................................................... 83
3 .3b
Summary table of results for all the models analyzed
with fishing excluded ........................................................................ 86
3.4
The pooled table of predictions for the Basic Set 1 models shows
the highest percentage of agreement among the model
configurations that produced alternations between sardines and
anchovies ........................................................................................... 88
3.5
Conditions for stabilizing the conditionally stable
models from Basic Set 1 ................................................................... 93
3.6
Tables 3.6a, 3.6b, and 3.6c. The pooled table of predictions
for the models in Basic Set 2, Subset A, B, and C ............................ 96
3.7
The pooled table of predictions for the Basic Set 6,
Subset C models shows the percentage of agreement among the
model configurations that produced alternations between sardines
and anchovies .................................................................................... 104
3.8
The pooled table of predictions for the models in Basic Set 2,
all Subsets combined ......................................................................... 105
4.1
The questions used in interviews with the fishers in Oman .............. 119
4.2
Geographic distribution of the number of fishers
interviewed along the Omani coast ................................................... 123
QUALITATIVE ANALYSIS OF SARDINE AND ANCHOVY
OSCILLATIONS AND IMPLICATIONS FOR THE MANAGEMENT OF
THE SARDINE AND ANCHOVY FISHERIES IN OMAN
CHAPTER ONE: INTRODUCTION
INTRODUCTION
The many cases around the world of important fisheries relying on small
shoaling pelagic fish species have captured the interest of hundreds of fisheries
scientist, and managers, e.g., Pacific sardine, Japanese sardine, California sardine
and anchovy, South African sardine, and the Peruvian anchovy (Liuch-Belda et al.
1992, Murphy 1988, Csirke 1988, Kawasaki 1983). The management of these
fisheries is often limited by uncertainties due to poor understanding of stock
expansions and contractions (Shuntov and Vasilkov 1981). For example, the
Pacific sardine
(Sardinops sagax)
stock sustained a fishery in British Columbia for
nearly 20 years but collapsed in the mid 1940s before new catches in 1950s began
to be reported (Hargreaves et al. 1994, Shuntov and Vasilkov 1981) (see also
Bentley et al. 1996 for the U.S. West Coast sardine population). Fluctuations in the
Japanese sardine (Sardinops melanosticus) also have been observed over the years
(Watanabe et al. 1995, Watanabe et al. 1996; Coppola et al. 1994). In many cases
the collapse of the sardine stocks were accompanied by a build up of anchovies. In
the case of the California Current system it has been shown that this cyclic
behavior has been occurring for the past 1500 years (Baumgartner et al. 1996).
Soutar and Isaacs (1969) first described this when they discovered layered deposits
of sardine-anchovy scales in anaerobic sediments off Santa Barbara, California.
Variation in the number of scales in the sediment indicated periodic changes in the
relative abundance of sardines and anchovies, with sardines most abundant when
anchovies were least abundant, and vice versa.
2
The Sultanate of Oman, capital Muscat, is one of the developing countries
in the Middle East that has a rich fishery resource when compared with its
neighbors.
The Omani coastline is estimated to be 1700 km with valuable
resources like abalone, lobsters, tunas, kingfish, small pelagic fishes and many
other exploited stocks that substantially contribute to the total income of the
country. Although a few of these fisheries are managed with closed seasons, for
instance the lobster and the abalone fisheries, Oman has also a large number of
marine species, known to be valuable overseas that are not exploited, for example,
sea urchins, sea cucumber, squids, and crabs. Most, if not all, Omani fishermen use
artisanal fishing methods and only operate near shore and never exploit offshore
stocks.
Fisheries for small pelagic fishes, mainly sardines, are among the Omani
fisheries that are not yet managed. There is a need to manage this fishery because
of its importance in the food web and its important contribution to the country's
annual income. In the Muscat region in 1994, excessive landings of sardines
(mostly by artisanal fishermen) were worth 4.2 million Omani Rials (OR), roughly
10.8 million US dollars (Omani Annual Statistical Report 1995), which was 45% of
the total fisheries revenue reported by Muscat region fisheries. Sardine fishermen
in Oman use mostly traditional beach and purse seines, which require little effort
but are rarely deployed far from the beach. Sardines are dried in the sun and used as
fertilizer, as food for cattle and humans, or as bait. Anchovies are utilized similarly
as sardines and are caught mostly by the same fishermen, with the same gear. So
far, Oman has had large catches of sardines but not of anchovies. No studies have
been conducted yet to estimate the biomass of this fish stock and no one has
carefully studied the landings of this fishery. It remains unknown whether the
Omani sardine and anchovy fisheries behave the same as the sardine-anchovy
fisheries elsewhere.
3
PROBLEM STATEMENT
The sardine and anchovy stocks often exhibit long term alternating
fluctuations in abundance. The causes of these fluctuations and details of the
processes involved are not well understood. There are many difficulties in studying
sardine-anchovy systems, especially in countries like Oman where the fisheries are
not well developed. How can the Omani sardine-anchovy fishery in particular and
other similar fisheries benefit from the lessons learned from sardine-anchovy
fisheries that have been more thoroughly monitored and studied? Are the sardines
and anchovies stocks in Oman inversely related? What precautions should be taken
to carefully study and manage such fisheries? These are some of the questions
explored in this thesis.
RESEARCH OBJECTIVES
The main objective of the second chapter of this thesis is to review the
inverse cyclic behavior in the abundance of sardines and anchovies. Four sardine
and anchovy systems were examined for this purpose: occurring in California,
Japan, Peru, and South Africa. The review focuses on identifying the processes
involved in expansions and contractions of the two species, the causes of the
fluctuations, and finally, what changes occur with the sardines and anchovies when
they are at high levels versus when they are at low levels.
Using qualitative loop analysis techniques in Chapter Three I identify
feasible sardine-anchovy model configurations, that result in the inverse
relationship between sardine and anchovy, based on a simple community model
that considers fishing, sardine and anchovy biomass, and sardine-anchovy food
resources. I investigate the stability of these model configurations and detennine
the conditions that should be met to stabilize the unstable model configurations.
Second I determine whether the behavior of the feasible sardine-anchovy model
ru
configurations is altered when fishing is removed from the model. Based on the
two analyses I identify the model configurations that best represents the sardine-
anchovy system in terms of predicting change in the system variables. These
models define the crucial interactions between sardines and anchovies that require
further studies.
In Chapter Four I use the results from earlier chapters and survey of Omani
fishers to investigate whether the sardines and anchovies in Oman are inversely
related or not. Based on the results I develop recommendations about how to
proceed with research, management, and development of fisheries for sardines and
anchovies in Oman.
5
CHAPTER TWO: INVERSECYCLIC BEHAVIOUR BETWEEN SARDINE
AND ANCHOVY STOCKS AROUND THE WORLD
INTRODUCTION
Sardines and anchovies, widely used for human consumption, fertilizer and
animal feed, exist in inshore regions with active upwelling systems. Sardine and
anchovy stocks appear in high volumes and colonize a wide geographic range
along the shore. New fisheries are established and old ones revived as stocks
fluctuate. While both commercial and traditional fishers use purse and beach seines
to catch these fish, the near shore distribution of sardines and anchovies supports
more traditional fishers than commercial fishers. This is mostly true with traditional
in the developing countries where fishermen do not have enough money to upgrade
their fishing gear. On the other hand, in developed countries commercial fishers can
benefit more. Because the sardines and anchovies are easy to catch and can be used
in many ways, and because sardines and anchovies stay for a long time before they
disappear, fishers become dependent on the sardines or anchovies. This dependence
leads to the development of new fishing technology such as the use of airplanes to
locate the fish schools, thus reducing the searching time and increasing the fishing
power. Given this dependence, the collapse of a stock usually means the collapse
of the coastal fishing villages as well, and the eventual closing of the fish
processing firms and other businesses.
Sardine and anchovy stocks have been fluctuating in abundance in a cyclic
manner for a very long time. The cyclic alternations in sardine and anchovy
abundance are not limited to one area but have been observed worldwide. The
fluctuations in sardines and anchovies are inversely related; when the sardines are
abundant, the anchovies tend to be at low abundance, and vice versa. Because the
stock collapse or contraction can occur rapidly, the fishers and fish processing
firms often do not have enough time to react and overcome the situation. It is not
clear what triggers sardine or anchovy stocks to increase in abundance and then
contract. The timing of expansion and contraction, the height of the fish stocks
peak, and the lengths of its disappearance are all unpredictable.
The literature suggests various reasons for the phenomenon of alternating
fluctuations between the two fishes. Scientists are divided over whether the
fluctuations are due to heavy fishing, changes in the environment, or both. The
average time for a complete cycle of sardine and anchovy appearance and
disappearance is 30-34 years. Fishing is controlled by market supply and demand
functions, and cannot trigger such long-term phenomena. Excessive fishing and
market supply and demand functions could act to cause temporary fluctuations in
the stocks of the two fish, but management will typically interfere and try to
stabilize the system.
Many scientists have tried to understand the reasons behind the fluctuation
in the sardines and anchovies stocks. One of the accepted hypotheses for the
fluctuation for example is that the fish stock is heavily overfished and as a result it
goes through several recruitment failures and cause it eventually to disappear. The
fluctuations however are so significant in terms of size and duration that heavy
fishing alone cannot explain them (Clark and Man 1955 and Murphy 1960).
Studies of the sardine and anchovy scale deposition in anaerobic sediments indicate
that both populations have fluctuated in an inverse cyclic manner, repeatedly
contracting and expanding for at least 1700 years, long before the existence of any
fishery (Souter and Issacs 1974, Baumgartner et al 1992). This study provides
strong evidence that natural phenomena, rather than human activities cause the
alternations between sardines and anchovies, at least for the California sardine and
anchovy stocks. Radovich (1982) speculated that shift of the Pacific sardines from
fishing grounds might reflect the fluctuation in the landings. This justification
ignores the long-term fluctuation of sardines and does not include anchovies as
well.
Perhaps the most accepted hypothesis for the sardine expansion and
contraction is given by Liuch-Belda et al (1991 a). The hypothesis states that the
Pacific sardines during warm periods expand to include wider geographic ranges.
7
During cold years sardines retreat to major spawning areas. The hypothesis does
not discuss what triggers the cold versus warm periods and does not relate
anchovies to it.
The cyclic behavior implies that both sardines and anchovies do not
disappear completely. For the cyclic behavior to continue, the contracted species
must continue to exist, but in lower abundance from which it expands to overcome
the current dominant species. As coastal species living particularly in upwelling
zones where the habitat is suitable, their migration is limited to the near shore. The
Japanese sardine, for example, completes its whole life cycle within the coastal
zones. Migration is along the coastline from south to north for feeding and
spawning purposes. The Pacific sardines and northern anchovies are reported also
to migrate along the Pacific shoreline to spawn (Lulch-Belda et al 1989). This life
history feature is inconsistent with the idea that sardines and anchovies migrate far
offshore during periods of apparent low abundance, but then emigrate from
offshore waters when they become abundant again.
The current study, however, is reviewing these studies and relates sardine
and anchovy fluctuations. The objective of this review is to support the author's
proposed hypotheses or theory behind the inverse cyclic behavior between sardines
and anchovies. When a stock contracts, the fish must remain somewhere in the
near shore in a refuge area where it can safely spawn and maintain itself far from
the dominant species. This refuge area is defined as a suitable habitat with optimal
food and environmental conditions where the contracted fish stock can be found all
the year around and persist in spite of the high abundance of the alternate fish.
An event, a natural phenomenon gradually or suddenly occurring, triggers
the contracted fish to expand and the expanded fish to contract. It is important to
understand and study the inverse relationship between sardines and anchovies so
that we can better manage and monitor the fisheries on these species. One way to
understand the sardines and anchovies is to analyze the phenomenon that triggers
their fluctuations.
From a conceptual model for sardine-anchovy alternations, I hypothesize
that when a sardine population contracts the fish retreat to one or more refuge areas
where the stock persists when it is at low abundance (Figure 2.1). An event (Xl)
occurs causing the sardines to expand to high levels and then to spread from their
refuge areas to colonize more areas. Anchovies are affected negatively by this
event (Xl) and are forced to contract to their refuge areas along the coast when
they are at low abundance. For the anchovies to expand again to high levels, event
(Xl) has to relax or another event (X2) has to occur. This event (the relaxation of
event (Xl) or event (X2)) affects the sardines negatively causing them to contract
back to their refuge areas (low levels).
The current chapter reviews four cases of sardine and anchovy fish stocks
fisheries around the world: California, Japan, Peru, and South Africa. My objective
is to compare and contrast the conceptual model hypothesized above with the stock
fluctuations observed in these chosen cases. The review is partly based on sardine
and anchovy landings as indices of abundance. Although fishery landings are a
crude index, their magnitude of change is so enormous that there is little doubt that
most of the change in landings reflects the change in real abundance (Liuch-Belda
etal 1991a).
High abundance levels
Low abundance levels
(Refuge areas)
r
Expansion
I
I
Contraction
ri
I
Relaxation of
Event Xl or
occurrence of
Event X2
Event Xi
Expansion
+
High abundance levels
-.
4
Low abundance levels
(Refuge areas)
Figure 2.1. A conceptual model for the inverse cyclical relationship between
sardine (solid) and anchovy abundance. An event is defined as a natural
phenomenon gradually or suddenly occurs.
10
THE CALIFORNIA SARDINE AND ANCHOVY CASE
The California coast is dominated by the California Current which is the
eastern extension of the clockwise flowing North Pacific Subtropical Gyre. The
California Current flows with mean speed of 10 cm
generally from the
s1
Columbia River to central Baja California (MacCall, 1986, Hickey 1998). The
California Current is associated with a pattern of wind-driven Ekman transport
directed offshore. Ekman transport is the net mass displacement of water from one
area to another, caused by wind blowing steadily over the surface; the net mass
transport is 90° to the right (in the Northern Hemisphere) of the wind's direction
(Tomczak and Godfrey 1994). The California current upwelling system, caused by
northerly flowing winds from April to September, occurs at the margin of
continental shelf, which is narrow off California, 20 km, while wider off British
Colombia, 46 km (Ware 1992). Major upwelling areas off California are centered
on Cape Mendocino (Bakun et al 1974, Ware 1992) (Figure 2.2). The coastal
upwelling and wind indices off California are 0.36
respectively (Cury et a! 1998).
m3.s1.m1
and 654
m3.s1,
The California Current is also associated with
warm eddies that occur in Southern California Bight, south of Point Conception,
called also the Southern California Counter Current (MacCall 1986). When
upwelling decreases in the fall, the northward-flowing California Undercurrent
(Davidson Current) surfaces and carries warm equatorial water inshore (Favorite et
al., 1976; Bottom et aL, 1993).
Two sardine species occur in the California Current system, the Pacific
sardine (Sardinops sagax) and a sub-species, California sardine (Sardinops sagax
caerulea), sometimes referred to as Sardinops caerulea
.
However, the two
sardines are often considered to be the same species, with the Pacific sardines and
California sardines being synonyms. In the literature some papers refer to
Sardinops caerulea as the Pacific sardine (as in Wolf 1992, among others). In the
FAO synopsis of the Clupeoid species of the world, Sardinops caerulea is indicated
11
to be distributed in the in the Pacific northeast while Sardinops sagax is said to be
in the southeast Pacific. I found that Sardinops sagax is more often used in the
literature of the northeast Pacific. Ahistrom (1960) indicated that S. sagax, and S.
caerulea are so similar that it is doubtful that they are distinct species. In reviewing
the literature I made sure that the species mentioned, whether it is S. sagax, S.
caeruleus, or S. sagax caeruleus, refers to the species in the northeast Pacific. Only
one species of anchovy exists in this system, the Northern anchovy (Engraulis
mordax) (Whitehead 1985).
From 1916 to the late 1 940s the pacific sardine dominated the fishery in this
system (Figure 2.3). Anchovies during this period were in very low in abundance
but did not vanish completely. Anchovies replaced the sardines from the late 1 940s
to early 1950s. During the period 1950 to 1958 there were landings of both sardines
and anchovies with higher catches recorded for sardines. From 1965 to 1993,
anchovies dominated the landings. Over the history of the California fishery
landings of sardines were recorded to fluctuate more than landings of anchovies
(Barnes et a! 1992). Currently the sardines dominate the system.
Sardines spawn over a much wider water temperature range (13-25 °C) than
anchovies (11.5-16.5 °C, Liuch-Belda et al 1991b). Generally, sardines spawn
between January and June (Ahistrom 1954). Based on egg and larval distributions
of anchovies many studies agree that the northern anchovy spawn in the southern
California Bight where it is found all the year around. Anchovies spawn in this
Bight between December and April (Moser et al. 1993). Sardines are also found to
use Southern California Bight for spawning but their concentration is not as high as
anchovies (Hernandez\Tazquez 1994). A clear reason why anchovies choose the
Southern California Bight to spawn all the year around is not given. However, there
is speculation that it is a function of temperature and upwelling intensity. A study
by Lluch-Belda et al. (1991b) of the California Current concluded that sardines are
eurythermic compared to anchovies but spawn in intermediate values of upwelling.
Anchovies are concluded to be stenothermic and spawn in wider upwelling values
especially at high and low values.
--
'VanCO;Ifld
-1I
I
P
I
\
I
\
I
450
j
12
1
I
I
i
1
1
P
1
I
I
Co1umbi River
I
1
j
I
I
I
I
I- ----------I--f- ------- I- ------------ I- ------------ I- ------- -1
I
I
P
P
I
I
P
I
P
I
I
400
350
iii
-----------SouThern
iiii_
<.
I
I
I
I
30°
Bidht
---I
4
I
I
O
I
I
I
4
0
I
I
4
0
O
1
1
1
0
1
I
I
I- ---------- 1- ----------i____.4
Current
I
il-ill
Calitoiiiia
4
4
/L
I
I
250
.-
130°
125°
120°
-
I
I
1150
110°
--
W
Figure 2.2. Ekman transport areas off the California coast (shaded areas,
magnitude is proportional to the symbol dimension). Solid arrows indicate the
California Current, no magnitude implied. An eddy within the Southern
California Bight is indicated by the dotted line. Redrawn from Bakun (1993).
13
PDO
700
Warm Phase
Cool Phase
Warm Phase
600
15oo
400
/
300
/
I\i
200
I
1oo
/
0
-
N
O
-
O
C\
-
-
C
-
N
C'
-
O
O
-
00
*
*
Year
Pacific Sardine
Northern Anchovy
Figure 2.3: Pacific sardine and northern anchovy landings. Data obtained
FISHSTAT Plus: Universal software for fishery statistical time series.
from
Version 2.3 2000.
In general global climate change causes changes in the sea surface
temperature and in the intensity of upwelling (Bakun 1990). Low levels of
upwelling were found to limit the abundance of sardine and very high levels limit
the abundance as well due to turbulence and instability of the water column (Lluch-
Belda et a! 1991a), On the other hand, anchovies' seasonal peak spawning success
is associated with high levels of upwelling, which results in lower temperatures
because cold bottom water is brought to the surface.
Many papers indicate that low abundance of California sardine occurs
during low temperature periods. These low temperature periods are when the
Northern anchovy are at high abundance (Hauto-Sobranis and Lluch-Belda 1987).
14
Temperature changes were found to coincide with shifts in the Pacific
sardine landings in California. A warming period starting in 1915 peaked in late
1930s and declined during the 1940s to early 1950s. A temporal warming began
again during the early 1 950s to early 60s, and temperatures decreased after that
until the mid 70s (Lulch-Belda 1991a, Shantov and Vasilkove 1982, Barnes et al
1992). Warm sea surface temperatures during the 1990s resulted in the sardine
stock re-occupying feeding grounds in central California, Oregon, Washington and
British Colombia (Schwartz!ose et a! 1999). Luich-Belda et al (1992) examined
data on global sea surface temperature, California air surface temperature,
California sea surface temperature, and the California sardine landings. He
concluded that these variables are positively correlated. The coherence in the
environmental conditions causing warm conditions or cold conditions is sometimes
called a regime shift. A regime shift is a variation in modes of atmospheric pressure
resulting from long-term changes (multi-decadal, 10-70 years) in patterns of
physical phenomena, hypothesized as variation in solar radiation and changes in the
North Pacific circulation that affect air-sea exchange of heat and momentum with
the Aleutian Low atmospheric pressure pattern (Minobe 1999, Hare and Mantua
2000, Schumacher 1999). The Pacific Decada! Oscillation (PDO) is thought to be a
measure regime shifts. Warm phases of the PDO occur during 1925-1947 and
1977-1990s, While a cool phase of the PDO occurs during 1947-1976. The warm
phase PDO is associated with increases in sardines and decreases in the anchovy.
While cool PDO was associated with increase in sardines and increase in the
anchovy (Figure 2.3) (Hare et al 1999, Francis and Hare 1994, Francis et al 1998).
Cold versus warm regimes coincide with alternations in the anchovy and
sardine landings. Cold regimes were reported for 1915-1930 and 1950-1970, while
warm regimes were reported for 1930-1950 (with a warm peaks in the 1940s) and
mid 1970s- 1980s (Lulch-Belda et al 1992). Klyashtorine and Smimov (1995) used
time series data for the annual sea surface air temperature anomalies of the
Northern Hemisphere and correlated the anomalies with the California sardine and
15
salmon abundance. Sardines were reported to respond more quickly than salmon to
the change in the environment.
First of all it is important to indicate that when the Pacific sardine or the
northern anchovy contract, they do not disappear completely. Rather, with declines
in abundance they contract their distribution, which becomes more limited to
certain areas along the pacific coast. When the California sardine is at low
abundance, their distribution is mainly limited to the Punta Eugenia region off Baja
California (Figure 2.4 and 2.5a). The Southern California Bight appears sometimes
to be a refuge area, as does San Francisco Bay. However, because the water
temperatures in these regions are not always suitable for sardine spawning, Punta
Eugenia remains the major refuge area for the California sardine (Liuch-Belda et a!
1992, Ahistrom 1960, Moser et al 1993). The sardines remain in this area as long as
they are at lower abundance than the anchovies. During this time period, the
Northern anchovy is distributed along the entire northern Pacific coast
(Schwartzlose et al 1999, Luich-Belda et al 1989) (Figure 2.4 and 2.6a). Anchovy
eggs and larvae can be found in the Punta Eugenia area as well but at lower
concentrations when compared with the sardines (Hernandez-Vazquez 1994). The
California Cooperative Oceanic Fisheries Investigations (Ca1COFI) plankton
surveys revealed that there are places in the California current where both sardines
and anchovy eggs and larvae are found together. There are places however where
the anchovy larvae outnumber the sardine larvae even during periods of high
sardine abundance (Ahistrom 1967).
16
Most of the Pacific
coast, British Columbia,
and parts of Canada
Punta Eugenia
Expansion
Contraction
Event
Expansion
Contraction
Event
(Coolin)
+
Most of the Pacific
coast, British Columbia,
and parts of Canada
Southern
California Bight
Figure 2.4. The sardine (solid) and anchovy cycle in the California system.
17
N
Co1umbiatRiver
45o
a--i --------
a-
a- --------- a- -------- a-
I
I
[Cape Mencdcino
I
a
a
I
a
a
(
i--
----------------- I ---------
I
a
a
I
a
a
a
i'SaniFrancisco
ç Poiit Conception
a
a
a
350
.ça1iforiia Bight
k
\ .I_J__J_
I
I --------- I ---------- I ------300
KCalifrnia
a
I
-
25°
1300
125°
120°
1150
110° W
Figure 2.5a. Distribution of California sardine during a low abundance period
(after Liuch-Belda et al. 1989).
18
------------------- 1
N
I
I
I
I
450
1CapeMendcino
L
I
I ---------------I
I
I
I
I
I
Li
.: .?
}
I
I
I
--------- I --------- I -------
I
I
I
I
4
I
I
I
4.'."-
San Francisco
''&Poiit Conception
350
I
F
*
I
4 ------ .'*-4 -------------------
I
-:oIt1ern
Bight
TZ_
I
I
I
F
F
I
I-
300
I
I
4 --------- I
--------
I
F
4_
I
I-- - F ------
!i
I
25°
-
4
I
aja California
*
I
I
1300
4
4
*
i
a
I
I
F
P
4
4
I
*
1250
120°
P
115°
1100
W
Figure 2.6a. Distribution of northern anchovy during a high abundance period
(after Liuch-Belda et al. 1989).
The sardines maintain themselves at this low level and not do expand except when
the environment allows. The size of the population could be a function of the size
of the refuge area, which is a function of the suitable environmental conditions
preferred by the sardines.
Prior to sardine expansion in 1994, most of the biomass was in the youngest
year classes and maturation was recorded to occur at less than 12 months (Butler et
al 1996). This indicates that sardines were experiencing very good conditions and
were ready to recover. When conditions were ripe for the sardine stock to expand, it
colonized most of the Pacific coast, including the waters off British Colombia
(Figure 2.4 and 2.5b) (Schwartzlose et al 1999), while the anchovies contracted to
their refuge areas in the Southern California Bight (Figure 2.4 and 2.6b).
Cury (1988) suggests that when the population is at low levels, rapid
changes can occur in the genome because of genetic drift, with the resulting
organism being better adapted to compete with the dominant population.
20
N
44o1umbia River
450
I
I
I
I
I
I
I
I
P
I
I
I
I
1
I
I
I
I
I
I
---------.j-J-- ------ I- --------- 1-
J1
I
I
1
I
---------- I-
Cape Mendoino
I
----------------------------------
400
I
}
i
I
i
anIrancisco
oin Conception
350
:.:ahforni Bight
3Øo
L
L\4'2t
L'aIjlflia
------------------- L -----
Pinta Eugenia
Pacifr Ocean
-
I
25°
r
--------------------------------
130°
125°
120°
115°
110° W
Figure 2.5b. Distribution of California sardine during a high abundance
period (after Lluch-Belda et al. 1989).
21
- ------
N
Cohimbia1River
450
I
-I-
--------- '"I' ------------------
I
I
I- --------- F------
11
I
1.1
II
I
I
1
1
1
I
I
I
CapeMendcino
I
I
J
I
I
i
I
I
I
I
I
I
I
(,.
r
....
---------- ----------------r
Sanl'rancisco
PtConcepfionL
35o
I
l
I
Southern
C aIilo-nia Bight
-
30°
F
PLntaEugenia
25°
I
at Ca hit )ffl W
1300
125°
120°
115°
c
'b
1100
"''
W
Figure 2.6b. Distribution of northern anchovy during a low abundance period.
The concentration area is the anchovy spawning area as referenced in the text
(Moser et al. 1993).
22
I consider increased water temperature to be the primary variable that
triggers the sardines to expand from their core or refuge areas and the anchovies to
contract to their core or refuge area.
The differences between sardines and
anchovies in their temperature preferences for spawning support the hypothesis that
regime shifts cause cyclic behavior in the sardine and anchovy abundance. LiuchBelda et a! (1992) found that warm local temperatures, global surface temperature
and California sardine abundance all follow the same pattern. Cold versus warm
regimes have coincided with the reversals in the sardine and anchovy landings.
Cold regimes occurred in 1915-1930, and in 1950-1970. Warm regimes, on the
other hand, occurred in 1930-1950 (peaked in 1940) and in 1970s to 1980s
(Klyashtorin and Smirnov 1 995).Warming of the sea surface water coupled with
appropriate levels of upwelling cause the sardines to increase their survival rates
and expand. During these periods the conditions were not good for the anchovies to
remain at high levels. Warmer conditions bring larger tropical predators to the
Pacific coast, which move northward along the coast and increase the mortality of
anchovies and hasten the population!s contraction (Smith 1995). High fishing
mortality on anchovies at this time reduces the adult population, further
contributing to the contraction of the anchovy population.
Sardine abundance seems to coincide with small scale environmental
changes as well. Sardine abundance was reported to positively correspond to 2-year
and 5-year sea level and sea surface temperature cycles (Hauto-Soberanis and
Liuch-Belda 1987), and these occur during the El Nino events off coastal California
when temperatures are elevated and sea level rises (Lenarz et a! 1995). El Nino
events result in intermediate levels of upwelling that normally favor sardine
reproduction (Lluch-Belda et al 1991). El Nino is characterized by a large scale
weakening of the trade winds and warming of the surface layers in the eastern and
central equatorial Pacific. It is believed that El Nino events occur at 2 to 10 year
intervals (Glantz 1999). El Nino is accompanied by an interannual see-saw in
tropical sea level pressure between the eastern and western hemispheres, called the
23
Southern Oscillation (SO). The strength of the Southern Oscillation is measured in
terms of the difference in surface pressure between Darwin, Australia and Easter
Island. The index therefore is called Southern Oscillation Index (SOT). Low values
of the Southern Oscillation Index are associated with El Nino events while stronger
values are associated with opposite conditions called La Nina events (Glantz and
Thompson 1981).
The expansion of the California sardine population is reported to have been
from south to north (Luch-Belda 1991b, Schwartzlose et al 1999). The further north
the California sardine population expands, the farther away they are from areas
with temperatures that are optimal for spawning. Off the Oregon coast the 14°C
isotherm was found to form a distinct boundary for spawning of Pacific sardine
(Bentley et al 1996). The sardines then have to migrate south to over winter and
spawn (Schwartzlose et al. 1999). The sardines also expand into offshore waters
but when they do so are limited by food availability (Lo et al. 1996). In addition to
Punta Eugenia off central Baja California, Point Conception off southern California
becomes a major sardine spawning area during warm periods (Liuch-Belda et al.
1991a).
As the sardines become more abundant, food availability becomes
problematic and the competition for food within the population increases (Lo et al.
1996). Food availability is particularly critical for larval survival. Therefore, the
less food available the lower the survival rate (Lasker and MacCall 1983). Due to
crowding of the sardines when they are at high biomass levels, there is less space
and food available, which causes the sardines' growth rate and size at age to decline
(Deriso et al. 1996). The slower growth causes individuals to be more vulnerable
to predation. Cannibalism, adult fish feeding on eggs and larvae, at high biomass
levels will also tend to increase (Hunter and Kimbrell 1980).
Recruitment is
reported to decrease when the population is at high levels as well (Tyler and
Gallucci 1980, Sheperd 1982). When the anchovy population increases, the
mortality rates of sardine larvae increase as well (Butler 1991, 1987), which
suggests the two species may compete for food and space.
24
It is important to note that the sardine population grows very rapidly during
the period of population growth. Sardines will expand to occupy all the suitable
habitats. When sardines fill all the possible habitats and cannot expand further they
are at the peak of their abundance. At this point only adverse negative
environmental conditions and heavy fishing can cause sardines to contract.
Within the process of large fluctuations between the sardine and anchovies,
there are short-term variations in landings associated with the rise and fall of both
fishes. These small variations, especially drops in landings, have always caused
people to wonder whether they signal changes in the long-term trends. These
short-term variations complicate our understanding of the interplay between sardine
and anchovies and the environment. Temporary reversals in trend as the sardine
expand or contract could be due to any number of different transitory factors.
Predation by migrating Jumbo squid (Dosidus gigas), for example, may have partly
contributed to the drop in the sardine landings in the Gulf of California in 1981
(Ehrhardt 1991).
Whatever it is that causes the sardine to contract or expand, the change
eventually shows up as recruitment success or failure. During the California sardine
decline in the 1950s
,
Clark and Man (1955) reported very poor 1949 and 1950
year-classes as the cause of the population reduction (see also Smith and Moser
1988, Radovich 1982).
As the California sardine or the Northern anchovy stocks increased, the
California fishing industry grew as well and the fisherst experience of where to
locate the fish improved. Fishing technology and methods also improved. Modern
technology was used to locate the sardine schools, combining spotter planes with
vessels equipped with side-scan sonar (Cisneros-Mata et al. 1995).
The
combination of heavy fishing and un-favorable environmental conditions caused a
major reduction in abundance.
In the l930s and 1940s, sardine abundance
decreased when exploitation increased beyond 20% annually and sea surface
temperatures decreased below historical records, 16.9 °C (Barnes et a! 1992).
25
As the sardines and anchovies fluctuate, the fisheries management also
changes trying to adjust the quotas accordingly. The moratorium banning fishing
for sardines as long as their spawning biomass was less than 20,000 tons was
established in 1974 and in effect from 1974-85. It successftilly, but slowly helped
in rebuilding the sardine stocks. The quotas on sardines remained the same (1000
tons) from 1986 to 1990, and increased carefully and steadily after that as the
sardine spawning biomass increased. During this period, the spawning area also
increased steadily from 670 to 3840 n.mi2 (Wolf 1992).
Sardines in California generally receive a higher price than anchovies.
Sardines contain a larger edible portion than anchovies and are suitable for pet food
and reduction into fishmeal. Anchovies are only suitable for reduction. Sardine oil
also can be used in paint and in margarine (Butler 1987). During 1982-89, prices
paid by the reduction firms for Northern anchovies were $29-48
ton1
while the
prices for Pacific sardines were $150-200 ton1 (Jacobson and Thomson 1993). In
1967 when sardines were scarce and caught only as incidental catch, they were sold
as dead bait to the lucrative market in central California for $200-400 per ton (Wolf
1992). The demand for sardines continued but the supply was limited due to low
abundance.
When the sardine fishery collapsed in the mid 195 Os, the sardine fishers had
to look for alternative means of employment. Some fishers switched to fishing for
market crab and others to tuna and salmon. The number of fishers, however, had to
decline because fewer were required to fish for market crab, salmon, and albacore
than for sardines. The surplus fishers had to leave the fishing industry and look for
other employment. A number of fishers shifted to the anchovy fishery since the
anchovy population was growing at that time. Other fishers with large vessels
moved to Alaska for the king crab fishery (Paralithides camtschatica). Others sold
their vessels for a loss and transferred to other fisheries. Many of the vessels and
equipment involved in sardine canning and reduction were sold to Peru, South
Africa, and Chile. Some of the experts in this field, including managers and
scientists, use their experience with the California sardine and in connection with
26
managing the South America sardine and anchovy fishery through the United
Nations' support of Peru's Instituto del Mar (IMARPE) (Ueber and MacCall 1990).
THE JAPANESE SARDINE AND ANCHOVY CASE
The Japanese coastal system is characterized by the Kuroshio and Oyashio
Currents. The Kuroshio Current originates east of the Philippines and flows north
to through Japan. It has a width of 100 km and speed of 2-4 knots. The sea surface
temperature over the Kuroshio Current averaged from 27-24°C over 20 years
(Sawara, 1974). Kuroshio Current takes two paths: a straight path and a meander
path. The straight path is close to the Japanese coast while the meander is far from
it. The meander path includes offshore waters and loops back toward Japan and
rejoins the typical straight path (Kawai 1998). The standing crop of phytoplankton
in the Kuroshio area is between 0.07 mg.m3 to 0.3 mg.m3 chlorophyll and the total
amount of chlorophyll in the euphotic zone is about 30 mg.m2 to 50 mg.m2. While
the primary production is 50 mg C.m2.d' to 100 mg C.m2.d' for the pelagic area,
and 300 mg C.m2.d' to 500 mg C.m2.d' for the coastal waters (Ichimura 1965).
An extension of the Kuroshio current flows in to the Japan Sea via the Korean strait
forming another current called the Tsushima current (Tomczak and Godfrey 1994).
The Oyashio current on the other hand is the western limb of the sub- polargyre. It
is a cold current that flows southward along northern Japan. Surface temperatures
of Oyashio water range from about 0°C in early spring to 20°C in the summer
(Tomczak and Godfrey 1994).
There is one species of sardine and one of anchovy in the Japanese fishery,
Sardinops melanostictus and Engraulis japonicus. The landings of Japanese
sardines have fluctuated much more than the landings of Japanese anchovies
(Fankoshi 1992) with sardine peaks reported in 1633-1660, 1673-1725, 1817-1843,
1858-1882 (Kikuch 1958). Schwartzlose et al. 1999 reported additional peaks in
1920-1945, and 1975-1995. Like the California sardine and anchovy, the Japanese
27
sardine and anchovy apparently do not exist together in high abundance (Figure
2.7). It is clear from the sardines and anchovy landings in Japan that the landings
of sardines at peak period are very large, unlike the anchovy's landings.
Japanese anchovy dominated the Japanese system from 1954 to 1973. The
landings (Figure 2.7) show very little change in the anchovies landings when
compared with the sardines landings, which suggests that the increase in the
anchovies in 1954 to 1973 may have been a response to the decrease in the sardine
abundance. The Japanese sardine started to increase in 1972, reaching its peak
period during 1983-89. The Japanese sardine decreased after that giving the chance
to the anchovies to revive in 1997. Fluctuations in the landings of Japanese
anchovies are not obvious like those with the California anchovies. In fact, the
Japanese anchovy was reported to be stable and fluctuated only three to four times
in narrow ranges during the period 1906-1984 (Funakoshi 1992).
Japanese sardines spawn around the Kuroshio Current, which transports the
eggs to the nursery grounds. Meandering of the Kuroshio Current depresses the
production of food particles for the sardine and anchovy larvae causing high
mortality rates in the larvae (Nakata et al 1994) (Figure 2.8). The Kuroshio took a
meander path in 1934; as a result, a large-scale cold water mass intruded into the
route of drifting sardine eggs and larvae. The shift in the Kuroshio Current caused
food shortages for larval sardines and resulted in starvation of the larvae
population. This resulted in poor recruitment to the adult population and the
collapse of the sardine population (Shantov and Vasilkov 1982).
28
5000
4500
4000
3500
3000
2500
2000
Warm Phase
o1 Phase
PDO
1500
1000
CID
500
0
I
I
III
I
Lt
&r
00
I
C
00
If
00
00
Q
00
(
O
V
O
00
Q
- - - - - * - - - Year
L
Japanese sardine -
Japanese anchovy
Figure 2.7. Sardines and anchovies landings in the Japanese system. Data
obtained from FISHSTAT Plus: Universal software for fishery statistical time
series. Version 2.3 2000. The solid line on the top of the graph indicates the
start and the end of the warm phase of the Pacific Decadal Oscilation (PDO),
while the dashed line indicates the cool phase (Hare et al. 1999, Francis and
Hare 1994, Francis et al. 1998).
Generally, the sardines spawn in the area of mixing between the waters of
the Kuroshio Current and the coastal water masses. The main spawning seasons are
from February to May during the years of low abundance, but they spawn in
February and March during high abundance (Nakai and Hattori 1962). The area
south of Kyushu Island is considered to be the major spawning area (Shantov and
Vasilkov 1982). The eggs and larva are then transported by the Kuroshio Current to
the eastern waters of Honshu Island (Nakai 1962) (Figure 2.8). The eggs hatch two
or three days after release at average water temperature between 17-18 °C.
Recently hatched larvae utilize the yolk as their source of nutrition for three days
before starting independent feeding. The post-larvae feed on copepod eggs, nauplii,
29
and copepodites (Kondo 1980). The progeny remain near Honshu until they reach
sexual maturity at two years old and then migrate back to the spawning areas south
of Kyushu.
The south flowing Oyashio Current is another important oceanographic
feature for the sardines and anchovies in Japan. This high salinity, nutrient rich
current contains highly abundant phytoplankton that are available for feeding the
immature and adult sardine. The area of mixing between the Oyashio and Kuroshio
currents is very important in the life history of the sardines (Figure 2.8).
Anchovies are always found along the coastline, within the coastal side of
the warm Kuroshio Current and the Tsushima Current waters, and are not found in
the open ocean. They spawn in water temperatures of 12 and 30 °C in northern
waters and in water temperature of 18 °C in the southern waters. The Japanese
anchovy is recorded to spawn between March and December. The spawning
temperature depends on the oceanic region and season, while the spawning areas
are scattered on the Pacific ocean side and on the Sea of Japan side, and in the Seto
Inland Sea of the (Funakoshi 1992).
When the stock of Japanese sardine is at low levels of biomass, the fish are
confined to a small coastal area along western and southern Japan (Schwartzlose et
al 1999, Kawasaki 1991). Spawning grounds during periods of low abundance are
also limited to the Pacific coastal waters (Watanabe and Kuroki 1997). In general
terms, the Japanese sardine during periods of low abundance is a coastal species
that wanders inshore along southern Japan. During these periods the fish grow
faster and mature at younger ages (Kawasaki 1993). Kawasaki added that, at low
populations, sardines direct more energy to growth and maintenance for self
preservation rather than reproduction. The Japanese sardine, when at low levels, is
confined to the south-eastern coast of Kyushu, Shikoku and Honshu, and spawning
is restricted to an even smaller area (Figures 2.9 and 2.10) (Liuch-Belda et al.
1989). These areas are considered to be the refuge areas or core areas for the
Japanese sardine. When the Japanese sardine population is at low levels, fish
condition and the proportions of lipid in the body are good, which results in adults
30
that produce healthy offspring, i.e. better adapted to survive and compete with the
existence other offspring (Kawasaki and Omori 1995). Healthy offspring are
advantageously prepared to take over the system from the dominant species
(anchovy). However, this replacement process cannot occur unless there are
favorable environmental conditions.
Global atmospheric warming intensifies wind stress over the oceans,
leading to coastal upwelling or oceanic turbulence (Bakun 1990). These changes in
oceanic circulation result in a rise in the marine productivity. Sardines react
sensitively to high primary productivity and their biomass increases rapidly. The
sardine, by its nature a coastal species, but can easily utilize the oceanic waters
when environmental conditions permit. High productivity in the Oyashio and
Kuroshio waters in addition to warm temperatures enhances the expansion of the
Japanese sardine to oceanic waters (Watanabe and Wada 1997). The Japanese
sardine expanded widely to include the coastal range of South Sakalin in the USSR,
the Japan Sea side of the Korean Peninsula, and the seas surrounding the Japanese
islands (Figure 2.9 and 2.10) (Kondo 1980, Lluch-Belda et al 1989). Japanese
sardine expansion and contraction in particular was reported to be in phase with the
Pacific sardine. Which indicates that both fish could be responding to the same
environmental changes. The warm versus cool regime shift in the northeast Pacific
seemed to also overlap with the increase and decrease of the Japanese sardine
(Figure 2.7).
31
N
'tkhaLin
45
Hokai
i
I
40
Oyashio
Current
I.--
I
I
I
I
C: Chosh'
Jf!
T:Tokyo
8aagami
Tsushima curent
i
Rii
350
Ry:I'.yusin
\
St
Se: Seto In1andSca
)Shio
[('S:East China Sea
Current
hi
EC
1300
e
135°
140°
I
Pacific Ocean
145°
150° E
Figure 2.8. Important areas for the Japanese sardine and anchovy. The lower
bold lines represent the Kuroshio Current with its different paths, straight (St)
and meandering (Me). The upper bold line represents the Oyashio Current,
while the bold line at the right bottom corner is the Tsushima Current.
32
East China, Western
Japan, Seto Inland Sea,
Coastal area of central
to southern Japan.
South and east Kyushu,
small coastal areas in
western and southern
Japan.
I
Expansion
I
I
Event
(Warming)
Contraction
Expansion
Event
I
Contraction
(Cooling)
+
L..._._._._
South Sakhalin coast,
Japanese Sea coast of
Korea Peninsula,
Surrounding seas of the
Japanese Island
Ise and Mikawa Bays,
Segami Bay off Bojo
Peninsula
Figure 2.9. The sardine (solid) and anchovy cycle in the Japanese system.
Catches of sardine reflect fluctuations in sea surface temperature. Kawasaki
(1991) found a positive correlation between the sardine catch and air temperature,
and suggested the influence of global long-term climate change.
Long-term
variations in Japanese sardine appear to relate to interdecadal North Pacific oceanic
climate variability (Yasuda et al 1999), In 1988, increased solar input was found to
not only result in higher sea surface temperature but to encourage expansion of the
Japanese sardine population by increasing phytoplankton production (Kawasaki
and Omori 1988). High catches of sardine correspond to periods of warm water
33
temperature in spawning grounds and cool temperatures in feeding grounds
(Tomosada 1988).
Although there is little evidence in the literature on the fluctuation of the
Japanese anchovies, they were reported to expand and contract as the Japanese
sardine do, but with a narrower geographic distribution (Figures 2.9 and 2.11).
When Japanese anchovy are at high abundance they are distributed over a wide
range including the coastal areas of central to southern Japan on all sides including
the Pacific Ocean, Western Japan Sea, Seto Inland Sea, and East China Sea
(Funakoshi 1992). While at low levels, they were observed mainly in Isle and
Mikawa Bay and Sagami and Tokyo Bays off the Bojo Peninsula (Kondo 1980)
(Figure 2.9 and 2.11).
The increase in the Japanese anchovy could be partially a function of the
decrease in the Japanese sardine. In addition, long term cooling of the Tsushima
Warm Current and cooling in the feeding grounds on the Pacific Ocean side
coincided with the decline of the Japanese anchovy (Ogawa and Nakahara 1979).
The Japanese anchovy abundance is much more stable when compared with the
Japanese sardine abundance. Unlike with the California anchovy, there is no
evidence in the literature to support cyclic behavior in the Japanese anchovy
abundance.
34
'-.---J
-i.-
'kh1in
450
1okka1
s(
Sea of Jaan
400
I
I
I
I
I
I
I
I
--
I--.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Honshu
I
350
I
I
I
1
I
I
- -----.1- --------------- 3--.
I
1
I
I
p
F
I
I
I
I
130°
1350
1400
I
1
1
1
I
I
145°
150° B
Figure 2.10. Japanese sardine expansion areas (light) and its spawning areas
(dark) when the population is at high levels. The spawning areas are also the
contraction areas. (after Liuch-Belda et al. 1989).
35
There are no specific refuge areas for the Japanese sardine, however, in
1971, sardines were first recorded to start recovering around the Ku Peninsula
located at 136°E (Kawasaki 1993). Kondo (1980) reported three reasons for the
sardine recovery in 197 1-72. First, the gradual expansion of the spawning area led
to greater egg abundance. Second, there was increased survival at the onset of the
post-larval stage due to adequate abundance of prey in the form of copepod nauplii.
In addition, the increase in the 1 970s coincided with broad-scale warming. During
this period, along with the sardine, many other warm-water species increased
(Shantov and Vasil'kov 1982). In the 1970, the Kuroshio Current meandered
further offshore, which helped boost the food available for the sardine larvae. The
meandering of the Kuroshio does not always favor the sardine larvae. In 1934-45,
as indicated earlier, the meandering in the Kuroshio caused the sardine larvae to be
trapped within an anticyclonic rotating water mass that contained little food for the
larval sardine and caused high mortality of the larval population (Shantov and
Vasil'kov 1982). In 1970, however, the larvae concentration was slightly north of
the meandering Kuroshio and did not suffer such massive mortality. In fact, the
meandering in 1970 acted in favor of the sardine larvae. As a result of the
meandering, large vortices were formed, which facilitated transport of the larvae to
the nursery grounds. Increase in vertical water movement also resulted in increased
food supply for the larvae (Shantov and Vasil'kov 1982)
As the sardine stock expands rapidly, the oceanic waters act as a reserve
spawning grounds (Watanabe et ãl 1997, Kuroda 1991).
Expansion in the
spawning grounds for the Japanese sardine mean an increase in egg abundance and
survival with the Kuroshio frontal waters providing juvenile sardines with
favorable growth conditions (Watanabe and Saito 1998). High survival from eggs
to larvae was observed for the 1977, 1979 and 1980 year classes, which contributed
to the build up of the far eastern portion of the sardine population (Kawasaki 1993).
Like the California sardine, it seems that the Japanese sardine population continues
growing until it cannot expand more, i.e., when the carrying capacity of the
expanded niche favored by the Japanese sardines is filled. The carrying capacity in
36
this case is determined by the distribution of the favorable conditions: distance
from the nursery grounds available for the offspring, food availability, and
predation. Offshore expansion of the spawning grounds caused changes in
environmental conditions where sardine eggs were released that influenced
survivorship during the egg and early larval stages (Watanabe et a! 1995).
Copepod nauplii are abundant in the mixing area between Kuroshio and
coastal water masses as mentioned earlier. In the offshore side and within the
Kuroshio, however, food availability is not good for the sardine larvae and the
predation rate is high. These two conditions limit the expansion of the Japanese
sardine beyond the Kuroshio Current (Nakata et al. 1996). Based on critical food
concentrations, inshore waters are reported to be more favorable for first feeding
larvae than the offshore waters (Watanabe et al. 1998).
The Japanese sardine ecologically adapts when at high population levels by
enlarging its feeding area at the cost of individual growth (Wada and Kashiwi
1991). In this situation, the physical conditions of the fish become worse as a result
of overexpansion and overpopulation and the offspring produced by these become
smaller (Kawasaki and Omori 1995). Fish grow slowly and mature at older ages
(Kawaski 1993). Slower growth of the Japanese sardine leads to substantially
higher mortality rates (Houde 1987). Slow growth makes the larvae spend more
time in any given size class making the larvae more vulnerable to predation
(Shepherd and Cushing 1980). The lipid content in the muscle drops as the fish
population reaches high levels (Kawasaki and Omori 1995), which indicates poor
adult physical condition. Spawning distribution of the Japanese sardine changes in
relation to the stock size but always has been within the Kuroshio Current (Nakai
1962, Kuroda 1989 and 1991), As a result of the expansion of the spawning
grounds to offshore areas, especially the offshore side of the Kuroshio, food
available for first feeding larvae becomes a limiting factor in addition to predation
(Butler and Pickett 1988). When the spawning grounds shift to include the offshore
side of the Kuroshio Current, the sardine stock starts to decrease (Nakata et al.
1996).
37
1
I China
)
IUSSR
/
1
akhaJin
450
400
"H
(.)flSIUI
350
130°
135°
140°
145°
150° B
Figure 2.11. Japanese anchovy expansion areas (light) and its spawning areas
(dark) when the population is at high levels. The spawning areas are also the
contraction areas (after Liuch-Belda et a! 1989).
38
Survival of young sardines in the feeding grounds is important for
recruitment success (Watanabe et al. 1995).
Yearly recruitment is mainly
controlled by mortality from starvation and predation during the egg and larval
periods (Kasai and Suhimoto 1995). The population decline in 1989 was a result of
recruitment failure for four consecutive years. This failure, the causes of which
have not been established, represented cumulative mortality in all the sardine stages
and not only an instantaneous rate of mortality (Wada et al. 1992). This indicates
that there was not a massive die-off of sardines but rather it was a general sustained
decrease in survival.
As with the California sardine and anchovy fisheries, changes in the
Japanese sardine and anchovy fisheries cause development as well as collapse of
coastal fishing villages. For examples in Choshi, a city north of Tokyo Bay on the
Pacific Ocean side (Figure 2.8), sardines are used as food for Hamachi (young
yellowtail), trout and eel culture, as well as feed for livestock. The inhabitants of
Choshi depends on sardines because they are a low-value species that are very
plentiful. When the fish stock gets and remains large for a time, the fishers improve
their fishing technology and fishing success (Kawasaki and Omori 1995).
THE PERUVIAN SARDINE AND ANCHOVY CASE
The Peruvian system is composed of four currents. Northward flowing
currents include the Peru Oceanic Current, also known as the Humboldt Current,
and the 100-200 km wide Peru Coastal Current, nearer the coastline (Figure 2.12).
The southward flowing currents are the Peru Undercurrent and Countercurrent;
both located under the coastal and oceanic currents. The Humboldt Current is slow,
0.04 m/s, and cold with low salinity. The Peruvian coastal upwelling is one of the
most productive upwelling regions in the world with an upwelling index of 1.2
m3.s1 .n11
and wind mixing of 224
m3.s1
(Cury et al. 1998). The upwelling in Peru
persist all the year around (Aiheit and Bernal 1993).
39
One anchovy species exists in the Peruvian system, the Peruvian anchoveta,
Engraulis ringens, and one sardine species, Sardinops sagax, which is similar to
the species found in the California Current system (Ahlstrom 1960). The Peruvian
case is quite different from the previous two cases. The systems in California and
Japan mostly favor the sardine, whereas the Peruvian system favors the anchovy. In
the California and Japanese systems the peak biomass of the anchovy populations
is not nearly as large as the peak biomass of the sardines. In the Peruvian system,
by contrast, the peak biomass of the anchovies is much larger than the peak
biomass of the sardines.
The anchoveta spawns almost year round along the entire Peruvian coast,
with peaks in September and January (Cushing 1981). Due to its short life span,
failure to spawn causes a decline in the population. The fish begin recruiting to the
fishery at 5 months with a large percentage maturing at age one (Glantz and
Thompson 1981).
Anchoveta density and hence availability to the fishery are closely related to
offshore and near shore water conditions with high densities associated with cooler
water temperatures of less than 17°C (Nesterov 1996). On the other hand, strong
year classes of sardine appear during elevated water temperature (Nesterov 1993).
Sardines can be found in waters with temperatures between 17 to 20°C and spawn
in temperatures of 19-21°C.
The Peruvian upwelling system has the highest rates of upwelling in the
world, 1.20 m3.s' .m1 and therefore supports the largest pelagic fish stock (Cury et
al. 1998). Unlike the Japanese and the Californian anchovies, the Peruvian anchovy
population supports very large landings, millions of metric tons during peak
periods. The Peruvian anchoveta catch increased between 1956 to 1972, but then
decreased sharply in 1973. The anchovies started to increase again in 1985, reached
a maximum in 1994, and decreased again to low levels in 1998. The landings of
anchovies peaked again in 1999, with the amount in 1999 being almost six times
the 1988 landings (Figure 2.13).
40
Lf1.1
5°
r
10
\Li
a
150
_L
_L
I
U
L4
1/J
il
I
I
a
a
\
I
I
a
pf.1
a
I
I
I
200
11
I
I
kI
I
a
a
p
U
a
I
a.
a
$
I
I
I
a
Paci4 Ocean
I
I
a
-I
Oceanic
Caastel
current
Current
)
Ja I1
I
I
250
I I I
300
I 1111 I I
I
ii
fl50
000
75°
70° w
Figure 2.12. The Peruvian Current System (after Laws 1997).
14000
Warm Phase
Cool Pha_f
PDO
12000
10000
2
8000
'
57-58
72-73
/
82-83
97-98
/
6000
'
/
/
if
1950
1960
I
/
1980
1970
1990
2000
Year
sardine
Peruvian anchovj
Figure 2.13: The Peruvian sardine and anchovy landings. Data obtained from
FISHSTAT Plus: Universal software for fishery statistical time series. Version
2.3 2000. The dark circles represent the El Nino years as cited in the text.
Relatively cooler temperatures and strong upwelling are important features
of the Peruvian system. Anchoveta prefer cooler temperatures, as do the California
and Japanese anchovies. Therefore the anchovies will dominate the Peruvian
system until an event occurs to change the conditions to favor of the sardines.
During El Nino events the wind is sometimes weakened and the coastal upwelling
is reduced so the upwefled water is low in nutrients and warm (Laws 1997). The
stronger the El Nino event, the weaker the upwelling and the lower the nutrients. El
Nino changes the environmental characteristics, bringing warmer waters and poor
conditions for the food preferred by anchoveta, and reduced area available for
spawning. The effects of El Nino on the anchovies depends on the severity and
timing of the El Nino (Figure 2.13). The length and duration of an El Nino is
irregular and cannot yet predicted. The El Nino in 1957-5 8 affected the anchoveta's
42
distribution but did not affect its abundance. In fact, the anchoveta started to
increase in the mid 1950s. In 1965, El Nino conditions caused a one-year catch
decline but the fishery recovered and produced its maximum catch in 1970. In
1972-73, El Nino conditions lasted long enough to cause the anchoveta population
to decline. Average landings recorded for the two years prior to 1972-73 was 11.28
million metric ton. The average dropped to 3 million metric tons two years after the
72-73 El Nino conditions. Walsh et al. (1980) state that the high larval mortality of
anchoveta caused by the poor food conditions during the El Nino events of 1965,
72 and 76, together with overfishing of the adult population, were the main reasons
for the collapse of the Peruvian anchoveta and the partial replacement of anchoveta
by sardine. During the mid 1970s to 1985, the anchoveta stock was at low levels
while the sardines were increasing. The 1982-83 El Nino hit during a period of low
anehoveta abundance. The anchoveta started to increase in 1986 to mid 1990s
while the sardines started slowly disappearing. Stable water temperatures led the
anchoveta recovery after 1991 (Nesterov 1996). Recent El Nino events were
recorded in 199 1-1995 and 1997-98.
Both El Nino and the related phenomenon called El Nino Southern
Oscillation (ENSO), usually takes place around the month of December, have the
same effects on anchovies. During an El Nino event, water temperature increases in
the upwelling region and the phytoplankton production decreases reducing the
habitat preferred for anchovy spawning. This leads the anchovies to contract to
areas of cooler habitat in the south or perhaps to remain in deeper water to avoid
the upper poor water conditions (Walsh et al 1980) (Figure 2.14 and 2.15a),
However, there is no evidence that the Peruvian anchovy dive to deeper waters.
Therefore the southern Peru area could serve as refuge areas. These areas are less
favorable for anchovies but they are much better than the warm areas to the north.
Concentration of anchovies in these less favorable areas results in poor feeding
opportunities and poor larval survival (Ware and Tsukayama 1981).
Also, during El Nino events the intrusion of warm water expands the
migration route of large tropical predators such as marlin, tuna, and swordfish to
'1-,
't.)
the Peruvian coast, which is thought to result in an increase of the natural mortality
of the anchovies (Smith 1995). The negative effect could also occur with the
sardines but is more pronounced with the anchovies because the environment is
also changing to negatively impact the anchovies but not the sardines.
High fishing pressure during El Nino conditions causes the anchoveta
population to be in even worse shape. Low food concentration during El Nino
events coupled with high fishing mortality of adult spawners results in reduction in
the spawning activity and therefore a decrease in the number of eggs released. This
leads to recruitment failure and the replacement of the anchoveta by the sardines
(Lasker 1975, Walsh et al 1980, Ware and Tsukayamal 981).
Anchoveta rebound once the El Nino event abates and conditions return to
the normal combination of high upwelling and moderate wind speed, resulting in an
optimal environmental window for the anchovies (Cury et al. 1998). High
upwelling intensity is important for sustaining nutrients that maintain food
availability (Wroblewski and Richman 1987, Cushing 1990) and small scale
turbulence increases the encounter rate between food particles and larvae
(Rothschild and Osborn 1988, Mackenzi and Leggett 1991).
At low abundance the Peruvian sardines, on the other hand, spawn off
Chile, but only north of 27°S. When the sardine population expands, they are
recorded along the coast in the Talcahuano region as far south as 40°S (Figure 2.14
and 2.15b) (Lluch-Belda et al. 1989) and to the north near the northern boundary
with Peru (Zuta et al. 1983). The spawning also increased considerably during
periods of high abundance to include almost the entire Peruvian coast. Peruvian
sardines were reported to spawn between September and March (Santander and De
Castillo 1977, Muck et al, 1987).
Along the coasts of
Peru and Chile as far as
40 °S
1
Off Chile only north
27° S and 40° S
Expansion
Contraction
[
Event
(Warming)
A
Expansion
Event
Contraction
(Cooling)
A________________
The entire Peruvian and
the Chilean coast
In south Peru cool
areas
Figure 2.14. The sardine (solid) and anchovy cycle in the Peruvian Current
System.
5*
10*
15*
20G
-.
- -
--
-
Pacific Ocean
:
I
25G
I
I
-I--I--.-'- -----I
I
I
I
I
I
11
I
I
I
I
I
J
I
I
I
I
I
I
I
I
I
I
I
I
I
30*
65*
800
75°
70°
w
Figure 2.15a. Distribution of the Peruvian sardine during low abundance
(after Lluch-Belda et al. 1989).
5°
-
L
L
I
.\
I
Peru
I
I
I
I
i
i
I
H
10°
-L
isa
I
I
20°
Pacifip Ocean
:
:
25°
30°
LIILIIIILiII1I-!r
B5°
ao°
7QG
w
Figure 2.15b. Distribution of the Peruvian sardine during high abundance
(after Liuch-Belda et aL 1989).
47
The Peruvian anchoveta fishery developed very rapidly. Before 1950, the
anchoveta off Peru were harvested very little and mainly for human consumption.
In less than a decade, fish meal produced from anchoveta replaced cotton and sugar
as the main export product in Peru. The anchoveta were so abundant that major
industries developed in harvesting, processing (fishmeal and reduction), and
exportation. The anchoveta became the main source of foreign exchange for Peru.
In 1959, there were 426 anchoveta (wooden) boats; by 1970 the number had
jumped to 1500 (Canon 1973). As the fishery developed, more and more fishers
improved their fishing gear and used advanced fishing techniques. Fishers used
bigger "steel" boats with sonar and powerful engines to locate the anchoveta stocks
and used hydraulic equipment to harvest them. These two techniques reduced the
effort and labor involved with catching the fish. As a result of the increase in catch
efficiency came increases profits.
The fishing sector in Peru became very heavily dependent on anchovies. In
1968, the fishers in Peru landed a total of 10.5 mt of anchovies. Fearing a crash in
the stock, a group of scientists in the Peruvian government issued a warning in
1970 that too many fishers were catching too many anchovies. They estimated that
the sustainable yield should be around 9.5 million tons. Due to political and social
pressure, the Peruvian government allowed the fishers to catch more anchovy and
ignored the advice about the maximum sustainable catch. The government
surpassed the 9.5 mt maximum sustainable catch in 1970 and 1971, landing 12.5 mt
and 10.5 mt respectively. The issuing of catch quotas led the fishers to buy more
advanced fishing technologies and step up the race to catch more anchovies. This
management regime of heavy fishing in conjunction with the strong 1972 El Nino
caused the fishery landings to drop from 10.3 million metric tons in 1971 to 4.4
million metric tons in 1972.
The collapse of the anchoveta had other negative side effects including the
collapse of the guano birds. The droppings of these birds, mainly the guanay and
the piquero, were exported to Europe and North America where they were used as
fertilizer. These guano birds feed on the anchovies; hence, the collapse of the
anchovies meant the collapse of the guano industry as well.
THE SOUTH AFRICAN SARDINE AND ANCHOVY CASE
The South African fisheries for sardines and anchovies occur in the
Benguela Current system.
The Benguela Current is a classic coastal upwelling
system along the southwestern Africa, Namibia and southern Angola. The coastal
upwelling of the Benguela Current system is bounded in the north by the Angola
Current and in the south by the Aguihas Current, both of which are warm systems.
The biologically productive area of the Benguela system in the south (105 000 km2)
is less than the California Current productive area (234 000
2)
This southern
Benguela system, however, has higher primary production rates (735 gC.m2.yeaf')
than California (343 gC.m2.yeaf') (Brwon et al. 1991, Ware 1992). The center of
the coastal upwelling is in Luderitz (27°S), where the upwelling occurs all the year
around. Intensity of upwelling is particularly high between the Orange River and
Luderitz and lessens substantially north of Walvis Bay (Stander 1964). Other
secondary upwelling areas are located at 18°, 20°, and 24° S and at about 31°, 33°,
34° S (BCLME 2001). The continental shelf along the west coast of South Africa is
variable. It is narrow at Luderitz and Cape Columbine and wider at Orange River
and Aguihas Bank in the South (Ware 1992) (Figure 2.16).
Every ten years on average an El Nino-like event hits the Namibian and the
South African coast. This El Nino type event is called the "Benguela Nino"
(Shannon et al 1986). The Benguela Nino condition in the Benguela Current refers
to intrusions of warm surface waters from the north that reduce nutrient
concentrations even at deeper waters (Boyd and Thomas 1984, Shannon et al. 1986,
Le Clus 1990). The Benguela Nino is a result of relaxation of winds off Brazil that
causes warm water in tropical Atlantic to travel eastwards and southwards and
become trapped along the South African coast. Benguela Ninos are not necessarily
in phase with the Pacific El Nino. The Benguela Nino events were recorded in
1934, 1949, 1963, 1984, and 1995 (Gammeisrod et al. 1998).
The major small pelagic fish stocks that the Benguela system support
include the South African sardine, Sardinops ocellatus (described locally as
pilehard), the South African anchovy, Engraulis capensis, Cape horse mackerel,
Trachurus trachurus capensis, Cunene horse mackerel, T tracae, round herring
Etrumeus whiteheadi, and round and flat Sardinella, Sardinella aurita and S.
maderensis respectively. The last two species are entirely caught by the Angola
fishing fleets (I3CLME 2001).
The sardine stock dominated the Benguela Current system from the early
1 950s to mid 1 960s and collapsed after that. This collapse was attributed to heavy
fishing. The sardine purse seine landings increased from 100,000 tons per year in
1950s to reach 1400,000 tons by the early 1960s. The decline was worsened by the
Benguela Nino in 1963. South African sardine remained at lower levels from the
early 1980s to mid 1990s. The sardine biomass in the South African system has
been increasing during the period of 1985-1997, partially due to the management
strategy to rebuild the stock (Cochrane et al. 1998). Anchovies, on the other hand,
started to increase in the early 1960s reaching a high peak period in 1968-1987 and
collapsing after that through the mid 1 990s. The decline in the anchovies in 1989 is
attributed to poor feeding conditions in the spawning grounds in 1988 and due to
poor transport to nursery grounds. The increase and decrease in the South Africa is
more likely a function of optimal environmental conditions that include wind
stress, food availability, and transport mechanism (Hutchings et al. 1998, Cury et
al. 1998), Anchovies have been steadily increasing since 1997 (Figure 2.17).
Variations in such conditions were found to explain much of the variation in the
anchovy recruitment (Shannon 1998).
50
Ango'a
River
/lJ,%
200
Jj \Caç
Cross
Nami
I
\
\
Walvis Bay
25°
:
ange Riwer
30°
CC: CaIte Columbine
Ca: Cape Aguihas
Doiing Bay South Aftica
St. IIelena Bay
CT:CafieTo
CP:CaePoint
I\CT
I
--
350
Agulha
AguthasBank Currenti
I
I
10°
I
15°
20°
25°
E
Figure 2.16. The Benguela upwelling and current system and some localities
mentioned in the text. The shaded areas are the upwelling areas. Redrawn
from Ware 1992 and based on information from BCLME (2001).
51
There are two major spawning areas for sardines and anchovies in South
Africa. In Cape Aguihas area in the south and in between 18° and 25°S (Shelton
and Hutchings 1982; Le Clus 1991). The two species prefer two spawning places
within this range, one is north of 22°S and the other one is around the Walvis Bay
from 22° to 24°S. In the Aguihas Bank area anchovy spawn in cooler inshore
waters in October and February while sardine spawn between August and March,
when upwelling is strong (King 1977a). Sardines are also reported to spawn in
temperatures between 16.5 to 22.8°C (Stander 1963, King 1977b). Sardines spawn
over a wider temperature range (15.2-20.5 °C) than anchovies (17.4-21.1 °C) and
preferentially spawn at cooler water temperatures (15.5 to 17.5 and 18.7 to 20.5°C)
than anchovies (19.5-20.5°C) (Van der Linger et al. 1999). However at 20 m both
of them spawn within salinities of 35.135.6*1 0 and temperatures of 13-19°C (Le
Clus 1991). The larvae and juveniles are transported northwards along the South
African west coast. Larvae then move inshore to the productive west coast
environment in autumn and then begin a southerly return migration to the major
fishing ground around Doring Bay in the south (Hutchings et a! 1998, Hutchings
1992). Adult sardines and anchovies are also encountered in the vicinity of Walvis
Bay inshore north of Cape Cross from autunm to early winter (O'Tool 1977).
In the southern Benguela system during a period of high sardine abundance
the spawning area was reported to include the area west of Columbine and east of
Cape Point (Figure 2.1 8a). When the sardine declined in the mid 1 960s the
spawning areas were restricted further south and included only the west side of the
Agulhas Bank (Schwartzlose et al. 1999, Lluch-Belda et al. 1989). In the north
prior to their collapse, the sardines were available south of Walvis Bay and for
fishers from Luderitz. As the fishery collapsed, the fishers had to go north of
Walvis Bay for sardines, and eventually the processing plant at Luderitz coastal had
to close. When the sardines are at lower levels, spawning is restricted to waters
north of Walvis Bay (King 1 977a, Le Clus and Kruger 1982). When the sardines
52
are at high levels, the catches along the Namibian coast are also high (Crawford et
al. 1988) (Figure 2.18b).
700
600
1965
1984
\''
300
200
Al
100
95
.__) .
f.\.__\
,'L-\/'
I,
is'
"
A
,1
['I
Ci
If
O\
O
\O
O
N
L(
C
C\
C\
C\
c
C
o
C
c
r
C
r-
C
South African sardine
r-
c
Q
oc
Q\
Q
oc
Q\
N
c
o\
O
c
C
a
South African anchovy
Figure 2.17. Southern African sardine and anchovy landings. Data obtained
from FISUSTAT Plus: Universal software for fishery statistical time series.
Version 2.3 2000. The dark circles represent the Benguela Nino years as cited
in the text.
The South African sardine and anchovy have a number of life history
similarities as opposed to the species in the previous three systems, where the
sardines and anchovies tended to be geographically separate due to differing
environmental preferences. In the South African system both species have the same
distribution area and are found together in the same spawning area. They do not
have distinct refuge areas like the California sardines and anchovies. Sardines and
53
anchovies in South Africa might avoid complete overlap by one species preferring
offshore waters for spawning and by having different time of spawning as well.
Sardines are distributed inshore within 50 km from the shore. Anchovies have
similar distribution pattern along the shore but also can be found commonly more
than 100 km offshore in the Aguihas Bank area during the spawning season
(BCLME 2001). Adult anchovies and sardines aged between two and four years
tend to spawn in the warmer waters found east of Cape Point and avoid the cool
upwelled water of the west coast (Crawford et al. 1983). Both species spawn in an
area of minimum offshore wind stress (Parrish et al. 1983, Boyd 1987). Lower
lethal temperature limits for eggs of sardine and anchovy are very close, 13°C and
14°C respectively.
According to Le Clus (1991, 1990) sardine and anchovy in the Benguela
Current system spawn north of 22°S in all environmental regimes, cold, warm, or
1983/84 El Nino type (Benguela Nino) (Figure 2.17). Anchovies and sardine,
however, avoid spawning in the area between 22° and 24°S during cooler regimes
because of strong offshore Ekman transport and the turbulence associated with
upwelling wind stress (Le Clus 1991). During these El Nino type conditions, the
anchovy spawning season is reduced more severely than that of the sardine (Boyd
et al. 1985). The distribution of temperature in the sardine and anchovy spawning
areas ranged from 13 to 21°C. During cooler years surface water temperatures
peaked at 14 to 17°C but at 15 to 18°C during warmer years. Temperatures get
higher during El Nino type event (>21°C). Anchovies prefer and expand during
warmer conditions while the sardines do better during cooler conditions.
Decadal-scale changes in the Benguela current system were detected in an
analysis of long-term sea surface temperatures. The data suggested that cooler
periods existed in the 1920s through l930s, with warmer periods taking over in the
1940s. Warmer periods also recorded in the 1960s through the 1980s (TauntonClark and Shannon (1988)
Density dependent changes have been identified in both the South African
sardine and anchovy. The length at sexual maturity of the sardine, for example,
decreased with the decline in the stock during the 1 960s (Crawford et al. 1980).
The length at sexual maturity of anchovy increased with their biomass in the 1 970s
(Shelton and Armstrong 1983).
Angofa
S
I
r
\\
Namllar
20°
L1____.
I
I
i
.
.-%
--t \
.-4..-' -------
I
I
I
I
4
i
I
I
I
I
I
I
I
25°
30°
IIIIIIIItIITTTi
South Africa
1
I
I
J
I
.\
I
I
I
I
I
I
4
4
I
.-.-e---J
...
350
4.
I
I
100
I
1°
Figure 2.18a. Distribution of the South African sardine and anchovy during a
period of high abundance (after Liuch-Belda et al. 1989).
55
Ango
I
'-L
--.-.4---
200
'
4
i
I
I
I
I
I
I
I
I
..r
Namibia
\\:
":
I
4
I
I
_j
I
I
I
25°
30°
South Africa
4
I
I
1
1
1
1
I
I
I
$
350
4
1
I
4- ---------
I..-..-
I
I
I
I
10°
15°
i
I
i
20°
25U
J
Figure 2.18b. Distribution of the South African sardine and anchovy during a
period of low abundance (after Liuch-Belda et al 1989).
56
Unlike the other the cases where the changes in the environment can favor
the expansion of one species over the other, the South African species attain almost
the same peak levels of biomass. I found few references relating the sardines and
anchovies to the environmental changes. Although it is strongly believed that these
two fish alternate in abundance, like the cases in the Pacific Rim (BCLME 2001),
the wide overlap in the temperatures preferred by the two species suggests that
temperature change alone is not enough to explain the inverse cyclic abundance in
South Africa. Fishing pressure, inter- and intra-specific competition, predation, and
cannibalism may be important factors influencing the inverse cyclic behavior in
South Africa.
SYNTHESIS AND DISCUSSION
The theory outlined at the beginning of this chapter to explain the inverse
cyclic behavior in the abundance of sardines and anchovies varies in its strength
and applicability with each of the case studies. The theory is well supported by the
California sardine and anchovy case. The stages of expansion, contraction and the
cause of fluctuation process are well defined with the California case when
compared with the rest of the cases. The two species seem to have distinct
spawning and refuge areas and reversals in their abundance trends coincide with
long-scale environmental changes The next best-defined case in terms of
expansion, contraction, and causes of fluctuation is the Peruvian case and finally
the Japanese and the South African cases. In California and Peru the contraction
areas between the sardines and the anchovies are very different, while an overlap in
the contraction areas can be found with the sardines and anchovies in the Japanese
and the South African (Appendix A). For the South African system the sardines
and anchovies have similar environmental preferences, while with the Japanese
case, the anchovies hardly respond to environmental changes.
57
Sardines and anchovies are well connected to their environment. Their
fluctuations in abundance in all the cases were a function of environmental change.
The changes in the sardine stocks in California, Japan and Peru are all in phase
(i.e., changes in abundance trends occur at almost the same time) and all these
stocks were reported to prefer warmer waters for spawning than anchovies
(Appendix A). The sardines and anchovies in South Africa are out of phase with
those in the Pacific and are quite different from those in the Pacific. The anchovies
in the South Africa prefer to spawn in warm waters while the ones in the Pacific in
cooler waters. Sardines in contrast behave the opposite. The Pacific sardine cases
are all in phase indicating that they may all respond to the same ocean-scale
environmental changes that contribute to triggering the fluctuation process.
Anchovies stocks around the Pacific, however, are not always in phase. Part of the
reason for this is that not all the anchovies in the Pacific respond to environmental
regime shifts as do the sardines. The Japanese anchovy for example was reported to
fluctuate only three to four times in narrow ranges during the period 1906-1984
(Funakoshi 1992).
That the sardine abundance fluctuations in the Pacific are in phase indicates
that their change may be related to the decadal-scale environmental changes in the
Pacific Ocean basin. That the South African sardines are not in phase with the other
cases indicates that the decadal-scale changes in the Pacific Ocean are not well
coupled with those in the Atlantic Ocean.
Many studies have indicated that regime shifts in the air-sea environment
from cold to warm are correlated with the sardine and anchovy fluctuations
worldwide. Four time-scales of fluctuation were identified by Ware (1995) using a
spectral analysis of 21 climate records: 2-3 years (quasi-biennial oscillation), 5-7
year (ENSO), 20-25 year, and a very low frequency (VLF) oscillation with a 50-75
year period. Environmental changes that occur every 2 to 10 years like ENSO
events and other local temporary environmental changes are not enough to trigger
long-term fluctuations of sardines in the Pacific rim (Rothchild 1995). When
sardines disappear, they do so for longer periods of time than 10 years. Therefore,
58
what triggers these fish to fluctuate must be longer-term environmental events like
the VLF and or the bidecadal variations.
In the Northeast Pacific, the Pacific Decadal Oscillation (PDO) reoccurs
every 20-30 years. The environmental changes associated with the PDO are similar
with those associated with the ENSO. The ENSO events, however, persist for 6 to
19 months and the changes associated with the PDO persist for 20 to 30 years. The
effects of the PDO are mostly observed in the North Pacific and in North America.
Cool regimes of the PDO were identified as occurring in 1890-1924 and again in
1947-1976. Warm PDO regimes occurred in 1925-1946 and from 1977 through the
1990s. The timing of these PDO regimes is similar to the California sardine and
anchovy fluctuations reported in the Baumgartner study (Hare et al. 1999, Francis
and Hare 1994, Francis et al. 1998).
Long term environmental regime shifts affect not only sardines and
anchovies but other fish such as shark, pomfrit, mackerel, albacore and squid
(Brodeur and Ware 1995). In California coastal waters, warming of the surface
waters is correlated with decreased zooplankton abundance as well (Roemmich and
McGowan 1995). The Pacific salmon stocks (sockeye, pink and chum) were found
to be significantly correlated with indices of atmospheric conditions (Noakes et al.
1998). Klyashtorin and Smirnove (1995) used time series data for the annual
surface temperature anomalies of the northern hemisphere and correlated the series
with the California sardine and total salmon abundance. The abundance of both fish
species was correlated with surface temperature. Sardines, however, responded
more quickly to changes in the environment than the salmon. Frances and Hare
(1994) concluded that Pacific salmon stocks were responding to long term changes
in global climate, This change was also found to be in phase with the changes in
sardine abundance in California, Japan, and Peru, as well as with abundance
changes in Alaska Pollock and Chilean mackerel (Klyashtorin and Rukhlov 1998).
Yasuda et al. (1999) show similarities between the abundance of Japanese sardine
and the regime shift indexed by the North Pacific Index (NPI), which represents
trends in primary productivity resulting from changes in upwelling or other
biological and physical processes.
The general conceptual model for the inverse cyclic behavior in the
abundance of sardines and anchovy is given in Figure 2.19. My review of the major
sardine/anchovy systems around the world suggests that the fluctuation process is
due to long-term environmental changes. The sardines/anchovies expand to cover
more area as the environmental conditions allow. Temporary fluctuations might
occur due to short-term environmental perturbations or heavy fishing. These
conditions are temporary and will not cause the sardines or anchovies to retreat to
their refuge or contraction areas. The expansion of sardines or anchovies might be
well limited by density dependent factors. Other variables like distance from
optimal nursery and spawning grounds also limit expansion of the sardines and
anchovies. The sardines/anchovies persist at high levels as long as the surrounding
conditions allow. When the environment "shifts" from cold to warm or vice versa
the sardines and anchovies are also change. Cold regimes result in the anchovy
stocks being dominant in California, Peru, and Japan, and sardine stocks in South
Africa. While warm regimes result in the sardine stocks becoming dominant in
California, Peru, and Japan, and the anchovy stock in South Africa.
When
conditions are unfavorable, the sardines/anchovies contract to their refuge areas
where they can spawn and persist for long periods until the environment shifts
again.
The refuge areas have environmental conditions that permit the
sardines/anchovies to spawn and grow optimally.
The conceptual model of sardine and anchovy expansions and contractions
is consistent with the "Basin model" proposed by MacCall (1990) as a model for
density-dependent changes in a stock's geographic range. MacCall asserted that the
ability of different habitats to support a population can be represented as a basin,
with the population filling this basin as if it were a liquid. The deepest areas in the
basin have the best habitats and are able to support higher population densities.
These centers of the population have the best environmental conditions. The
population density becomes shallower at the edges of the basin. As the population
increases in size, it expands its range into the shallow portions of the basin. In
terms of sardines and anchovies, as these populations increase they spread
geographically into less favorable habitats, but are limited in their spread by the
distance from spawning grounds. These spawning grounds can be thought of as the
deep central areas of the basin with the best habitat. When conditions are poor, the
population contracts to the center of the "Basin" where the conditions are the most
favorable. In this model sardine and anchovy expansions and contractions are
functions of habitat size and suitability.
Other sardine cases around the world also appear to fluctuate and their
fluctuations seem related to environmental changes. In Morocco, for example,
sardines have expanded and contracted regularly during the past five decades,
apparently because of changes in the intensity of Moroccan upwelling, which is a
function of trade wind intensity (Kifani 1998). During periods with reduced trade
wind intensity the upwelling is decreased along the Moroccan coast and
temperatures are elevated. During these periods sardines tend to concentrate in
relatively small areas where they are easily targeted by purse-seine vessels (Kifani
and Gohin 1992). Catches of the Adriatic sardine
Clupea pilchardus
have been
reported as being correlated with the 11 year solar cycle (Regner and Gacic 1974).
Sardines and anchovies in Australia reportedly concentrate in different areas.
Sardines are found almost exclusively in continental shelf waters and sometimes
further off shore. Anchovies, in contrast, are generally restricted to inlets and
estuarine areas and are rarely found in the open oceanic waters (Schwartzlose et al.
1999).
In all the cases reviewed for this study the sardine and anchovy species
occurred in pairs. That is one sardine species and one anchovy species. This low
diversity in the number of sardine and anchovy species in these systems could be
due to relatively uniform environmental conditions and restricted geographic
variability.
61
- Faster growth rates Maturation at younger ages
- Less predation rates Good adult conditions
Healthy offspring
Less removal by fishing
Low Cannibalism - Low inter- and intracompetition
El-Nino
Global air temperature
- Local air temperature
Local sea surface
temperature
- Interdecadal
environmental change
- Regime shifts
+
/
Refuge areas
High levels
Contraction
Event
(Warming)
Expansion
A
Event
(Warming)
Contraction
I
Expansion
_Refuge areas
- - - - -.
High levels
Overexpansion
Growth rate decreases
Interspecific competition increases
- Cannibalism increases
-Poor egg and larval survival
- Predation rate increases
Overcrowding Space becomes limited
Food becomes limited
- Intraspecific competition increases
Poor adult condition
Diseases due to crowding
- Fish at the edges are far from the optimal
environment
-Fishing removes more adult spawners, resulting in reduced spawning activity.
-
Figure 2.19. Conceptual model of sardine (solid) and anchovy (dotted)
cyclic changes in abundance.
62
The literature review for this study was based on papers published in
English. As a consequence, the review may have missed some relevant
information, especially regarding smaller sardine and anchovy populations. In
addition, the conceptual model for the inverse cyclic behavior in sardines and
anchovies is mostly based on speculations about changes in sardine and anchovy
abundance rather than highly accurate measurements or experimentally verifiable
facts. For example, long-term measurements of sardine or anchovy stock sizes were
unavailable. Instead, landings were used as indices of abundance.
If market
demands for sardine and anchovy products are not consistently strong, then
changes in landings will not provide a true representation of changes in the sardine
and anchovy abundance.
CONCLUSION
The conceptual model hypothesized by the author is strongly supported in
some cases and not in others. The California system is the most consistent with the
proposed scenario. The model is weakly support by the system in South Africa
because sardines and anchovies there have very similar temperature preferences
and both expand and contract within the same region. In this case, the biological
interactions between the sardines and anchovies might be very important in
determining which fish population will dominate the system. As long as the
sardines and anchovies have their own refuge areas and their replacement process is
due to environmental change, the inverse cyclic behavior in abundance will
continue.
Sardines and anchovies are intimately coimected to short- and long-term
environmental changes on both local and oceanic scales. Local environmental
changes can cause temporary increases and decreases in one fish stock or the other,
but do not necessarily trigger the replacement process. Instead, broad-scale regime
shifts appear to cause these long-term inverse relationships between sardine and
63
anchovies. Understanding how long term air-sea environmental changes affect the
sardine and anchovy populations and their ecosystems is essential to understanding
the process of sardine and anchovy fluctuations.
64
CHAPTER THREE: QUALITATIVE ANALYSIS OF THE IN VERSECYCLIC BEHAVIOR BETWEEN SARDINE AND ANCHOVY
INTRODUCTION
The previous chapter reviewed the inverse relationship in the abundance of
sardines and anchovies around the world and concluded that environmental change
triggers sardines and anchovies to alternately dominate the system. Although there
are other differences in the environmental characteristics preferred by the two
fishes, it seems that water temperature, which is a function of global weather,
regime shifts, El Nino, and local environmental conditions, is the most critical in
the process of alternating dominance between sardines and anchovies. Water
temperature controls the timing and extent of spawning by sardines and anchovies
and therefore their expansion and contraction. In the California, South African, and
Peruvian systems the alternation between sardines and anchovies is clearly
reflected in fishery landings. In the Japanese case, however, the Japanese sardine
abundance fluctuates more frequently than the Japanese anchovy, which appears to
be more stable.
There are many possible interactions between sardines and anchovies due to
overlap in their biological and life history characteristics. Sardines and anchovies
use the same ecological system and therefore eat almost the same food resources.
Furthermore, beside the possibility of food competition, sardines and anchovies
may interfere directly with each other through predation, particularly when bigger
stages of sardines or anchovies (e.g., adults) feed on smaller stages of the other
species (eggs, larvae, juveniles). Another possible interaction between sardines and
anchovies is mutualism, where both benefit from the presence of each other. The
mechanisms leading to mutualism with sardines and anchovies is somewhat
different from the usual forms of mutualism. Assuming that the predators and
fishing on sardines and anchovies are not specialized, then when the sardines, for
65
example, are in relatively high abundance, fishing activities and predation may
concentrate mostly on the more abundant sardines; thus benefiting the less
abundant anchovies. This could be a negative effect on the sardines the most
abundant fish population. When the anchovies are at higher abundance, they
provide protection for the sardines. Thus, under this form of mutualism high
abundance of one species provides protection for the other less abundant species.
When one species dominates the ecosystem, the other retreats to its refuge area. In
this situation the persistence of anchovies depends on the existence of sardines, and
vice versa. This form of mutualism involves a long-term interaction between the
sardines and anchovies and carmot be expressed directly in terms of a classical
predator-prey relationship.
Commensalism also could occur between sardines and anchovies, where
only one species (either sardine or anchovy) benefits from the present of the other.
Commensalism involves only part of the mutualism interaction between the
sardines and anchovies and is observed over shorter time periods, when compared
with mutualism. In this situation the anchovies, for example, would benefit from
the existence of the sardines when the sardines were in high abundance, but the
sardines would not benefit from the existence of the anchovies when the anchovies
were very abundant. Amensalism, where one species is negatively affected by the
presence of the other species, could also occur between sardines and anchovies.
When the sardines are at higher abundance than the anchovies, the sardines could
consume more food than the anchovies. Therefore the anchovies are negatively
affected by the presence of the sardines. The low abundance of anchovy makes the
fishing activity and predators concentrate on the sardines. So the low abundance of
anchovy has a negative effect on the sardines. Each of the above interactions can
occur at different sardine and anchovy life stages or within the replacement process
itself i.e. when sardines overtake the anchovies or vice versa.
None of the above interactions can be assumed to exclusively represent the
interaction between sardines and anchovies because all of them could occur, but at
different stages in the sardine and anchovy life history.
In this chapter I use qualitative loop analysis techniques (Puccia and Levins
1985) to explore the behavior of different ways of modeling a sardine-anchovy
system. The objectives of this chapter are:
1) To identify feasible sardine-anchovy model configurations, which result in the
inverse relationship between sardine and anchovy, in a simple community
model that considers fishing, sardine and anchovy biomass, and sardineanchovy food resources. In addition, I investigate the stability of these model
configurations and determine the conditions that should be met to stabilize the
unstable model configurations.
2) To determine whether the behavior of the feasible sardine-anchovy model
configurations is altered when fishing is excluded from the model.
3) To identify the model configuration that best represents the sardine-anchovy
system in terms of predicting change in the system variables and thus identify
the crucial interactions between sardines and anchovies that require further
studies.
METHODOLOGY
Loop analysis, also called network analysis, is a method that can be used to
understand the qualitative behavior of dynamic systems. Variables in loop analysis
are represented by circles and the interactions between variables are represented by
lines connecting one variable to another. The connecting lines will have either
arrowheads (+-), dark circles (.-), or nothing at the ends (); each representing
positive, negative, or no effect respectively. The connectors having different types
of ends represent different kinds of interactions: no interaction, amensalism,
commensalism, competition, and predator-prey (Table 3.1). A variable having a
loop to itself with a small dark circle at one end (Table 3.1) indicates selfregulation or a negative self-effect, which means that this variable has resources
available to it that are external to the model system and the variable will not go
extinct. A negative self-effect can be represented mathematically by a logistic
model in which population size tends to stabilize at some finite carrying capacity.
Loop analysis provides a simplified method for evaluating the performance of the
system of differential equations that underlies the ioop diagram. Loop analysis does
not allow one to determine the exact reasons behind the fluctuations typically
observed in sardine-anchovy systems, but it
can help to identify model
configurations where this phenomenon occurs. Loop analysis provides a qualitative
evaluation of a system when the coefficients for the interactions between the
variables are given as 1, -1, or zero according to the interaction type. By replacing
the 1, -1, and zeroes by actual values for the coefficients, loop analysis can also
provide a quantitative evaluation that measures how rapidly the system will return
to equilibrium, and determines the equilibrium values for the variables. With many
variables in a given model, qualitative loop analysis can be very beneficial because
the analysis does not require one to measure the strength of the flows between all
the variables, which in practice is a very difficult task.
Lsr
Table 3.1. Different possible interactions between sardine (S) and anchovy (A)
populations and their ioop analysis representations.
Interaction
Loop Representation
No interaction
There is no interaction between the two species
(S&A)
Amensalism
One species (S) is negatively affected,
but the other is not affected (A)
EE1
Commensalim
One species benefits (S), but the other is
not affected (A).
Competition
Both species (S&A) are negatively affected
Mutualism
Both species (S&A) are positively affected
Predator-Prey
One species is positively affected and the other one
is negatively affected.
Sardine
Negative self-regulation, the species (S) never
goes extinct and always regulate itself
Anchovy
Negative self-regulation, the species (A) never
goes extinct and always regulate itself
.
The table of predictions under qualitative loop analysis is a useful tool for
predicting or forecasting the responses of the different system variables to changes
in the birth or death rate of each of variables (Bodini 1998, Bodini and Giavelli
1989). All the predicted changes for the different variables are put into a matrix
called the table of predictions. The direction of change in the table of predictions is
expressed as a (+) for a variable that responds by increasing to a higher equilibrium
level, a (-) for a variable that decreases to a lower equilibrium level, and a (0) for a
variable that has no change. The direction of change can also be indicated by
arrows increasing
(1')
or decreasing (1).
In loop analysis the interactions between the different variables in the model
system can be represented by coefficients in a square matrix, called the community
matrix, that has one row and colunm for each variable. The coefficients determine
the direction and magnitude of the flows between the variables. In general, change
in the equilibrium value of variable X in any given model, due to change in
coefficient c is given by the following formula (as described by Puccia and Levins
1985),
AX
V I Af,') pk YF(0hh1P))
j\ n-k
frLAc)
Ac
where
P1
is the path to j from 1. A path is the interaction between two variables. If
we need to determine the effect of j on i, then the path is written as P, and reads as
the path from j to i or simply i from j. In the above equation, k is the number of
variables involved in the open path, which is defined as the series of links that
coimect variable i to j without crossing any variable twice, and Fflk0m is the
feedback of the complementary subsystem. The complementary subsystem is
defined as all the n-k variables that are not included in the open path, where n is the
total number of variables in the model. In the term Af1/Ac, f is the function for the
time rate of change of the
ith
variable. Finally, the term F is the overall feedback of
70
the entire system, which cannot be zero (Puccia and Levins 1985). A list of
terminology definitions is given in Appendix B
To illustrate these ideas, consider model Y with three variables, the Sardine
and Anchovy biomass (S and A), numbered as variables 1 and 2, and the Food
Resource (FR), numbered as variable 3 (Table 3.2, Figure 3.1). In this model the
change in the equilibrium abundance of S, due to an increase (+) in its own growth
rate (c) is
+(l)(aAFJ?aF)
zXc
P1 in this case is
==+
which is always defined as 1, k is 1, the path length is
k-l=0, and the complementary feedback
F..k
is 3-12. The complementary
subsystem is (aFR)(-ap) and has a negative sign. The overall feedback
negative. The resulting sign of the formula for S!C is put in cell
the so-called table of predictions. Cell
a11
a11
(Fe)
is
(or ass) of
(or ass) shows the prediction of what will
happen to the sardine biomass if the sardine population growth rate increases.
Similarly the asA cell (or
a12)
shows the prediction of the response by S to an
increase in the anchovy growth rate. Appendix C provides more information about
the model Y example and describes how to obtain the complementary feedback
(FomP) and other calculations.
Loop models can be checked for stability. If a system is perturbed from a
stable equilibrium, it will return to that equilibrium. If the equilibrium is unstable,
the system will not return to the original equilibrium. In many cases there are
conditions that must be met in order to stabilize the system. If such conditions exist
then the system is described as conditionally stable. According to Routh-Hurwitz
(May 1973) there are two conditions that must be met in order for a system to be
stable. First, the feedback, which is the overall sign of the interaction among
variables involved in a path should be negative for all paths at all levels. Second,
71
feedback at lower levels has to be stronger than the feedback at upper levels. Lower
level feedback involves fewer variables in a path than upper level feedback.
For a
system with three and four variables, the second condition mathematically will be
Fl. F2 + F3 > 0, similarly for a five variable system
(F1.F2+F3).F3+(F1.F4+F5).F1 >0 where F is the feedback and the number
associated with it indicates the level of feedback.
In this study loop analysis methods were used to evaluate the different kinds
of interactions that might occur between sardines and anchovies in a system with
four variables: sardine biomass, anchovy biomass, a food resource, and fishing. The
environment carmot be used as a variable in the loop analysis because it is not
affected by the other variables in the system and so will always appear as a constant
in the analysis of the system. The effect of the environment on sardines and
anchovies and its role in the alternating cyclic behavior of these species is
described in a previous chapter.
A study by Baumgartner et al. (1992) on deposits of sardine and anchovy
scales in anaerobic sediments in Santa Barbara, California revealed that the inverse
cyclic behavior in the abundance of the Pacific sardine and the northern anchovy
has existed for a long time, well before the existence of any fishery. Therefore,
models that produce the cyclic behavior, even after removing the fishing variable,
should be more realistic representations of the sardine-anchovy systems. Although
there are no studies showing very long-term and strong cyclical fluctuations in the
worlds other sardine-anchovy systems, I assume the Japanese, South African, and
the Peruvian systems are similar to the California system.
72
FR
Figure 3.1. Example Model Y, where sardines are denoted by S, anchovies by
A, and their shared food resource by FR. The interaction between S and A
with FR is a predator-prey interaction, while there is no interaction between S
and A. All the variables (S, A, and S) are self-regulated.
Table 3.2. Illustrative table of predictions. The column labels indicate the
variable whose birth rate has increased. The cells show the responses of the
variables. For example, cell a23 (or aAFR) shows the effect on A of increasing
the birth rate of FR. See Appendix C for more details.
Variables
Sardine
'1'
Anchovy
4'
Food Resource
1
Sardine
a11 or ass
a12 or asA
a13 or asFR
Anchovy
a21 or aAs
a22 or a
a23 or aR
Food Resource
a31
a32 or aF4
a33 or aFRFR
or aFRs
For the first objective four basic systems were investigated (Figure 3.2). In
Basic Set 1 the sardines and anchovies share the same food resource. In Basic
Set 2 the sardine and anchovy components each have their own food resource (FRS
and FRA) instead of competing for the same food resource. In Basic Set 3, separate
fisheries operate simultaneously but independently on the sardines (FS) and
anchovies (FA), while the sardines and anchovies have a shared food resource (as
in Basic Set 1). Basic Set 4 is like Basic set 3, however, in Basic Set 4 the sardines
73
and anchovies each have their own food resource (as in Basic Set 2). The
interactions between fishing and sardines, and fishing and anchovies are fixed, such
that fishing reduces the fish biomass. Fishing consumes both the sardine and
anchovy stocks, and the sardine and anchovy stocks, in turn, consume the food
resource. Fishing, however, is not reduced by any variable in the system, which
corresponds to the assumption that fishing costs and fish prices, which are likely to
influence the intensity of fishing, are fixed. Natural mortality of the fish stocks and
the food resource, caused by starvation and predation by organisms other than
sardines or anchovies, is embedded in the self-effect loops.
F
S
A
FRA
Set I
Set 2
FS
I
S
FR
Set 3
Set4
Figure 3.2. The Basic Sets of sardine-anchovy systems.
74
Each Basic Set was divided into subsets based on the food resource share.
The subsets then were divided into three groups based on the self-regulation by
sardines and anchovies. Basic Set 1 has only one food resource, which is shared by
both sardines and anchovies; therefore this Basic Set is only divided into groups,
Basic Sets la to ic (Figure 3.3). Basic Set la (Figure 3.3a) assumes that both
sardines and anchovies share the same food resource but they do not rely
exclusively on this food resource. Each species has a ioop to itself to represent the
availability of alternative food resources. In Basic Set lb only the sardines have an
alternative source of food. A parallel configuration (not shown) would have the
self-effect on anchovies but not on sardines. The Basic Set 1 c models assume that
both sardines and anchovies prey exclusively on the same food resource and have
no alternative sources, hence there are no self-effects.
For each subset under all of the Basic Sets I examined model configurations
with the following types of interactions between sardines and anchovies: no
interaction,
amensalism,
commensalism,
competition,
predator-prey,
and
mutualism, illustrated for the Basic Set 1 subsets in Figure 3.4.
Due to symmetry under each Basic Set; there are six different models or
system "configurations" in all the type a and c subsets (Figure 3.3) (no interaction,
amensalism,
cornmensalism,
competition,
predator-prey,
and
mutualism).
However, due to asymmetry there are a total of nine configurations under the type b
subsets because switching species in the amensalism, commensalism, and predatorprey links results in different system behavior.
75
0
FR
a
0
FR
C
Figure 3.3. Different possible groups derived from Basic Set 1.
The configurations in Basic Set 2 were divided into subsets A, B, and C
based on the interaction between sardines and anchovies with their food resources
(Figure 3.5). Subset A assumes that sardines can feed on the anchovy's food
resources. Subset B assumes that sardines share their food resource (FRA) but not
the opposite. Subset C assumes that sardines and anchovies share each others food
resources. Subset C is slightly different from a four variable system with a shared
food resource (Basic Set 1) in that the models in Subset C assume that sardines and
anchovies could be specialized predators on the two different food resources.
76
0
/0
\
c®<
(FR
FRI
No interaction
Amensalism
Commensalism
Competition
0
c®
FR
Mutualism
Figure 3.4. Different sardine and anchovy model configurations considered
under Basic Set 1, Subset a.
Given the Basic Sets 1, 2, 3, and 4 and their possible subsets and groups I
examined a total of 333 possible model configurations. For each one I evaluated the
corresponding qualitative table of predictions to determine whether the model
configurations were consistent with sardine-anchovy alternations. Any model that
77
predicted an inverse relation between sardine and anchovy was selected for further
analysis, while any model that did not predict an inverse relation was considered to
be an implausible model for a real sardine-anchovy system. An inverse relation
between sardines and anchovies was determined by looking for +7- or -1+ in the
tables of prediction.
For the second objective, which is to determine whether the behavior of the
feasible sardine-anchovy model configurations is altered by fishing, the fishing
variable was removed from Basic Sets 1, 2, 3, and 4, resulting in two sets: a 3*3 set
(derived from Basic Sets 1 and 3), and a 4*4 set (derived from Basic Sets 2 and 4).
These new 3*3 and 4*4 sets are called Basic Sets
5
and 6 (Figure 3.6). These
models were analyzed for alternation of the sardines and anchovies by looking at
the tables of predictions and were analyzed for stability as well. Basic Set 5 was
divided into three different groups based on the sardines and anchovy self-effects,
groups a, b, and c, while Basic Set 6 divided into three different subsets based on
the food resource sharing; subset A, B, and C (Figure 3.7). Each subset than was
divided into three groups based on the self-effects, for a total of nine groups.
All the model configurations with the fishing effect removed (Basic Sets
5
and 6) were tested for the inverse relationship between the sardines and anchovies
and compared with the corresponding models that included the fishing effect (Basic
Sets 1-4).
78
FRA
Subset A
FRA
Subset B
Subset C
Figure 3.5. Different Subsets (A, B, and C) from Basic Set 2. Models in Subset
A have completely separate food resources for sardines and anchovies. For the
models in Subset B the sardines can feed from their own as well as the
anchovies' food resource while anchovies cannot feed from the sardines' food
resource. Finally, in Subset C both sardines and anchovies can feed on each
others food resource.
S®
Set 5
Set 6
Figure 3.6. Basic Sets 5 and 6 were derived by removing fishing from Basic
Sets 1 and 2.
Figure 3.7. Different Subsets (A, B, and C) under Basic Set 6. Subset A
assumes that the sardines and anchovies have entirely separate food resources.
Subset B assumes that sardines consume the anchovies' food resource in
addition to their own, but anchovies only have their own food resource.
Subset C assumes that both sardines and anchovies consume each other's food
resources.
79
80
Under objective three, to identify the model configuration that best
represents the sardine-anchovy system, I derived a "pooled" table of predictions
(PTP) for each Basic Set from the model configurations that resulted in an inverse
relationship between sardines and anchovies. This table highlighted the similarities
and differences among those configurations that produced alternations in the
sardines and anchovies.
For each model configuration under a given Basic Set, the tables of
predictions were from those configurations that had inverse relationship between
sardines and anchovies and that were stable or conditionally stable obtained. The
predictions from these individual model configurations were combined into a single
table. For each cell in this combined table I calculated the percentage of cells in the
original models that predicted increase, decrease, or no change. The pooled table of
predictions showed the highest agreement between the different model
configurations regarding the direction of the predicted change (increase, decrease,
or no change).
For example, suppose there are a total of 25 model configurations for Basic
Set I that predict an inverse relationship between sardines and anchovies, and that
for cell a11, there are 15 models that predict an increase, 7 that predict a decrease
and 3 that predict no change. The majority of the models (61%) predict an increase
in cell all and that most frequent prediction is put in the pooled table of predictions.
The cells in the pooled table of predictions show the most likely directions of
change in the corresponding variables for the different models.
Mathematically the pooled prediction for a cell is obtained from the formula
mode (sign)
*
freq [mode (sign)]
n
where n is the total number of stable and conditionally stable model configurations
that predict sardine-anchovy alternations in the given Basic Set, mode(sign) is the
most frequent sign in the different tables of predictions, sign indicates a predicted
81
increase, decrease, or no change, freq[mode(sign)] counts the number of repetitions
of the given sign in the different tables of predictions.
The pooled table of predictions identifies the most likely set of predictions
under each Basic Set of models. The different tables of predictions from the
individual model configurations were compared to the pooled table of predictions
to choose a single model configuration from each Basic Set that best represents the
sardine
anchovy system. Any table that completely overlaps the pooled table of
predictions was considered to reliably represent the sardine and anchovy system.
All the models that completely overlapped the pooled table of predictions were
further checked by removing the fishing effect to confirm that these models
retained the inverse relationship between sardines and anchovies. Models satisfying
all these criteria were considered the best ones to represent a sardine and anchovy
system. Hence, the interactions between the sardines and anchovies in these models
were considered to be important for further studies.
RESULTS
From Basic Sets 1 to 4 a total of 333 model configurations were checked.
Only 84 model configurations from Basic Sets 1 and 2 predicted the inverse
relation between sardines and anchovies (Table 3.3a). From these 84 models, a total
of 13 model configurations resulted in zero overall feedback and 7 had positive
overall feedback. In terms of predictions, when a model has zero overall feedback,
the model variables fail to respond to any changes. Therefore, it is not possible to
predict the direction of change of the variables. Models with positive overall
feedback, on the other hand, suggest that one or more of the system variables
increase exponentially and it is not possible to stabilize the system (Edelstei-Keshet
1988). Hence, only 64 models were further analyzed for the pooled table of
prediction. Because none of the model configurations in Basic Sets 3 and 4 resulted
82
in the inverse relationship between the sardine and anchovy populations, these
model systems were excluded from any further analysis. After removing fishing
from the systems (Basic Sets 5 and 6), only 34 model configurations retained the
inverse relation (Table 3 .3b). Four models resulted in zero overall feedback and six
had positive overall feedback.
Basic Set 1 and Basic set 5
All the models under Basic Set 1, a total of 16, predicted the inverse
relation between sardines and anchovies (Appendix D 3.1). Due to zero overall
feedback in these systems, tables of predictions could not be obtained for models in
Basic Set la where sardines and anchovies compete with each other. Under Basic
Set ib, amensalism also resulted in zero overall feedback, while competition
resulted in positive overall feedback. Under Set 1 c, the models with no-interaction
and the predator-prey interaction resulted in zero overall feedback. In general, it
appears that configurations where both sardines and anchovies have no self-effect
more frequently result in zero or positive feedback.
Model configurations with zero and positive overall feedback were not
included when calculating the pooled table of prediction. The table of predictions
does not exist for models with zero overall feedback. Models with positive overall
feedback are not stable and therefore produce persistent alternations between
sardines and anchovies. The pooled table of prediction included those models that
had an inverse relation between sardines and anchovies when fishing was included
in the model, regardless of whether the inverse relation persisted after fishing was
removed.
Table 3.3a. Summary table of results for all the models analyzed with fishing included.
Basic Set
Subset
Single food
Resource
2
Separate
Food
Resource
Based on Food
Resource (FR)
A None share
FR
B
-
One share
FR
C -
FR
Both share
Groups
Total
Models Models with Models with Number of Number of
Models with inverse Zero overall
Positive
Stable
Conditionally
examined
relation
Feedback overall feedback Models Stable models
a Both with self effect
b one with self effect
c None with self effect
6
5
9
6
7
4
a - Both with self effect
b - one with self effect
c - None with self effect
6
6
9
1
0
5
0
2
1
6
0
2
2
0
2
0
6
0
9
0
0
0
8
1
6
5
1
0
4
2
a - Both with self effect
6
6
0
0
18
8
2
0
0
6
b one with self effect
c None with self effect
8
1
6
5
1
1
3
1
a - Both with self effect
b - one with self effect
6
5
1
0
5
0
9
7
2
1
6
0
c None with self effect
6
4
2
2
1
1
Table 3.3a Continued
Basic Set
3
Single food
resource &
Separate
fishing
Subset
Models Models with Models with Number of Number of
with inverse Zero overall
Positive
Stable
Conditionally
examined
relation
Feedback overall feedback Models Stable models
Based on
Fishing (f)
A Separate
a Both with self effect
F
c None with self' effect
B One share
a Both with self effect
b one with self effect
c None with self effect
9
6
a Both with self effect
b one with self effect
6
9
c - None with self effect
6
0
0
a - Both with self effect
5
0
b one with self effect
8
c - None with self effect
5
0
0
B - One share
a - Both with self effect
6
0
FR
b one with self effect
c None with self effect
9
0
6
0
C - Both share
a - Both with self effect
6
0
FR
b one with self effect
c None with self effect
9
0
6
0
F
C - Both share
F
4
Groups
Total
Models
b - one with self effect
6
9
6
0
6
0
0
0
0
0
0
Based on FR
Separate food A None share
food resource food resource
& Separate
fishing
Table 3.3a continued
Basic Set 4
continued
Subset
Groups
Total
Models Models with Models with Number of Number of
Models with inverse Zero overall
Positive
Stable
Conditionally
examined relation
Feedback overall feedback Models Stable models
Based on
Fishing (F)
A - One share
a - Both with self effect
b - one with self effect
c - None with self effect
6
9
6
0
0
0
- Both with self effect
b - one with self effect
c - None with self effect
6
0
0
0
a - Both with self effect
b - one with self effect
c - None with self effect
6
9
6
0
a - Both with self effect
b - one with self effect
c - None with self effect
6
0
0
0
6
& one share FR
a - Both with self effect
b - one with self effect
c - None with self effect
D - Both share F
& both share FR
a - Both with self effect
b - one with self effect
c - None with self effect
6
9
6
F
B - Both share
F
a
9
6
Based on FR
&F
A - One share
F & one share
FR
B - One share F
& both share FR
C - Both share F
9
6
9
6
0
0
0
0
0
0
0
0
Table 3.3b, Summary table of results for all models analyzed with fishing excluded.
Basic Set
Subset
Groups
2
Based on Food
Resource (FR)
Separate Food A None share
Resource
FR
B
-
One share
FR
C -
(FR
Both share
Models
Models with Models with Number of Number of
with inverse Zero overall
Positive
Stable
Conditionally
relation
Feedback overall feedback Models Stable models
- Both with self effect
b - one with self effect
c - None with self effect
6
2
9
6
2
2
a - Both with self effect
b - one with self effect
5
8
None with self effect
a
Single food
Resource
Total
Models
examined
c
-
0
0
0
2
0
1
1
0
0
2
1
0
1
0
1
0
0
0
5
0
0
1
0
0
0
-
-
1
a - Both with self effect
b - one with self effect
6
3
0
0
3
0
18
7
1
1
6
0
c - None with self effect
6
2
0
0
2
0
a - Both with self effect
6
5
1
0
5
0
b - one with self effect
c - None with self effect
9
6
7
2
1
6
0
5
1
2
3
0
87
All the model configurations under Set 1 a met the Routh-Hurwitz stability
criteria. Under Set lb only the competition model resulted in positive overall
feedback and was thus unstable. Under Set 1 c amensalism and competition also
resulted in positive overall feedback. Commensalism and mutualism under Set ic
were conditionally stable and these models did not meet the second Routh-Hurwitz
criterion, that lower-level feedback must be stronger than upper-level feedback. All
the interactions between sardines and anchovies under Basic Set 1 seemed capable
of producing the inverse relationship between the two fishes.
Pooled Table of Prediction (PTP):
The pooled table of predictions for Basic Set 1 (Table 3.4) revealed that an
increase in the rate of fishing resulted in an increase in fishing (a 100% agreement
between the model configurations), a decrease in the sardine and anchovy
populations (69% and 77% agreement respectively), and finally an increase in the
food resource (100% agreement).
Increasing the sardine population growth rate resulted in an increase in the
fishing variable (62% agreement), an increase in the sardine population, a decrease
in the anchovy population, and no change in the food resource (100% agreement
for each). An increase in the anchovy population growth rate resulted in an increase
in the anchovy fishing (85% agreement), a decrease in the sardine population, an
increase in the anchovy population, and no change in the food resource (100%
agreement for each). An increase in the rate of food resource production resulted in
increases in both the fishing variable and the food resource (100% agreement) and
no change in the sardine and anchovy populations (100% agreement).
A total of five models had 100% agreement with the pooled table of
predictions (Figure 3.8). These models included three different types of interactions
between sardines and anchovies: no-interaction, commensalism, and mutualism.
When both sardines and anchovies are self-regulated, the three interaction types
88
(no-interaction, commensalism, and mutualsim) between the sardines and
anchovies will produce alternations. When neither or only one of the two fishes are
self-regulated, mutualism is the only type of interaction that will reliably produce
the sardine-anchovy alternations. The pooled table of prediction associated with
Basic Set 1 had very high levels of agreement except for the prediction between
sardines and anchovies and their fishing.
Table 3.4: The pooled table predictions for the Basic Set 1 models shows the
highest percentage of agreement among the model configurations that
produced alternations between sardines and anchovies. The arrows indicate
direction of prediction; (1') increase, (a') decrease, (0) no change. The
percentages within brackets indicate the degree of agreement between the
model configurations regarding the direction of the prediction. A total of 13
models in this Subset produced alternations between sardines and anchovies
and were included in this pooled table of predictions.
Variables
Fishing
Sardine Population
Anchovy
Population
Food Resource
F
S
A
FR
(100%)
(62%)
(85%)
(100%)
1'
1'
1'
(69%)
(100%)
(100%)
(100%)
'If
1'
..If
0
(77%)
(100%)
(100%)
(100%)
1'
0
'if
.if
(100%)
(100%)
(100%)
(100%)
1'
0
0
'1'
The fact that the other model configurations and other types of interactions
(amensalism and competition) were not as consistent with the pooled table of
predictions does not mean that these interactions and configurations are not
important. The five model configurations that were chosen, and the sardineanchovy interactions in them, are the most reliable for predicting changes in all the
given variables. To refine the model selections, and to be consistent with results
from the Baumgartner study, the best models for representing real-world sardine
89
and anchovy systems will be ones that retain the inverse relationship between
sardines and anchovies when fishing is removed.
When fishing was removed from Basic Set 1 (resulting in Basic Set 5) only
six of 13 models still kept the inverse relation between sardines and anchovies
(Figure 3.9). These models had only three types of interactions between sardines
and anchovies: no-interaction, competition, and amensalism. From these models,
models d, e, and f resulted in positive overall feedback, leaving only models a, b,
C
as plausible sardine and anchovy models when there is no fishing. However, only
model a, which corresponds to the model where both sardines and anchovies are
self-regulated and yet share the same food resource with no other interaction from
Basic Set 1, had 100% overlap with the pooled table of predictions from Basic Set
1. Thus, this model (Figure 3.10) is the simplest model that most closely conforms
to the alternating behavior observed in real sardine and anchovy systems. This
result indicates that when both sardines and anchovies are self-regulated, the type
of interaction between them is not important.
Figure 3.8. Model configurations from Basic Set 1 that had 100% agreement
with the pooled table of predictions.
G®\/?
A
FR
Figure 3.9. Model configurations from Basic Set 5 that still held the inverse
relationship between sardines and anchovies even after the Fishing effect was
removed.
91
Figure 3.10. The model from Basic Set 1 that most reliably predicts changes in
the different variables. This is the only model that maintains the inverse
relationship between sardines and anchovies when the fishing effect is
removed.
Effect of fishing on the interaction between sardine and anchovy:
None of the models under Basic Set 5 with mutualsim and commensalism
as the interaction between sardines and anchovies resulted in the inverse relation
between sardines and anchovies (Figure 3.9). This result is very logical because if
sardines and anchovies are mutually related then the increase of one fish means the
other increases as well, or vice versa.
Similarly, when the interaction is
commensalism one fish benefits from the increase of the other fish.
In relation to the model configurations with and without fishing, the models
can be divided into two groups based on the cause of the inverse relation between
sardines and anchovies. In one group, with models that have commensalism and
mutualism as the interaction between sardines and anchovies, the inverse relation is
due to fishing. In the other group, with models that have amensalism and
competition, the inverse relation is due to factors other than fishing.
92
Conditionally Stable Models:
Under Basic Set 1, whose models all include fishing, there were two
conditionally stable models, configurations having amensalism and mutualism
between the sardines and anchovies, which are the types of models where fishing
seems to cause the inverse relation. Therefore to stabilize these systems one has to
control the fishing. Furthermore, fishing is the only variable in the system that is
subject to human control. These conditionally stable models appeared in
configurations where neither the sardines or anchovies had any negative self-effect.
Both these models did not satisfy the second Routh-Hurwitz condition, which is
that feedback at lower levels has to be stronger than feedback at higher levels,
equivalentto F1F2+F3>O.
Control of fishing can satisfy the above inequality if one can regulate the
sardine and/or anchovies catch rate coefficients, aFs and aFA respectively, or their
death rates coefficients, asF and/or a. Sardines and anchovies death rate
coefficients can be larger than their catch because the fishers kill more fish than
what they actually retain as catch, the difference being fish that are discarded. To
satisfy the above inequality; I increased/decreased the sardine and anchovy death
and catch rate coefficients one at a time. The coefficient changes that satisfy the
inequality are then identified. I divided the conditions that satisfied the inequality
into two according
to the interaction between sardines
commensalism, and mutualism (Table 3.5).
and anchovies,
93
Table 3.5. Conditions for stabilizing the conditionally stable models from Basic
Set 1. The models are divided into two categories based on the interaction
between sardines and anchovies; commensalism and mutualism.
Basic Set 1 and 2
Commensalism
Satisfy the R.H
Condition
No
Yes
Decrease sardine
death rate (aSF)
Decrease anchovy
death rate (aAF)
Decrease sardien
catch (aFS)
Decrease anchovy
catch (aFA)
Mutualism
Satisfy the R.H
No
No effect
No effect Yes
X
X
X
X
X
X
X
X
The commensalism model could be stabilized by decreasing either the
anchovy death rate or the sardine catch coefficients. Which one of these two
condition to follow depends on the table of prediction. If the table predicted that
sardines are the one that increases, then decreasing the anchovy death rate will
stabilize the system. With the mutualism model, there are more options than with
the commensalism model. This is because the commensalism interaction makes the
model asymmetric, unlike the mutualsim. The condition to be adopted again will
depend on the direction of the fish population, whether it is increasing or
decreasing.
Basic Set 2
In Basic Set I the sardines and anchovies share a food resource, but in Basic
Set 2 each fish species has its own food resource. Basic Set 2 consists therefore of
five variables rather than four as in Basic Set I, removing fishing will leave four
variables in the model configurations, Basic Set 6. Having food resource for each
fish stock simply allows me to assume that each fish stock could have its own food
resource. The Basic Set 2 models were divided into subsets A, B, and C based on
the amount of separation of the food resources: entirely separate, partially separate,
or shared. A total of 72 models were examined. Nine had in zero overall feedback
and four had a positive overall feedback (Table 3.3a), leaving model configurations
for constructing the pooled table of predictions.
The model configurations under Basic Set 2 Subset A a (Appendix D.3.2),
where sardines and anchovies had separate food resources and both were selfregulated all satisfied the inverse relationship between sardines and anchovies and
satisfied the Routh-Hurwitz stability criteria. Under Subset A b mutualism model
did not meet the second Routh-Hurwitz stability criterion. Under Subset A c the
models with commensalism and mutualism did not meet the second Routh-Hurwitz
criteria, and the competition model had zero overall feedback. More conditionally
stable models appear as both fish become more dependent on the same food
resource. Under Basic Set 2 Subset A a total of 20 model configurations went into
constructing the pooled table of predictions.
Under Basic Set 2 Subset B, where one fish shared its food resource, a total
of 30 model configurations were examined. Three models had zero overall
feedback and one had positive overall feedback. Only two model configurations
were conditionally stable: mutualsim under Basic Set 2 Subset B b, and subset B c
(Appendix D.3.3). These models did not meet the second Routh-Hurwitz criteria.
Under Basic Set 2 Subset C, where both fish share each other's food
resource, a total of 21 models were examined. Five models had zero overall
feedback and 3 had positive overall feedback, leaving thirteen models for
95
constructing the pooled table of predictions. Only one model was conditionally
stable, mutualism under Subset C c (Appendix D.3.4).
Pooled Tables of Predictions:
All the models in the pooled tables of predictions for Basic Set 2 Subsets A
and B (Tables 3.6a and 3.6b) predicted that an increased sardine growth rate would
lead to an increase in the sardine population and therefore a decrease in the sardine
food resource. Under Subset C, however, all the models agreed that the sardine
food resource would remain unchanged when the sardine population increased
(Table 3.6c). The same scenario is true for the anchovies and their food resource.
For Subsets A and B the pooled tables of predictions predicted that an increase in
the sardine food resource would cause the sardine population to increase (100%
agreement) but under Subset C the sardine population remained unchanged (100%
agreement). Under Subset A an increase in the anchovy's food resource caused the
anchovy population to increase (100% agreement) but under Subsets B and C the
anchovy population remained unchanged (100% agreement in both tables).
Table 3.6. Tables 3.6a, 3.6b, and 3.6c. The pooled table of predictions for the
models in Basic Set 2, Subset A, B, and C. The arrows indicate the direction of
change; (1') increase, (.L) decrease, (0) no change. The percentages within
brackets indicate the degree of agreement between the model configurations
regarding the direction of change.
Table 3.6a. A total of 20 models in Subset A produced alternations between
sardines and anchovies and were included in this pooled table of predictions.
Variables
Fishing
Sardine Population
________________
F
S
A
FRS
FRA
(95%)
(80%)
(95%)
(80%)
(95%)
1
1'
1'
1
1'
(80%)
(100%)
(100%)
(100%)
(100%)
1'
1'
Anchovy Population
(95%)
(100%)
(100%)
(100%)
(100%)
Sardine Food
Resource
Anchovy Food
Resource
(80%)
(100%)
(90%)
(90%)
(100%)
1
1'
1'
(100%)
(100%)
(90%)
1
1'
1'
(95%)
(100%)
1
1'
I
l-__
I
Table 3.6b. A total of 26 models in Subset B produced alternations between
sardines and anchovies and were included in this pooled table of predictions.
Variables
Fishing
F
5
A
FRS
FRA
(100%)
(54%)
(92%)
(62%)
(100%)
(70%)
(70%)
0
(70%)
0
(96%)
0
(92%)
1
-r
Sardine Population
(62%)
1
1
1
1'
Anchovy Population
(88%)
(100%)
(100%)
(70%)
'1'
1
Sardine Food
Resource
Anchovy Food
Resource
(70%)
(70%)
(70%)
(81%)
1'
1'
(70%)
0
(100%)
0
I
(92%)
1
(100%)
I
I
(70%)
0
(100%)
1
(Continued)
97
Table 3.6c. A total of 13 models in Subset C produced alternations between
sardines and anchovies and were included in this pooled table of predictions.
Variables
Fishing
F
S
A
FRS
FRA
(100%)
(62%)
(85%)
(100%)
(100%)
Sardine Population
(69%)
Anchovy Population
(77%)
1'
1'
(100%)
(100%)
0
0
(100%)
(100%)
(100%)
1'
0
0
(100%)
(100%)
1'
Sardine Food
Resource
Anchovy Food
Resource
(100%)
(100%)
(100%)
(100%)
(100%)
(100%)
1'
0
0
1'
0
(100%)
(100%)
(100%)
(100%)
(100%)
0
0
0
j
Under Basic Set 2 Subset A, a total of eleven models (Figure 3.11)
overlapped 100% with the pooled table of prediction (Table 3.6a). All of the types
of interactions between sardines and anchovies listed in Table 3.1 were represented
in these different models. The no-interaction, amensalism, commensalism,
mutualism, competition, and predator-prey interactions appear in the model
configurations when both sardines and anchovies are self-regulated. The nointeraction, commensalism, and mutualism appear in the models when only one
fish species is self-regulated or when neither is self-regulated.
98
F
F
F
F
C)
O
c4
t
t
t
/0
/O\
C/34
t
t
®
I
Figure 3.11. The model configurations from Basic Set 2, Subset A; that had
100% agreement with the pooled table of predictions (Table 3.6a).
The models in Basic Set 6 were derived from Basic Set 2 models by
removing fishing. When fishing was removed from the Subset A models, the nointeraction models had no link between the sardine and anchovy systems. Only two
models in Basic Set 6 Subset A maintained the inverse relationship between the
sardines and anchovies (Figure 3.12). The model configuration with self-effects on
both the sardines and anchovies (Figure 3.12, panel a) had 100% overlap with the
corresponding pooled table of predictions from Basic Set 2 subset A, in which
fishing was included in the model. The model with a self-effect only on the
sardines (Figure 3.11, panel b), however, does not.
(a)
(b)
Figure 3.12. Model configurations from Basic Set 6, Subset A that maintained
the inverse relationship between sardines and anchovies when the Fishing
effect was removed.
Hence the only model from Basic Set 2, Subset A that seems a good candidate
model for a sardine and anchovy system is the one that has competition between
the sardines and anchovies where both fish populations are self-regulated (Figure
3.13). So, sardines and anchovies do not share food resources, only seems to be a
plausible interaction between sardines and anchovies. When fishing is included in
the models, commensalism and mutualsim also produce the inverse relation.
100
/0
Cl®.
t
Figure 3.13. The model from Basic Set 1, Subset A that most plausibly predicts
changes in the different variables. It is the only model in this subset that
maintains the inverse relation between sardines and anchovies when the
fishing effect is removed.
Under Basic Set 2 Subset B a total of five models (Figure 3.14) overlapped
100% with the pooled table of prediction (Table 3.6b). The interactions between
sardines and anchovies appearing in these models were mutualsim, conmiensalism,
and no-interaction.
When fishing was removed from the Basic Set 2 Subset B, twelve of them
continued to satisfy the criterion that sardine and anchovy abundance be inversely
related (Figure 3.15). In terms of the interaction between sardines and anchovies,
the results here for Basic Set 6 Subset B were almost the same as the results for
Basic Set 1 with fishing excluded. The interactions between sardines and anchovies
that maintained the sardine-anchovy alternations were competition and amensalism,
which support the idea that mutualsim and commensalism are not plausible
interactions between sardines and anchovies. Model (h), in which only the sardines
have a self-effect and the interaction between sardines and anchovies
is
competition, was the only model with positive overall feedback. The rest of the
models were stable.
101
0
0
0
/0
tt
Figure 3.14. The models from Basic Set 2, Subset B that had 100% agreement
with the pooled table of predictions (Table 3.6b).
Only the no-interaction model configuration where both fishes were self-
regulated continued to have sardine-anchovy alternations when fishing was
removed from the model. Therefore this model is considered to be the most
plausible of the Subset B model configurations (Figure 3.16).
Under Basic Set 2 Subset C, five model configurations (Figure 3.17)
resulted in 100% agreement among the model predictions combined in the pooled
table of prediction (Table 3 .6c). Three types of interaction appear in these models.
No-interaction, amensalism, and commensalism appear in models where both
sardines and anchovies are self-regulated. The mutualism interaction also appears
in model configurations where only one or neither of the fish species is selfregulated.
102
ç®®
dd
Qec
gd
chc_ci
cç®
9kccIc
Figure 3.15: Model configurations from Basic Set 6, Subset B that maintained
the inverse relationship between sardines and anchovies when the fishing
effect was removed.
When fishing was removed from the Subset C model configurations that
were in the pooled table of predictions, all the model configurations maintained the
inverse relationship between sardines and anchovies, which implies that there is no
single model that best represents the Subset C model configurations This also
indicates that mutualsim and commensalism under Basic Set 2 Subset C are real.
This means that fishing has a very insignificant role in causing the inverse relation.
However, the pooled table of predictions for the Basic Set 6, Subset C models
103
(Table 3.7) revealed that the configurations overlapping 100% with the pooled
table of predictions were the same ones that overlapped with the pooled table of
predictions form Basic Set 2, Subset C, which included fishing. However, the
simplest of the candidate models in Subset C is the one with no interaction between
sardines and anchovies (Figure 3.18).
F
TT
S\
(A
Figure 3.16. The most reliable model from Basic Set 2, Subset B. It is the only
model that maintains the inverse relationship between sardines and anchovies
when the Fishing effect is removed.
0
0
0
0
0
Figure 3.17. The models from Basic Set 2, Subset C that had 100% agreement
with the pooled table of predictions (Table 3.6c).
Table 3.7: The pooled table of predictions for the Basic Set 6 Subset C models
shows the percentage of agreement among the model configurations that
produced alternations between sardines and anchovies. The arrows indicate
direction of prediction; (1') increase,
decrease, (0) no change. The
percentages within brackets indicate the degree of agreement between the
model configurations regarding the direction of the prediction. A total of 13
models in this Subset produced alternations between sardines and anchovies
(.1-)
and were included in this pooled table of predictions.
Variables
Fishing
F
S
A
FR
(100%)
(100%)
(71%)
(71%)
1'
1'
(100%)
(71%)
(71%)
(71%)
(93%)
(100%)
(93%)
1'
1'
(64%)
(86%)
(93%)
(100%)
1'
Sardine Population
(100%)
1'
Anchovy
Population
Food Resource
.1
'1'
1'
104
105
F
S\
(A
Figure 3.18. The simplest model from Basic Set 2, Subset C.
Table 3.8. The pooled table of predictions for the models in Basic Set 2, all
Subsets combined. Increase, decrease, and no change in a variable are
represented by 1', , and by 0, respectively. A total of 63 models in this Set
produced alternations between sardines and anchovies and were included in
this pooled table of predictions. The numbers in parentheses indicated the
percentage agreement among these models.
Variables
Fishing
Sardine
Anchovy
F
A
(97%)
(98%)
5
(62%)
1'
1'
(71%)
(93%)
(93%)
1'
'1'
(93%)
(93%)
(84%)
FRS
FRA
(87%)
(98%)
1'
1'
(98%)
(64%)
0
(98%)
Sardine Food
Resource
Anchovy Food
Resource
(67%)
(64%)
0
1'
(96%)
(64%)
(98%)
(87%)
1'
1'
0
(98%)
(98%)
(98%)
(65%)
(98%)
1'
0
0
0
1'
1'
The pooled table of predictions for the Basic Set 2, Subsets A, B, and C were
combined into a single pooled table of predictions. This overall pooled table of
predictions was the same as the pooled table of predictions for Subset B with
slightly different (Table 3.8) agreement percentages. This means that the best
106
model to represent the configurations of Basic Set 2 is the model that best
represented the Subset B models, shown in Figure 3.16.
Conditionally stable models:
There were a total of six conditionally stable model configurations among
the 72 models in Basic Set 2 that were included in the tables of predictions
(Figure
3.19).
These models did not meet the second Routh-Hurwitz condition and
had commensalism and mutualism as the interaction between the sardines and
anchovies. In addition, in three models neither the sardines nor the anchovies had a
negative self-effect.. The second Routh-Hurwitz condition requires that feedback at
lower feedback levels should be stronger than feedback at the upper feedback
levels. In terms of the Basic Set 2 models this condition can be written as
(F1F2 + F3) x F3 + (F1F4 + F5)
x Fl > 0 where Fn is the feedback at level n.
where Fn is the feedback at level n.
The inequality can be written in various ways that satisfy the condition and
stabilize the models. However, the only feasible actions that could be taken to
satisfy the inequalities are controls over fishing as indicated previously under
conditionally stable models section for Basic Set 1 models. The conditionally stable
models were divided into two according to the results based on the interaction
between sardines and anchovy. The conditions to satisfy commensalism and
mutualsim models were the same ones as for Basic Set I given in Table
3.5.
107
/0
t
0
0
c.® '0
0 '0
c{
t
0
'®
Figure 3.19. Conditionally stable models from Basic Set 2.
DISCUSSION
Only the Basic Sets 1 and 2 model configurations resulted in the inverse
relationship between sardines and anchovies. All the model configurations in these
sets agreed that amensalism and competition are plausible interactions between
sardines and anchovies in the sense that these interactions maintain the inverse
relation between sardines and anchovies regardless of whether fishing is included
or not in a given model configuration. The analyses of the configurations under
Basic Set 1 and 2 agreed also that commensalism and mutualism are implausible
interactions between sardines and anchovies because the inverse relation between
sardines and anchovies is not maintained when there is no fishing. When sardines
108
and anchovies share independent food resources with each other (Basic Set 2
Subset C) then mutualism and commensalism can be considered plausible
interactions. This means that when sardines and anchovies are very similar in their
resource use yet they alternate in abundance, the environment is the likely cause for
the replacement of one species by the other. As sardines and anchovies are known
to prefer opposite environmental conditions, warm versus cold, fishing under this
particular subset has a very insignificant role, if any, in triggering the replacement
process.
All of the conditionally stable model configurations under Basic Sets 1 and
2 involved implausible interactions (amensalism and mutualsim) and fishing was
needed to induce the fluctuations between sardines and anchovies. To stabilize
these systems, fishing has to be monitored and controlled. Fishing is controlled
either by decreasing the catch rate or death rate coefficients of the less abundant
fish. One could instead increase the catch of the more abundant fish, but such a
policy could lead to overcapitalization and market problems.
Although the environment seems to be the main cause of the fluctuations
between sardines and anchovies, fisheries management can sometimes help to
rebuild a depressed stock. The moratorium on fishing for California sardines during
1974-85 successfully helped in rebuilding the sardine stocks (Wolf 1992).
Sometimes the environmental change is so significant that management measures
are not adequate. A strong El Nino event in 1972 caused the fishery to almost
completely collapse (Walsh et al 1980)
In the plausible models for representing the sardines and anchovies, at least
one fish population was self regulated and the interaction between sardines and
anchovies was either no interaction, amensalism, or competition. This result
suggests that as long as one fish species is self-regulated then this population
cannot be driven to extinction and will always recover whenever the conditions
allow. The negative self-effect could be caused by density dependent growth rates
or by having external food resources for the population (Puccia and Levins 1991).
Although there are many forms of population regulation (Krebs 2001), density
109
dependent control of sardines and anchovies probably involves habitat selection.
The greater the geographic expanse of suitable habitat, the larger the population
becomes. For the sardines and anchovies to persist at low levels of abundance,
they need a suitable refuge area. For example, sardines are believed to concentrate
in the Punta Eugenia area when they are at low levels or when the environment
elsewhere is not good for reproduction (Liuch-Belda et al 1992, Ahlstrom 1960,
Hernandez-Vazquez 1994, Schwartzlose et a! 1999). The anchovies, in contrast, are
believed to concentrate in the Southern California Bight, where they can be found
all year round (Parrish et al. 1986, Moser et al. 1993).
In all the non-stable models, neither fish population had a negative selfeffect (refuge). Hence an increase in one fish could cause the other to disappear
completely. These models remained unstable even when there was fishing, which
suggests that having sardine and anchovy fish stocks without some form of self-
regulation is inconsistent with population alternations. Matsuda et al.
1991
indicated that two fish species can coexist if each has its own "habitat" refuge with
individuals dispersing between the refuges and the common habitat. For the
sardines and anchovies to exhibit persistent fluctuations, they must have such
refuge areas. Without the refuge areas, the sardines and anchovies are complete
competitors and they could not coexist (Gotelli 1998).
There were some conditionally stable model configurations where the fish
were not self-regulated and that had the inverse relation between sardines and
anchovies only when fishing was included. To stabilize these models so that neither
the sardines nor anchovies go extinct fishing has to be controlled.
In any system where there is fishing, there is a danger that uncontrolled
fishing could cause serious depletion of weak stocks. This is especially a problem
in a system where the fish populations rely on refuge areas. If the fishers can locate
the refuge areas, they may irreversibly deplete the stock. Regardless of whether the
fish populations are self-regulated or not, it is important to develop fisheries
management processes to prevent the build up of excess fishing capacity.
110
The pooled table of predictions used in the current project was developed as
a method for finding a single model configuration whose predicted qualitative
behavior is most consistent with the average predictions of a set of models. The
method needs to be evaluated quantitatively as a verification of its validity. In a
future study the best qualitative model, as identified by the pooled table of
predictions, should be compared with the best quantitative model, as determined
from a quantitative pooled table of predictions. Do the quantitative versus
qualitative methods result in selection of the same model?
The analyses of the Basic Sets all indicated that the type of interaction
between sardines and anchovies was not the main factor responsible for sardine-
anchovy alternations. Instead, the results suggest that it was very important that
both populations be self-regulated.
CONCLUSION
The plausible models for representing sardine and anchovy systems are
ones that involve amensalism and competition between sardines and anchovies. In
these models, the inverse relation between sardines and anchovies occurs regardless
of whether there is fishing. Therefore environmental factors are the main cause of
the fluctuations between sardines and anchovies. In contrast, commensalism and
mutualism interactions between sardines and anchovies cannot result in the inverse
relation between sardines and anchovies.
Negative self-effects are important features for stability of the sardineanchovy systems and for the persistence of the sardine-anchovy fluctuations. The
previous chapter concluded that these negative self-effects in sardines and
anchovies are a function of refuge areas.
Without the negative self-effects,
alternation between the sardine and anchovy populations is conditionally achieved.
Another form of density dependence is required to protect the less abundant fish
111
and ensure its recovery. This can be achieved by controlling fishing activity, by
lowering the fishing on the less abundant fish, hence giving it a chance to recover.
In the simplest model for representing the alternations in sardine and
anchovy populations there are no direct interactions between sardines and
anchovies and both populations are self-regulated. This applies regardless of
whether sardines and anchovies have shared food resources. In models where
interaction between sardines and anchovy is applicable, competition become the
simplest model to consider.
112
CHAPTER FOUR: SARDINE AND ANCHOVY FISHERIES IN OMAN
INTRODUCTION
The Sultanate of Oman, a developing country in the Middle East, has a
considerably richer and longer coastline in comparison to its neighbors.
The
Oman's approximately 1700 kilometers of coastline is divided into different
regions, from North to South, Musandam, Batina, Muscat, Sharqiya, Wusta, and
Dhofar (Figure 4.1). Valuable shellfish and fish resources such as abalone, lobsters,
tunas, kingfish small pelagic fishes and many other stocks substantially contribute
to the total income of the country. Oman also has a large number of fish species,
known to be valuable overseas, that are not exploited, such as sea urchins, sea
cucumber, squids, and crabs. Most, if not all, Omani fishers use artisanal fishing
methods, operate only near shore and hardly exploit the offshore stocks.
A few of the Omani fisheries (e.g. the lobster and the abalone fisheries) are
managed with closed seasons. Most, including sardines and anchovies, are among
the Omani fisheries that are not yet managed. Currently there is a growing interest
in the Omani government to study the large pelagic fishes including tunas and
kingfishes as these are considered to be very valuable to the fishers' income and the
country's economy. The Omani government will soon need to understand the food
resources of such pelagic fish to assure their proper management and sustainability.
Along with the increasing interest in tunas and kingfishes is a perception that
sardines and anchovies are the main diet for these fishes even though there are few
studies on Omani sardines and anchovies (Siddeek et al 1994, Dorr Ill 1991, AlBarwani et al 1987).
A better understanding of these f isheries and government
management of them will be needed to ensure that they exist in enough abundance
to attract the tunas and kingfish and keep these large pelagic fish in Omani waters
for extended periods of time.
113
Previous chapters showed that sardine and anchovy stocks from California,
Japan, Peru, and from South Africa fluctuate in an inverse manner. The available
evidence suggests that long-term environmental changes trigger the sardines and
anchovies stocks to alternate in abundance. As the sardines or anchovies expand
they cover a much wider geographic range along the coastal region. This expansion
is a function of having optimal environmental conditions along the coast. Sardines
(or anchovies) persist in high abundance for as long as the optimal conditions
exist(typically for a period of 30 to 50 years,) before abundance collapses and the
range shrinks. When the sardine (or anchovy) population shrinks, the fish retreat to
"refuge areas" that allow the population to exist and reproduce even though the
other fish population has become dominant. When the environment changes to
favor the sardines, the anchovies retreat to their own refuge areas while the sardines
expand.
Given the dramatic changes that occur in the major sardine-anchovy
systems, an essential step to better understand the Omani sardines and anchovies is
to define whether they behave in a similar inverse cyclical manner or not.
Oceanographic Conditions in the Waters of Oman
The Omani coast is open to the Arabian Sea and therefore to the Indian
Ocean, and is affected by the Southwest (SW) and Northeast (NE) monsoon
regimes. The April SW monsoon alternates with the NE monsoon winds, which
blow from November to March. The winds strengthen during June through August,
weakening in September and then reversing at the end of October (Elliott and
Savidge 1990, Fieux and Stommel 1977). The SW and NE monsoons are functions
of environmental changes in the Indian Ocean. The Arabian Sea is the only basin
that fully reverses its circulation pattern on a semi annual basis (Burkill 1999).
During summer, changes in pressure over the north and south Indian Ocean result
in a strong south to north airflow. This airflow blows as steady southwest surface
114
winds over the Arabian Sea, with their onset occurring in April or May and ending
in September (SW monsoon). This surface wind causes a strong, low level stream
known as the Findlater Jet (Findlater 1969). It is located 300-400 km offshore and
blows parallel with the coast of southern Arabia.
Coastal upwelling is driven by Ekman transport and affects hydrographic
conditions to a depth of about 400 m (Brock and Macklain 1992). This upwelling
extends along the shore from Ra's Fartak in Yemen to Ra's Al-Hadd in Oman. The
nutrient-rich upwelled water supports massive blooms of phytoplankton (Owens et
al 1993). Upwelling strength along the coast of the Arabian Sea is a function of
many factors including sea surface temperatures in the Indian Ocean and even the
El-Nino phenomena in the tropical Pacific Ocean (Brock and Macklain 1992).
When the northern Indian landmass becomes cooler than the surrounding
ocean, the pressure gradient reverses and a cool, dry,
weak northeast wind
develops over the Arabian Sea from about November to April (the NE monsoon,
described by Burkill (1999)). It can extend from December to February.
The mean monthly air temperature from 1977 to 2000 collected by Al-Seeb
International Airport in different Omani regions shows that the temperature in
Oman rises from May to September and lowers from October through February.
This pattern can be also seen in the different regions of Oman (Figure 4.2). The
temperatures in Dhofar and Batina are generally cooler than in Musandam and
Sharqiya.
115
25°
200
(
150
t
I
t
I Arabian Sea
- - -- - + - * .-- - * - * - -.*_ -. - .4. - - -
550
60°
E
Figure 4.1. Map of the different regions in Oman and other localities
mentioned in the text.
116
Omani Sardines and Anchovies
In all the cases reviewed in this project, sardines and anchovies existed in
pairs with only one species of each involved. The situation is different in Oman
where there are at least five species of sardines and five species of anchovies.
These species are recorded from the Arabian Sea, the Arabian Gulf, and the Gulf of
Oman (Greg Hermosa, personal contact). The species are:
Sardine species
Anchovies species
Sardinella longiceps
Encrasicholina devisi
S. gibbosa
E. heteroloba
S. aibelia
E. punctfer
S. melanura
Stolephprus commersonii
S. sindensis
S. indicus
Sardinelia ion giceps is the most abundant of the sardine species in Omani waters
according to Siddeek et a! (1994, Don III (1991), and A1-Barwani et a! (1987).
These studies do not indicate, however, the contribution of the rest of the species to
the total catches or biomass of sardines. S. ion giceps, also known as the Indian Oil
sardine, is well known in the northern Indian Ocean and the Gulf of Aden, but not
in the Red Sea (FishBase 1997, Rosa and Laevastu 1960).
The above species are also reported to exist in many other places. S. albeiia
(white sardinella) and S. gibbosa are distributed broadly in Indo-West Pacific along
the East African coast, and from Madagascar eastward to Indonesia. Sardinella
albella is also found in the Red Sea while S. gibbosa is found in the Persian Gulf
but not in the Red Sea. (Munroe and Nizinski 1997). Sardinella n'zeianura is
distributed from the Gulf of Aden south to Madagascar (Suvatti 1981).
Encrasicholina devisi is also distributed throughout the Indo-Pacific and is
widespread in the northern part of the Indian Ocean and the western central Pacific.
It does not thrive in the Red Sea, or the coastal waters of Kenya. Encrasicholina
117
hereroloba and E. punctfer are found in the Red Sea and along the east African
coast to at least northern Madagascar.
Stolephprus commersonii is widespread along the eastern coast of Africa, from the
Gulf of Aden to Zanzibar, northern Madagascar, Mauritius, and eastward to Hong
Kong and Papua New Guinea. It is not found in the Red Sea or the Persian Gulf.
Stolephprus indicus is found in the Red Sea (Russell and Houston 1989).
I
I
I
I
-
-
-
I
O
.<
rn
0
Z
Time
Figure 4.2. Mean monthly air temperature in different regions in Oman
calculated from 1977-2000.
As indicated earlier, few have studies the Omani sardines and anchovies
and there are limited historical data to draw inferences regarding whether the
sardine-anchovy alternation process exists in Oman. Therefore, the objectives of
this chapter is first to investigate whether sardines and anchovies stocks in Oman
are inversely related and identify any problems associated with the sardines and
118
anchovy fisheries of Oman. The second objective is to recommend and define steps
and precautions that should be taken to understand and develop the Omani sardines
and anchovies fisheries.
These recommendations will be based on the lessons
learned from the sardine-anchovy cases described in Chapter 2, the analysis of the
sardine and anchovy qualitative loop models in Chapter 3, and the results from
objective one in this chapter.
METHODODLOGY
I was unable to locate any published references on the history of the fishery
for sardines and anchovies in Oman. To fill this gap, I interviewed a number of
sardine and anchovy fishers in Oman. I selected the questions for the interviews so
the answers would make it easy to infer whether the Omani sardines and anchovies
fluctuate inversely as in the other sardine-anchovy systems.
I conducted all the interviews personally during June 2001 using a set of 17
questions (Table 4.1). Because most of the fishers interviewed were illiterate, I
conducted the interviews face to face and verbally, rather than distributing
questionnaires. Also, the oral interview format allowed me to make sure that the
fishers understood the questions and that their answers were relevant to the
question asked. The questions were designed for sardine and anchovy fishers that
had extensive experience fishing for the two types of fish. Hence, I made sure that
the fishers interviewed were old enough to have a long historical perspective. Their
ages were from 60 to 70 years and they had at least 30 years experience in the
sardine and anchovy fisheries.
119
Table 4.1. The questions used in interviews with the fishers in Oman.
1. How old were the fishermen and how long had they been in
the
sardine/anchovy fishery?
2. What are the sardine/anchovy seasons? Are the seasons related to cold versus
hot months or to the monsoon?
3. Are sardines/anchovy caught together in the same gear?
4. What type of gear is used for the sardines and anchovy and which gear types
are preferred and why?
5. How far
along the shore and offshore does the fishery occur for
sardine/anchovy?
6. How are the sardine and anchovy used and how much are they worth?
7. Which of the two types of fish do the fishermen prefer to fish for and why?
8.
Have there been periods when sardines/anchovy disappeared for a long time? If
so, for how long did they disappear before returning? Do the two types of fish
disappear at the same time or do they alternate?
9. When sardine/anchovy disappear do they disappear completely or can they still
be found in some places? Where?
10, Sometimes a number of fishermen share the same gear. What do they do when
the sardines increase or decrease? Do they buy more gear and fish separately?
11. When sardine/anchovy populations increase/decrease, do the fishermen increase
their fishing frequency? Do they hire new employees to increase their fishing
activities? What do the fishermen do with the extra gear and employees (if
any)?
120
Table 4.1. Continued
12. When sardine/anchovy return after a long disappearance, which of them come
back in higher numbers?
13. When sardine/anchovy expand do they expand along the shore or offshore?
14. Are there periods when the fishermen stop fishing sardine/anchovy even though
they are available? Why do they stop fishing? For how long?
15. Do the fishermen catch for other types of fish besides sardines/anchovies?
16. When fish other than sardines/anchovies are available do the fishermen
continue their sardines/anchovy fishing or do they decrease their activities so
they can catch other fish as well?
17. Have the fishermen noticed any changes in the oceanographic conditions (water
temperatures, currents, or appearance of rare species) that they think might be
related with changes in sardine and anchovy abundance?
The responses to question 1 indicate how long the fishermen have been in
the sardine and anchovy fishery. The fishermen that had over thirty years of
experience were asked the remaining interview questions. These fishermen are
most likely to be able to describe any major historical changes in the abundance of
sardines and anchovies.
Sardines and anchovies in the other cases studied are distinguished by their
temperature tolerance, with the sardines usually preferring warm waters and the
anchovies preferring cold waters. The purpose of question 2 (Table 4.1) was to find
out whether the sardines and anchovies in Oman have such temperature
preferences. Question 3 was included to confirm the response to question (2). If
the sardines and anchovies are rarely caught together, it could be because they have
121
different temperature preferences. Regional differences in the responses to these
questions are of particular interest.
The type of gear used for sardines and anchovies and how far offshore the
fishermen operate (questions 4 and 5) probably limit the fishermen's knowledge.
Catch preference may be an important issue with regard to fluctuations in the
landings. The price and the use of the catch (questions 6 and 7) might orient the
fishers to the more expensive fish, in which case landings fluctuations are a
function of market price rather than changes in fish abundance or distribution.
Questions 8 and 9 address the phenomena of sardine and anchovy
expansions and contractions, which occur in other sardine-anchovy systems. The
intent of the questions is to determine whether sardines or anchovies in Oman have
experienced long-term or short-term disappearances and whether there are refuge
areas. Questions 10 to 13 are intended to establish the relative dominance of
sardines versus anchovies and geographic details of the expansion process. In
addition the questions deal with the fishermen's willingness and ability to increase
their fishing activities when the sardines or anchovies increase.
Questions 14-16 deal with the importance of sardines and anchovies to the
Omani fishermen compared to other fisheries. If fishers prefer other fish stocks,
and rank sardines and anchovies as being unimportant, then catch trends in Oman
for sardine and anchovies provide a biased representation of sardine and anchovy
abundance.
Finally, question 17 is a general question that explores the fishers'
observations of changes in ocean conditions, and whether they can relate this
change in the ocean condition with changes in the sardine and anchovy availability.
RESULTS
A total of 30 fishermen were interviewed along the Omani coast (Table
4.2). The interviews covered five out six Omani regions. Musandam region was not
122
included in the survey. A few fishers interviewed were slightly younger (from AlWusta), but still had extensive experience.
In most of the areas, it was difficult to interview one fisher alone. The
majority of the interviews were with more than two fishers who spoke for all of the
sardine and anchovy fishers in that area.
Fishing operations for sardines or
anchovies involve an average of 12 fishers; so the one particular fisher that I
interviewed in fact spoke for the entire crew of 12. Therefore the 30 fishers that I
interviewed represent roughly 360 sardine and anchovy fishers. While interviewing
the fishers, I never interrupted them and I did not follow a fixed interview protocol
and sequence of questions. Rather, I allowed the fishers to describe their work and
the fisheries within the general framework of the questions in Table 4.1. Even with
this less structured approach, most of the answers were consistent. A summary of
the
fishers1
responses to the questions by region is given in Appendix E.
Sardines and anchovies are distributed all along the Omani coast. The
abundance of the two fishes and their seasonal availability vary. The fishers agreed
that the sardine season is during the cooler season of September to April, with peak
abundance during December and January. The fishers added that sardines might
disappear from their gears for a short period of time (up to two months), but
generally they are always available somewhere along the Omani coast. Anchovies,
on the other hand, do not have a particular season but could appear in Omani
coastal waters at any time of the year. After appearing in coastal waters, the
anchovies disappear after one to two months and hardly overlap with the sardines.
The fishers agreed that sardines and anchovies are not found together and agreed
that anchovies mostly tend to appear during the summer months (May-September).
However, the fishers again repeated that this is not always true.
123
Table 4.2. Geographic distribution of the number of fishers interviewed along
the Omani coast. The regions are listed from North to South.
Region
Number of
Fishermen
Al-Batinah
Shinas
Majis
Saham
alkhabora
2
4
2
3
Muscat
Alseeb
Alazaiba
Barka
6
2
3
Alsharqiya
Sur
Alashkhara
1
3
Al-Wusta
Masira Island
2
Dhofar
Mirbat
Taqa
Total
1
1
30
If it happens that sardines and anchovies overlap in abundance, which rarely
happens, the fishers in Oman target them separately. Fishers in all regions use the
same types of gear for sardines and anchovies: beach seine, purse seine, and,
occasionally, cast nets. Larger mesh sizes are used for sardines and smaller and
thinner mesh for anchovies.
Because the fisheries occur next to the beach, the
fishers can shift the gears easily to catch one species over the other, according to
the market demand, especially the fishers in the Muscat region. Fishers can shift to
using purse seines when the schools are big or the demand is high. In all regions,
the unused gear is put aside on the beach and covered with a heavy blanket until the
next season. Beach seines and purse seines are modified gilinets that require up to
124
12 people to set and haul. The fishers can catch from 30-40 tons a day when the
sardines are plentiful. Beach seines, in general, are the most common type of gear
used for sardines or anchovies. The geographic scope of the fishing operations are
limited, however, by the traditional gear types used by the fishers, which can only
operate near shore to depths of 3 to 5 meters. Fishing also is limited by
political/social boundaries with the adjacent fishing villages.
The landings of both sardines and anchovies are used as food for humans
(either dried or fresh), as fertilizer (dried) and as food for cattle (dried). Even
though dried sardines and anchovies fetch a higher price, most fishers only sell
fresh fish because the dried fish require attention and space to prepare. Also,
conflicts can arise with neighbors because of the odor resulting from the drying
process.
People in the different regions prefer different fish, which is reflected in the
sardine versus anchovy prices. In the Dhofar region, including Masirah Island and
Wusta, people always prefer sardines, which are more expensive than anchovies. In
contrast, most of the people in the northern
part
of Al-Batina coast, Shinas and
Majis (Figure 4.1), like to eat anchovies, which drives up the price for anchovies in
this region and explains their greater abundance in the market. Dried anchovies are
brought from all over Oman to the Al-Batina market, making them plentiful there.
Fishers in the Muscat area are close to the Al-Batina, yet target sardines because
they are plentiful. However, the Muscat fishers do not mind fishing for anchovies
when they are available due to the favorable price for anchovies in the nearby Al-
Batina market.
Fishers from the Muscat area also have other advantages over
fishers in other regions. There are more fishing companies willing to buy sardines
or anchovies at any time.
In the Al-Batina region anchovies are greatly preferred over sardines and
are treated more as a delicacy than in any other region. Anchovies in this region can
therefore cost twice as much as sardines. In Dhofar sardines are preferred and they
exist in higher abundance. In Al-Wusta the people prefer sardines over the
125
anchovies but fishers also monitor the Omani market demand. Hence, fishers in AlWusta target either sardines or anchovies according to market demand.
In the Muscat region, targeting sardines and/or anchovies depends on the
fishers. There are two types of fishers in Muscat - the year-round and the part time.
The year-round sardine and anchovy fishers target sardines and anchovies
regardless of the market demand. The other fishers enter the fishery on a temporary
basis when there is strong market demand. The year-round fishers supply the local
markets and sell their surplus catches to certain fishing companies that "give away"
these sardines and anchovies to other fishers that use the fish as bait for more
valuable fishes. These other fishers, in turn, are obligated to sell their catch to the
companies. The part-time sardine-anchovy fishers enter the fishery when the high
demand includes nearby country markets.
Anchovies sometimes appear for short periods in very large volumes and
when they do, the fishers make a good living. When there are high volumes of
anchovies, numerous additional fishers enter the fishery. The fishers use the same
gear to catch anchovies whenever the anchovies appear so they have no expense for
new gear. The fishing is very intense on the anchovies but lasts a very short period
of time.
Sardines and anchovies are sold by the truckload, with each truck holding
up to one ton. The price per truckload differs from one region to another and varies
according to the distance to the market place. The main markets for sardines and
anchovies are located in Muscat, Al-Batina, and in Dubai in the United Arab
Emirates. Most of the sardines landed in Oman are sold in the Dubai market.
Anchovies, however, are often sold in local markets because the demand is very
high.
Because the Dhofar region is very far from all these major markets, all of
region's sardine and anchovy landings are sold locally to local farmers and cattle
owners when the catch is small. When the catch is larger, the fishers sell their catch
to middlemen who transport the fish in refrigerated trucks to markets in Al-Batina
or Dubai where the prices are higher. The farmers in Dhofar create enough demand
126
to generate
a year-round fishery for sardines and anchovies. The price in the
Dhofar region ranges from 25-30 Omani Rials per truckload (1 Omani Rial is
equivalent to 2.58 US dollars). In the Al-Batina region the price ranges from 50200 O.R. per truckload.
The fishers have two ways to sell their catch to middlemen: sell directly for
cash, or allow the middleman to sell the catch in the market on commission,
splitting the proceeds after deducting for the middleman's fixed costs. Some
middlemen ensure their access to of fish by providing the fishers with fuel and bait.
In the Dhofar and Wusta regions, the price and fish abundance must be very
high for new fishers to enter the sardine or the anchovy fishery because of the
greater distance between these regions and the major markets. The Al-Wusta
fishers do not enter the fishery until the price goes up to 80 O.R. per truckload, at
which point they transport their catch either dried or refrigerated to Muscat and AlBatina, or to Dubai if the prices are higher.
Al-Sharqiya was the only region where there was no fishery targeting
sardines or anchovies, regardless of the market demand. The fishers at Al-Sharqiya
believe that sardines and anchovies are food for the larger pelagic fish species that
the Al-Sharqiya region is famous for. In this area, the fishers from the ports of Sur
and Alashkhara (Figure 4.1) focus on tuna. The fishers believe that small pelagic
fish are the primary food for the tunas and other valuable large pelagic fishes, and
so they rarely target sardines. The fishers in some villages in the Sharqiya region
have mutual agreements to not fish for sardines in order to maintain the tuna in
their waters. Target fishing for sardines is allowed only if the catch is to be used as
bait for tuna or other fish, The fishers also report to the local authorities any fishers
that target the sardines for any other use.
All the fishers in all the regions agreed that it is not the abundance of
sardines or anchovies that is the major control over whether or not they enter the
fishery;
it is the market demand for the fish. When the demand for sardines
increases, the fishers do not buy more gear or expand their operations. They
continue fishing each day until the demand is satisfied. They believe that what they
127
can catch in a day is enough. If the fishers need to add employees, they do so with
family members and not outsiders. These part-time helpers receive wages for the
short periods of employment and are laid off when their help is no longer needed.
This flexibility in the fishing workforce does not negatively affect the workers or
their communities. When the extra market demand for fish relaxes, the extra fishers
are laid off and most fishers quit the fishery. The fishing gear is covered and stored
on the beach until the next increase in market demand. The fishers also stop fishing
for sardines or anchovies when there are market demands for other more valuable
fish.
In the Al-Wusta and Dhofar regions, the fishers stop fishing for sardines
and anchovies from June to August, when sea conditions are too rough for fishers
to operate their traditional gears. They resume fishing for sardines when sea
conditions become favorable again. But not all the fishers follow this pattern. In
some places, fishers can still fish for sardines even during the monsoon. In areas
such as Mirbat in the Dhofar region, the sea is generally calm and fishers can safely
operate in all seasons. None of the fishers in the Muscat and Batina regions stop
fishing due to bad oceanic conditions, but they do stop fishing when there are poor
market conditions or the fish are unavailable.
Around Masirah Island in the Al-Wusta region, the fishers indicated that
anchovies could always be found in deeper waters or in specific geographic
locations but they are not targeted. Other fishers from Muscat pointed out that
anchovies are always present in coastal waters in areas such as Khabora, Shinas,
and Majis along the Batina coast. When the fishers from these areas were
interviewed, they all answered the questions no differently than the others from
Muscat. The fishers find anchovies only for a short time but have sardines almost
all of the time.
The fishers generally believed that sardines were present in Omani coastal
waters all year and in greater abundance than anchovies. However, the fishers from
Al-Wusta said that anchovies were more abundant in the 1 970s, and the fishers
from Al-Batjna said that anchovies were more abundant in the 1950s. During those
128
periods the sardines did not have a defined season. The fishers did not remember
how long the anchovies remained plentiful before they disappeared. The fishers
from Muscat and Dhofar also indicated that anchovies had been abundant in the
past but they did not know when. The fishers from Al-Sharqiya never targeted
sardines or anchovies and therefore were not familiar with their patterns of
historical abundance.
The fishers interviewed did not indicate any notable relationship between
changes in sea conditions and changes in the sardine and anchovy populations.
However, they did say that when anchovies were available more fish in general
were available, including tunas and kingfishers. The fishers from Al-Batina and
Muscat said that even agriculture was much better during those periods.
DISCUSSION
There was no evidence from the fishers that the abundance of sardines and
anchovies in Oman is inversely related as it is with the cases reviewed in earlier
chapters. The fishers' knowledge of expansions and contractions of sardines and
anchovies, however, is limited by their small-scale fishing technology and the
relatively low importance of these two fish stocks to the Omani fishers. Studies of
sardines and anchovies in the Indian Ocean do not show any inverse relation,
although these studies indicate short-scale fluctuations in sardine biomass
(Figure 4.3). The reason given was that poor year classes resulted when summer
monsoon conditions had an adverse effect on sardine spawning (Gopinathan 1974,
Raja 1973).
Temperature and salinity fluctuations were associated with rapid
declines in the Indian oil sardine population, particularly in the juvenile portion.
Low sea surface temperatures prevent the juvenile sardines from entering the
fishing grounds (Bensam 1970). Increased phosphate concentrations in the water
during the monsoon were also found to be associated with decreases in the sardine
biomass (Subrahmanyan 1959, Annigeri 1969). Based on my review of the
129
literature, there is no evidence that sardines and anchovies in the Indian Ocean
undergo 30 to 40 year alternations as occur in Peru, Japan, and California.
355
305
255
205
155
105
V
I
,V_'
V.
V
,.- .-'
-
5
\.
/
\
1._I
/
F]
I
Ii 1) --
00 N
- N NN 00
C
O
I
III
00
11
00
Q O
C ,- C - - C- C- C- C- O- - - -
C
00
O
.-4
F
II
N
rO 00
O
O
O
C
- -
Year
Sardines - - - - Anchovies
Figure 4.3. Western India total sardines and anchovies landings. Data
obtained from FISHSTAT Plus: Universal software for fishery statistical time
series. Version 2.3 2000
There are a number of cyclical environmental changes that induce
variability in the abundance of sardines and anchovies around the world. These
environmental changes involve changes in the sea surface temperature and
corresponding atmospheric changes due to the shifting air-sea relation. Sardines
and anchovies fluctuate in abundance as the environment shifts from one regime
(cold) to another (wann). Yasuda et al. (1999) show similarities between the
abundance of the Japanese sardine and the regime shift indexed by North Pacific
Index (NPI). Scientists were able to identify a similar relation in the Indian Ocean
130
between the ocean and sea surface temperatures which causes interarmual climate
variability. The Indian Ocean air-sea interaction, called the dipole mode (Saji et al.
1999), is believed to be independent from the ENSO.
The dipole mode events are characterized by cooler sea surface
temperatures off Sumatra in the southeast tropical Indian Ocean when the western
tropical Indian Ocean is warmer than usual. It is also characterized by strong
easterly winds along the equator in the tropical eastern Indian Ocean. To quantify
the dipole mode, researchers defined an index called the dipole mode index. During
a positive dipole event, there are warm sea surface temperatures (SST) in the
regions around Oman; there are cold sea surface temperatures during a negative
dipole event. This dipole phenomenon occurs about every four to five years in the
Indian Ocean (Saji et al 1999). The dipole mode, however, does not correlate with
the landings of sardines or anchovies in the Indian Ocean and it is not known how it affects their abundance.
When compared with the sardine and anchovy species of California, Peru,
Japan, and South Africa, Oman has many more species of each, six sardine species
and six anchovy species. In the other sardine-anchovy systems there is generally
only one species of sardine and one anchovy. Speciation involves natural selection
of life-history traits that permit individual species to maintain the flow of genetic
material from one generation to next, but prevent the exchange of genetic material
between individuals of different species. The life-history traits and life cycle
become tuned to recurring predictable environmental conditions such as seasonal
changes in temperatures and currents, Hence, individuals from different species
must be separated when spawning to prevent the flow of genetic materials between
species. Individuals of the same species must aggregate to maintain genetic
exchange among individuals of the same species (Sinclair 1987). The variety of
sardine and anchovy species in Oman could indicate greater diversity of
environmental habitats when compared to the other sardine-anchovy systems.
This diversity allows or perhaps encourages speciation.
131
The sardine and anchovy species found in Oman are also found in the
countries that border the Indian Ocean. The seasonal shift in winds due to changes
in the Indian Ocean atmospheric pressure in summer versus winter causes reversals
of the current along the coast of East Africa that extends to Oman and Iran
(Sheppard 2000). The existence of several species of sardines and anchovies along
the Omani coast could be due to the migration of fish to the north during the
southwest monsoon and southward during the southeast monsoon.
The geography of the area around Oman is also diverse. The oceanic bodies
include the Arabian Gulf (Persian Gulf), the Red Sea, and the Gulf of Oman. This
diversity of environments could be the reason for the existence of so many species.
Some of the sardine and anchovy species in the region have been recorded in the
Red Sea but not in Oman; others have been reported in northern Oman but not in
the south. In California, Peru, Japan, and South Africa the geographic diversity is
much more limited than around Oman.
It is easier for the inverse cyclic behavior to occur in a system composed of
two species, one sardine and one anchovy, than in a system with six species of
sardines and six of anchovies. A system with so many species of sardines and
anchovies indicates a diverse system supporting speciation. A system with only one
species of each suggests that an homogenous system does not support speciation.
Therefore shifts in the environment in a homogenous system are likely to either
support the one species or the other. In a diversified system, it is difficult to
conceive how an environmental change could support the complete set of, say, the
anchovy species and depress the set of sardine species, while simultaneously
providing sufficient diversity to maintain all the individual species within each set.
Some of the fishers in Oman indicated that when anchovies used to be
plentiful, agricultural production was much better and fish were more abundant
near the shore. One possible the reason for the simultaneous decline in agriculture
and fish abundance could be the construction of dams that occurred in Oman during
the 1 980s. The dams might prevent underground fresh water from reaching the
coastal agricultural areas in Muscat and south AlBatina resulting
in the
132
groundwater becoming more saline. All the fishers complained about declines in
overall fish landings. In the 1970s and 1950s, the overall fish landings were much
larger than now. The fishers related these years with the high abundance of
anchovies as well. The fishers all agreed that the bigger vessels that now fish in the
offshore waters are the reason behind the decreased availability of fish inshore.
The offshore fishing fleet did not operate in Omani waters until the 198 Os.
Regarding possible refuge areas for anchovies, some fishers indicated that
anchovies are always abundant in Al-Batina coast, but the fishers from Al-Batina
denied that anchovies are always present in their waters. The fishers from AlBatina suggested that the demand for anchovies in Al-Batina is always high and
that fishers from all over Oman bring their anchovy catch to the Al-B atina market.
Although anchovies might always be available in the market, they are not always in
the waters around Al-Batina.
The sardine and anchovy cyclic behavior in Oman is not evident as far as
the results of this study are concerned. However, the current study should be
considered a preliminary investigation into the fluctuation phenomena between the
sardines and anchovies. The results of this research should be reconfirmed in the
near future. The following sections include recommendations on how to proceed
with the Omani sardine and anchovy fisheries.
Future of the Omani Sardine and Anchovy Fishery
Based on my review of the sardine-anchovy systems in California, Peru,
Japan, and South Africa, I concluded that refuge areas are very important for
persistence of the sardine and anchovy inverse cyclic relation. Since this inverse
cyclic behavior is not evident in Oman, this may indicate that refuge areas do not
exist in Oman. As indicated by the loop analysis of this thesis, when refuge areas
are not included in a system that has fishing, then fisheries management should
focus on protecting the less abundant fish stock. Fishery management in this case
133
can be viewed as providing the self-regulation for the less abundant fish stocks.
Sardines and anchovies spawning areas in Omanhave not been identified.
Fishermen might accidentally be fishing these areas. Therefore these areas should
be identified to prevent the fishers from exploiting them. These spawning areas
could be equivalent to the refuge areas in California, Japan, Peru, and South Africa,
as all the refuge areas in the different cases were also identified to be the major
spawning areas as well.
Based on the interviews with sardine and anchovy fishers in Oman, the
lessons learned from other sardine-anchovy systems, and the qualitative analysis of
sardine-anchovy systems, the points given below should be addressed for managing
and developing the sardine and anchovy fisheries in Oman. Although dramatic
alternations in sardine and anchovy abundance are not an important feature in
Oman, there are other crucial issues that should be considered.
Research Issues
The diversity and stock composition of Omani sardines and anchovies
should be closely studied. It is important to identify the less abundant species
because in a mixed-stock fishery it is possible to drive the less productive stocks to
extinction even when overall rates of exploitation are moderate. The sardine and
anchovy fishers in Oman use the same gear to fish all the sardine and anchovy
species. If the weaker species intermix on the fishing grounds with the productive
ones, the lack of differentiation by the fishing operations could unknowingly harm
the less abundant species. Therefore. research is needed to monitor the species
composition of the sardine and anchovy catches in Omani waters.
Knowledge of where and when the sardines and anchovies live, spawn, and
feed is important for determining whether Oman needs to collaborate with
neighboring countries to study and manage the fish stocks. Defining concentration
areas or spawning areas along the Omani coast is the current priority because there
Ii
is a danger that fishers might unknowingly discover and target these zones when
market conditions are favorable. These zones should be identified by egg and larva
surveys and the characteristics of the fish populations in them carefully studied.
Closing these zones to sardine and anchovy fishing and restricting the by-catch of
sardines or anchovies through gear changes may be necessary. Special attention
should be given to the anchovy populations because they exist in lower abundance
than the sardines. Increases in market prices could encourage the fishers to
overexploit anchovies if fishing is not controlled. Sardines and anchovies are short-
lived species, and high levels of exploitation could result in recruitment failure and
a succession of poor year classes. So far, fluctuations in sardine and anchovy
landings in Oman have been due mainly to changes in market conditions.
Studies of the relationships between sardines and anchovies and conditions
in their spawning areas are very important for understanding the preferred
environments. The dynamics of the local ocean environment in Oman needs to be
more closely studied to establish the oceanographic processes that generate the
conditions preferred by sardines and anchovies. Correlation studies should be
conducted among the local ocean environment and atmospheric change and the
response of the sardines and anchovies to such changes. For example, upwelling
indices, sea surface temperatures, salinity, and the type and amount of food
available should be correlated with the sardine and anchovy spawning activity.
These variables proved to be important in the sardines and anchovies life cycles in
the cases studied. Therefore measurements of these variables should be taken on a
regular basis. Long term environmental data are important for understanding
fluctuations in the sardine and anchovy catches. Ocean currents and the monsoon
cycle in Oman and their combined effect on the sardine and anchovy populations
are poorly understood at present and should be also studied.
135
Management Issues
The sardine and anchovy fisheries are a secondary source of income for
most fishers in Oman. Consequently, a simple form of management should control
the fishery, although greater complexity may be required if the fisheries develop.
The fishers involved in these two fisheries and the gears they use should be
licensed, with the government limiting the number of licenses according to their
assessment of the stocks via monitoring of the landings. Because the fishery is
market driven, closed seasons or areas could be established as needed to protect the
spawners or the less abundant species.
To avoid the involvement of too many fishers when the market price
increases , the number of fishers has to be controlled. Fishers who are licensed to
harvest sardines and/or anchovies could transfer or sell their license to other fishers.
Since no harvest quota is involved, only licenses are going to be transferred and the
transaction might involve new fishers. License transfers or sales should be
officially registered to keep track of the license holders
The middlemen in Oman manipulate the fisheries because they are monitor
the market prices. The number of these middlemen should be controlled. As more
middlemen become involved, there could be more pressure to catch sardines and
anchovies during periods of high demand. The middlemen who transport sardines
and anchovies within Oman or to neighboring countries should be licensed as well.
Border taxes can be established on the middlemen as needed
136
RECOMMENDATIONS
The following recommendations are in priority order.
Recommendation 1: Study and describe the Omani sardines and anchovies,
their environment, and their interactions with other species.
The most immediate task is to determine the magnitude of the sardines and
anchovy stocks in Oman. Sardine and anchovy population size can be assessed
acoustically as is done with the stocks in South Africa. Acoustic techniques involve
the use of an echo-sounder or sonar to estimate the dimensions and packing
densities of schools of fish. Pulses of acoustic energy released from an echosounder strike the fish school and bounce back to be detected by a transducer. Fish
then may be counted and their size estimated by the density and amplitude of the
echo signals. Mark-recapture methods also can be used to estimate and assess
stock size. A known number of marked fish is released into a fish stock. The
proportion of recaptured marked fish in subsequent catches is used to estimate the
stock size (King 1995). These methods of stock size estimation can be repeated as
needed.
Sardine and anchovy spawning grounds can be identified through egg
surveys and gonad studies Thereafter, different environmental measurements
should be taken in the identified spawning areas, including sea surface temperature
and upwelling indices. The environmental data should then be correlated to the
sardines and anchovies spawning activities. Having better knowledge of these
spawning areas will allow the fishery managers to protect the areas and close them
whenever it is necessary. More studies on the food habits of pelagic fish species
should be conducted so that there is better understanding of the linkages between
prey species such as sardines and anchovies, and the valuable predator species such
as tunas and kingfish.
In the past, the Omani government has supported
development of fisheries that target large pelagic fishes. It may be very important
137
to preserve the small pelagic fish species as a food resource for these larger, more
valuable fish.
Sampling programs for collecting long term environmental data should be
established in Oman and the data evaluated in relation to changes in the sardines
and anchovies stocks. Measurements of sea surface temperature, air temperature,
wind direction and speed, salinity, and phytoplankton and zooplankton
concentrations should be regularly collected from different stations along the
Omani coast.
Recommendation 2: Establish a sardine and anchovy management plan and
fishers license system , and monitor effort and landings.
Stock assessment and monitoring techniques for sardines and anchovies
should be designed to minimize the risk of underestimating the amount of fish
harvested and landed. To assist the monitoring process, the government should
license the number of fishers and the types of gear. Better reporting mechanisms
for catch and landings need to be designed and implemented.
Additionally, to
improve the tracking of landings, the middlemen should be licensed to transport
sardines and anchovies across the border.
Recommendation 3: Develop the sardine and anchovy fisheries cautiously.
In general, the landings of sardines and anchovies are highly variable. This
variability can be of serious consequence for fishers that become overly dependent
on these fish species. Full development of the Omani fisheries for sardines and
anchovies will make both fishers and industry subject to fluctuations in abundance.
Fluctuating fish stocks will result in variable market prices, which, in turn,
influence the fishers' earnings.
Development of a modern fishery for sardines and anchovies would require
education of the fishers and financing of new equipment. The fishers would have to
138
replace their boats with larger ones that can travel further offshore and deploy
larger fishing nets to catch more sardine and anchovies. This will be difficult
because the fishers at present cannot afford to buy new boats.
The Omani fishers do not complain about fluctuations in the abundance of
sardines and anchovies because they do not depend completely on them. In fact,
fishers mostly catch sardines and anchovy stocks when prices are favorable.
Establishing steady markets for sardines and anchovies is likely to be the factor
limiting development of these fisheries.
The critical question that the Omani government must answer is whether
sardines and anchovies in Oman are worth all this attention.
The answer is yes,
but not necessarily for direct economic reasons but rather to preserve the sardines
and anchovies as feed for other valuable fish species.
1-'
13
CHAPTER FIVE: SUMMARY AND CONCLUSIONS
The overall objective of this research was to understand the inverse relation
in the abundance, of sardine and anchovy stocks around the world. Four cases were
used for this purpose: the California, Japanese, Peruvian, and South African sardine
and anchovy stocks. The literature review (Chapter Two) resulted in a conceptual
model for sardine-anchovy systems. The model hypothesizes that when sardines
expand to high levels, anchovies contract to lower levels and retreat to their refuge
areas. These refuge areas provide suitable habitat with optimal food and
environmental conditions where the contracted fish stock can be found all the year
around and persist in spite of the dominance of the other species. The contracted
fish, anchovies for example, will remain in their refuge areas until an "event"
allows them to expand and take over the system again. Simultaneously, the
previously expanded sardine population contracts to its own refuge area.
The review focused on identifying the "event" that triggers the fluctuation
between sardines and anchovies, where the sardines and anchovies expand to and
what limits their expansion, and where the fish contract to. The review also focused
on identifying what changes occur with the fish while they are at high levels versus
when they are at low levels. I found that the sardines and anchovy alternations are
mainly triggered by long-term environmental changes referred to as regime shifts in
which environmental changes from cold to warm and vice versa. The patterns of
sardines and anchovies fluctuations generally coincide with such regime shifts.
Sardines were correlated positively with the warm regimes and anchovies with cold
regimes for the Pacific Rim cases, California, Japan, and Peru. With the South
Africa case, I found less support for the hypothesized scenario because both
sardines and anchovies in South Africa have almost the same environmental
preferences. In addition, regime shifts in South Africa are not as evident as with the
Pacific Rim cases. I found that the expansion of the sardines and anchovies is a
function
of habitat suitability,
mainly in
terms
of temperature.
The
140
sardines/anchovies will expand as far as the environmental conditions allow. The
expansion of sardines and anchovies is limited by many factors including
temperature, predation, and distance between the spawning and nursery grounds.
The sardines/anchovies retreat to their refuge areas when environmental conditions
become unfavorable. These refuge areas are generally coincident with the spawning
areas where fish are found all year round. Changes that occur with the
sardines/anchovies when at high levels reflect density dependent processes.
In Chapter Three, I used loop analysis techniques to find plausible
qualitative models that predict the inverse relation between sardines and anchovies.
The basic models analyzed included four variables: fishing, the sardine and
anchovy populations, and their food resources. The objective also was to find the
best models for representing the sardine and anchovy system and to identify
important interactions between sardines and anchovies.
In the four by four and the five by five models, I found that in the model
that represents the sardine and anchovy system both fish stocks are self-regulated.
Without self-regulation, the sardine-anchovy populations do not alternate or the
system is unstable and one stock goes extinct. Based on the review chapter, self-
regulation in sardines and anchovies was found to be a function of habitat
suitability. Sardine/anchovies will continue provided there are regions of suitable
habitat. In the four and five variable models, the type of interaction between the
sardines and anchovies was unimportant
The number of conditionally stable models increased when one or both fish
species had no self-regulation. This emphasizes the importance of the refuge areas
for the continued existence of sardines and anchovies. The conditions required to
stabilize the system could be satisfied by decreasing the fishing on the contracted
stock. Fishing in this case acts as a form of self-regulation for the contracted fish.
In a similar study, fishing was found to affect sardines stock that experience
environmental forcing negatively (Cisneros-Mata et all 1996). Such a stock takes
more time to recover when compared when there is no fishing. Adequate
141
management therefore was also found to be important to prevent economic, if not
biological collapse.
Based on the results from Chapters Two and Three in Chapter Four, I
investigated whether there is any indication of the inverse cyclic behavior in
sardine and anchovy populations in Oman. A set of interview questions was
developed and a series of interviews conducted with sardine-anchovy fishers in
Oman. Based on the interviews results and review of literature on the fisheries in
the region of Oman, recommendations were developed on how to proceed with the
research, management, and development Omani sardine and anchovy.
Sardines and anchovies in Oman are secondary targets to the fishers and
landings a function of market demand. The fishers provided no evidence that
sardines and anchovies in Oman are inversely related. However, the fishers'
experience with the sardines and anchovies is limited by the primitive fishing
technology they use and their poor interest in the two fish stocks. The general
scarcity of scientific research on sardines and anchovies in Oman makes it difficult
to draw strong conclusions about possible sardine-anchovy alternations.
The findings of this study have important implications for understanding
sardine and anchovy populations around the world. Refuge areas are critically
important for the persistence of sardines and anchovies. It is important that fishing
agencies identify these areas and protect the fish when they are at low levels. Also,
the results suggest that the refuge areas have the environmental conditions
preferred by the sardines and anchovies. The proposed conceptual model for the
fluctuations of sardine and anchovy stocks should help to simplify the design of
research on stocks that have not been well studied.
The results from the modeling part of this research, although qualitative,
suggest that studies of possible interactions between sardines and anchovies should
not be given high priority. Instead, it is important to identify the stocks' refuge
areas and closely monitor fishing activities on the weaker stock. This study and
most studies in literature use landings as indices for abundance, because the
142
fisheries have grown proportionally with the fish stocks. Egg and larval surveys
would be more reliable abundance indices, especially for lightly exploited stocks.
The results of the study emphasize the importance of oceanography in the
life of sardines and anchovies. Fishery management calculations of Total
Allowable Catches for sardines and anchovies should consider unpredictable
oceanographic changes and monitor the fishery closely. The ioop analysis can be
approached differently by replacing the fishing with biological predators and the
inverse predictability between sardines and anchovies analyzed. Are the results the
same as when using fishing only? The research also considered only two fish
populations.
Would the fluctuation process change if more species were
considered in the models.
In addition the results of the current research are qualitative. The models
need to be quantitatively approached. The current research also does not indicate
the length of the sardineanchovy cycle. This could be identified with the
quantitative approach. The research indicates that fluctuations between sardines and
anchovies are a function of long-term environmental changes. Few studies correlate
sardines and anchovies with such environmental changes. More such studies are
needed to confirm the hypothesis.
143
BIBLIOGRAPHY
Ahistrom, E. H 1954. Pacific sardine (sardine) eggs and larvae and other fish
larvae, Pacific coast-1952. U.S. Fish and Wild. Service, Spec. Sci. Rept.; Fish.
(123): 1-76
Ahistrom, E. H 1960. Synopsis on the biology of the Pacific sardine (Sardinops
caerulea). Fish. Biol., 17: 415-451
Al-Barwani, M.A. Prabhakar, A., Don III, J. , and Al Mandhery, M. Studies on the
biology of Sardinella longiceps (Valenciennes) in the Sultanate of Oman, 19851986. Kuwait Bulletin of Marine Science 1989 (10):201-209
Alheit, J, and Bernal, P. 1993. Effects of Physical and Biological Changes on the
Biomass Yield of the Humboldt Current Ecosystem. In Kenneth Sherman, Lewis
M. Alexander, and Barry D. Gold (eds.) 1993. Large Marine Ecosystems, Stress,
Mitigation, and Sustainability. American Association for the Advancement of
Science. Washington, D.C. 53-68 pp
Annigeri G. g. 1969. Fishery and biology of oil-sardine at Karwar. Indian J. Fish.,
16:35-50
Antony Raja 1973. The Indian Oil-sardine: problems and perspective. J. Mar. Biol.
Ass. India, 15 (2): 735-749
Bakun, A. 1990. Global surface air temperatures: update through ocean upwelling.
Science, N.Y 247:198-201.
Bakun, A. 1993. The California current, Benguela current, and south Atlantic shelf
Ecosystems: a comparative approach to identifying factors regulating biomass
yield. In Sherman K., Lewis M. A., and Gold B. D. (eds), 1993. Large Marine
Ecosystems, Stress, Mitigation and Sustainability. American Association for the
Advancement of Science. Washington, D.C. p. 199221
Barnes, J. T., Jacobson, L. D., MacCall, A.D., and Wolf, P. 1992. Recent
population trends and abundance of the Pacific sardine (Sardineops sagax). Calif.
Coop. Oceanic Fish. Invest. Rep 33:60-75
Baugartner, T. R., Soutar, A., and Ferrira-Bartina, V. 1992. Reconstruction of the
history of Pacific sardine and northern anchovy populations over the past two
millennia from sediments of the Santa Barbara Basin. California. Calif. Coop.
Oceanic Fish. Invest. Rep 33:24-40
144
Bensam, P. 1970. On the fluctuation of the Oil-sardine fishery at Cannanore during
1961-1964. Ibid., 17:132-148
Bentley, P. J., Emmet, R. L., Lo N. C.}I., and Moser G. 1996. Egg production of
Pacific sardine (Sardinops sagax) off Oregon in 1994. CALCOFI REP., vol. 37,
193-200, pp.
BCLME 2001 .Benguela Current Large Marine Ecosystem. Available online:
http://www.ioinst.org/bclme/index.htm
Bodini , A 1998. Representing ecosystem structure through signed digraphs. Model
reconstruction, qualitative predictions and management: the case of a fresh water
ecosystem. Oikos 83:93-106.
Bodini , A, and Giavelli G. 1989. The qualitative approach in investigating the role
of species interactioson stability of natural communities. Biosystems, 22, 289-299.
Bottom, D. L., Jones, K. K., Rodgers, J. D., and Brown, R. F. 1993. Research and
Management in the Northern California Current Ecosystem. In Sherman K., Lewis
M. A., and Gold B. D. (eds), 1993. Large Marine Ecosystems, Stress, Mitigation
and Sustainability. American Association for the Advancement of Science.
Washington, D.C. p. 259-271
Boyd, A. J. and Thomas, R. M. 1984. A southward intrusion of equatorial water off
northern and central Namibian in March 1984. Trop. Ocean-Atmos. Newsl. 27:1617.
Boyd, A. J. 1987. The oceanography of the Namibian shelf. Ph.D. thesis.
University of Cape Twon: (xv) +190 pp.(i).
Boyd, A. J., Hewitson, J. D., Kruger, I. andLe Clus F. 1985. Temperature and
salinity trends off Namibia from August 1982 to 1984, and their relation to
plankton abundance and reproductive success of pelagic fish. Cohn scient. Pap.
Commn Se. Atl. Fish 12(1):53-60.
mt.
Brock, J. C., and McClain, C. R,, 1992. Interannual variability in phytoplankton
blooms observed in the northwestern Arabian Sea during the southwestern
monsoon, J. Geophys. Re. 97(C1), 733-750
Brodeur, R. D., and Ware, D. M. 1995. Interdecadal variability in distribution and
catch rates of epipelagic nekton in the Northeast Pacific Ocean. In R. J. Beamish
(ed) Climate change and northern fish populations. Can. Spec. Pubi. Fish. Aquat.
Sci. 121.
145
Brown, P. C., Painting, S. J., and Cochrane K. L., 1991. Estimates of phytoplankton
and bacterial biomass and production in the northern and southern Benguela
ecosystems. S. Afr. J. Mar. Sci. 11:537-564
Burkill, P. H., 1999. ARABESUE: An overview. Deep-Sea Research II 46, 529547
Butler, J. L. and Pikett, D. 1988. Age-specific vulnerability of Pacific sardine,
Sardinops sagax, larvae to predation by northern anchovy, Engraulis mordax, Fish.
Bull., US 86:163-167.
Butler, J. L.
1987. Comparison of the early life history parameters of Pacific
sardine and northern anchovy and implications for species interactions Dissertation
Abstracts International Part B: Science and Engineering [DISS. ABST. TNT. PT. B
- Sd. & ENG.}, vol. 48, no. 2, 261 pp
Butler, J. L. 1991. Mortality and recruitment of Pacific sardine, Sardinops sagax
caerulea, larvae in the California Current Can. J. Fish. Aquat. Sci ., vol. 48, no. 9,
1713-1723 pp.
Butler, J. L., Granados, M. L. G., Barnes, J. T., Yaremko, M., and Macewicz, B. J.
1996. Age composition, growth, and maturation of the Pacific sardine (Sardinops
sagax) during 1994. CALCOFI REP., vol. 37, 152-159 pp.
Cisneros-Mata, M.A, Montemayor-Lopez, U.; Nevarez-Martinez, M.O. 1996.
Modeling deterministic effects of age structure, density dependence, environmental
forcing, and fishing on the population dynamics of Sardinops sagax caeruleus in the
Gulf of California. CALCOFI Rep., vol. 37, pp. 201-208
Cisneros-Mata, M. A., Nevarez-Martinez, M.O. and Hammann, M.G. 1995. The
rise and fall of the Pacific sardine, Sardinops sagax caerleus Girard, in the Gulf of
California. Mexico. Rep. Calif. Coop. Ocean. Fish Invest. 36:136-143
Clark, F. N., and Marr, J. C. 1955. Population dynamics of the Pacific sardine.
Progress Rep., Calif. Coop. Oceanic Fish. Invest., 1 July 1953-30 March 1955:1248.
Coppola S.R., Fischer W., Gribaldi L., Scialabba N., Carpentr K.E. 1994.
Speciesdab: Global database for fishery purpose. User's manual. FAO
Computerized information Series (Fisheries). No. 9. 103 p.
Crawford, R. J. M., Shelton, P.A. and Hutchings, L. 1980. Implications of
availability, distribution and movements of sardine (Sardinops ocellata) and
146
anchovy (Engraulis capensis) for assessment and management of the South African
purse-seine fishery. Rapp. P-v Reun. Ciem 177:355-373
Crawford, R. J. M., Shelton, P.A. and Hutchings, L. 1983. Aspects of variability of
some neritie stocks in the Southern Benguela system. In Proceedings of the Expert
Consultation to Examine Changes in Abundance and Species Composition of
Neritic Fish Resources. San Jose, Costa Rica, April 1983. Sharp, G.D. and J. Csirke
(Eds). FAO Fish. Rep. 291(2):407-448.
Crawford, R. 3. M., Shannon, L. V. and. Shelton, P. A 1988. Characteristics and
management of the Benguela as a large marine ecosystem. In Biomass yields and
Geography of Large Marine Ecosystems. Sheman, K. And L. M. Alexander (Eds.).
Washington, D.C.; American Association for the advancement of Science: 169-219
(AAAS Selected Symposium 111).
Csirke J. 1988. Small shoaling fish stocks. In Fish Population Dynamics, 2'
ed.
Gulland J. A., (ed). John Wiley pp. 27 1-302, 1988.
Cury, P. 1988. Selective pressures and evolutionary novelties: An hypothesis to
understand certain aspects of long-term fluctuations in coastal pelagic fish. Can. 3.
Fish. Aquat. Sci.; 45(6):1099-1107.
Cury, P., Roy C., and Faure, V.1998. Environmental constraints and pelagic
fisheries in upwelling areas: the Peruvian puzzle. S. Afr. J. Mar. Sci., vol. 19, pp.
159-167
Cushing, D.H. 1990. Plankton production and year-class strength in the fish
populations: an update of the match1mismatch hypothesis. Adv. Mar. Biol. 26:249293.
Cushing, D.H. 1981. The effect of Elnino upon the Peruvian anchoveta stock In
"Coastal upwelling" ed. Richards, F. A. Amer. Geophys. Union. 449-457
Deriso, R. B., Barnes, J. T., Jacobson, L. D., and Arenas, P. R. 1996. Catch-at-age
analysis for Pacific sardine (Sardinops sajax). Calif. Coop. Oceanic Fish. Invest.
Rep. 37: 188-192
Dorr III, J. A. 1991. Small pelagics final report for the contract for technical
services for staffing the Marine Science and fisheries Center. Project No. 2720101.1-i
Edeistein-Keshet, L. 1988. Mathematical Models in Biology. Random House, New
York, New York, USA.
147
Ehrhardt, N. M 1991. Potential impact of a seasonal migratory jumbo squid
(Dosidicus gigas) stock on a Gulf of California sardine (Sardinops sagax caerulea)
population. Bull. Mar. Sci, vol. 49, no. 1-2, p. 325-332
Elliott A. J. And Savidge G. 1990. Some features of the upwelling off Oman.
Journal of Marine research, 48, 319-333.
FAO fisheries Department, fishery information, Data and statistics unit. 2000.
FISHSTAT Plus:Universal software for fishery statistical time series. Version 2.3
2000.
Favorite, F., Dodimead, a. J. and Nasu, K. 1976. Oceanography of the subarctic
Pacific region, 1960-71. Bull. Tnt. North Pac. Fish. Comm., no. 33, p. 1-187
Fieux, M. And H. Stommel 1977. Onset of the southwest monbsoon over the
Arabian Sea from marine reports of surface winds: structure and variabilty. Mon.
Weth. Rev., 105, 23 1-236.
Findlater, J., 1969. A major low level air current near the Indian Ocean during the
northern summer, Q.J. R Meterol. Soc., 95, 362-3 80
FishBase 1997. FishBase 97 CD-ROM. Manila.
Francis, R. C., and Hare, S.R. 1994. Decadal-scale regime shifts in the large marine
ecosystems of the North-east Pacific: A case for historical science. Fish. Oceanogr,
vol. 3, no. 4, p. 279-29 1
Francis, R. C., Hare, S.R., Hollowed, A.B., and Wooster, W.S. 1998. Effects of
interdecadal climate variability on the oceanic ecosystems of the NE Pacific. Fish.
Oceanogr, vol. 7, no. 1, p. 1-2 1,
Funakoshi, S. 1992. Relationship between stock levles and the population structure
of the Japanese anchovy. Mar. Behav. Physiol., vol. 21, p. 1-84
Gammeisrod, T., O'Toole M. J., Barthouse, C. J., Boyer, D. and Filipe, V. L. 1998 Intrusion of warm surface water layers at the Southwest African coast in FebruaryMarch 1995:The BenguelaNino 95:S.Afr. J. Mar. Sci. 19 p,41-56
Glantz, M. H., and Thompson J. D.1981. Resource management and environmental
uncertainty: lessons from coastal upwelling fisheries. New York: Wily p. 491
Glantz, M. 1999. The Aral Sea: going, going
p. 20-21;
...
almost gone. Calypso Log; Feb(3),
148
Gopinathan, C. K. 1974. Early stages of upwelling and decline in oil-sardine
fishery of Kerela. J. Mar. Biol. Ass. India, 16 (3): 700-707
Gotelli, N. J. 1998. A primer of ecology. Sinuer Associates, Inc. Sunderland,
Massachusetts, USA 236 pp.
Hare, S.R., Mantua, N. J. and Francis, R. C., 1999. Inverse Production Regimes:
Alaska and West Coast Pacific Salmon. Fisheries, vol. 24, no. 1, p. 6-15
Hare, S.R., and Mantua, N. J.2000. Empirical evidence for North Pacific [climatic]
regime shifts in 1977 and 1989. Prog. Oceanogr., vol. 47, no. 2-4, p. 103-145
Hargreaves N.B; Ware D.M; McFarlane, GA. 1994. Return of Pacific sardine
(Sardinops sagax) to the British Columbia coast in 1992. Can. J. Fish. Aquat. Sci.,
vol. 51, no. 2, pp. 460-463
Hauto-soberanis, L., and Liuch-Belda, D. 1987. Mesoscale cycles in the series of
environmental indices related to the sardine fishery in the Gulf of California. Calif.
Coop. Oceanic Fish. Invest. Rep 28:128-134
Hernandez-Vazquez, S. 1994. Distribution of eggs and larvae from sardine and
anchovy off California and Baja California, 1951-1989. CALCOFI REP, vol. 35, p.
94-107.
Hickey B. M. 1998. Coastal oceanography of western North America from the tip
of Baja California to Vancouver Island. In A. R. Robinson and K. H. Brink, editors,
The Sea, Vol. 11: The Global Coastal Ocean - Regional Studies and Syntheses, p.
345-393.
Houde, E.D. 1987. Fish early life dynamics and recruitment variability. Am. Fis.
Soc. Symp. 2:17-29
Hunter, J. R. And Kimbrell, C. A. 1980. Egg cannibalism in the northern anchovy,
Engraulis mordax. Fish. Bull., 78(3), 811-816.
Hutchings, P., Frouin, P. and Hily, C. 1998. Two new species of Spionidae
(Polychaeta) from Tahiti, French Polynesia. Proc. Biol. Soc. Wash,, vol. 111, no. 4,
p. 799-806
Hutchings, P. 1982 Fish harvesting in a variable, productive environment searching
for rules or searching for exceptions? S. Afr. J. Mar. Sci. p. 297-318
Ichimura, S. 1965. A short review on primary production in the Kuroshio. Inform.
Bull. Planktol. Jap. 12:1-6.
149
Jacobson, L. D., and Thomson, C. J. 1993. Opportunity costs and the decision to
fish for northern anchovy. N. Am. J. Fish. Manage., vol. 13, no. 1, PP. 27-34
Kasai, A., Kish, M. J., and Sugimoto, T. 1995. Influence of wind-induced currents
on the recruitment of Japanese sardine (Sardinops melanostictus) eggs and larvae,
p. 547-552. In R.J. Beamish (ed.) Climate change and northern fish populations.
Can. Spec. Pubi. Fish Aquat. Sci. 121.
Kawai, H. 1998. A brief history of the recognition of the Kuroshio. Prog Oceanogr,
41 :-578,
Kawaski, T. and Omori, M. 1988. Fluctuations in the three major sardine stocks in
the Pacific and global trend in temperate. International symposium on long-Term
changes in Marine Fish population in Vigo, Spain in 1986, p. 3 7-52
Kawasaki , T. 1991. Long-term variability in the pelagic fish populations and their
environment. T. Kawasaki, S. Tanaka, Y. Toba and A. Taniguchi (eds). Oxford:
Pergamon Press, p 7-60.
Kawasaki , T. 1992. Mechanism governing fluctuations in pelagic fish populations.
S Aft. J. Mar. Sci. 12: 873-879.
Kawasaki, 1. 1993. Recovery and collapse of the Far Eastern sardine. Fish.
Oceanogr. 2:244-25 3
Kawasaki , T. and. Omori, M. 1995.. The impacts of climate change on Japanese
fisheries, p 523-528. In R. J. Beamish (ed) Climate change and northern fish
populations. Can. Spec. Publ. Fish. Aquat. Sci. 121.
Kifani, S. 1998. Climate dependent fluctuations of the Moroccan sardine and their
impact on fisheries. In Global versus local changes in upwelling systems -international conference under the auspices of the Climate and Eastern Ocean
Systemes Project (CEOS), ORSTOM, Paris (France), p. 236-248,
Kifani, S., and Gohin, F. 1992. Pelagic fish recruitment success and reproductive
strategy in upwelling areas: Environmental compromises. S. Afr. J. Mar. SCI vol.
l2p. 135-146
Kikuch, T. 1958. A relation between the alternation between good and poor
catches of sardine and the establishment of Shinden and Naya villages. Memorial
works dedicated to professor K. Uchida, p. 84-92
150
King, D. P. F. 1977a. Distribution and relative abundance of eggs of the South
West African sardine Sardinops ocellata and anchovy Engraulis capensis, 1971/72.
Fish. Bull. S. Afr. 9:23-3 1.
King, D. P. F. 1977b. Influence of temperature, dissolved oxygen and salinity on
incubation and early larval development of the South West African sardine
Sardinops ocellata. Investi Rep. Sea Fish. Brch S. Aft. p. 114:35.
Klyashtorin, L. B., and Smirnov, B. 1995. Climate-dependent salmon and sardine
stock fluctuations in the North Pacific. In R. J. Beamish (ed) Climate change and
northern fish populations. Can. Spec. Pubi. Fish. Aquat. Sci. 121.
Klyashtorin, L. B., and Rukhlov, F. N. 1998. Long-term climate change and pink
salmon stock fluctuations. NPAFC Bulletin Number 1: Assessment and Status of
Pacific Rim Salmonid Stocks, p. 464-479
Krebs, C.J. 2001. Ecology: The Experimental Analysis of Distribution and
Abundance. 5th ed. Benjamin Cummings: Menlo Park, California. 801 pp.
Kuroda, K. 1989. Long-period changes of the spawning grounds of sardine
(Sardinops melanosticta Temminck et Schiegel) at the Pacific coast of southern
Japan. (Results of the studies of biological resources in the Northwest pacific.
Tinro Viadivostock (USSR), p. 117-130
Kuroda, K. 1991. Studies on the recruitment process focusing on the early life
history of the Japanese sardine, Sardinops melanostictus (Schlegel). Bull. Nat. Res.
Inst. Fish. Sd. 3:25-278.
Kondo, K 1980. The recovery of the Japanese sardine-the biological basis of stock
size fluctuations. Rapp. P.-V. Reun. Cons. mt. Explor. Mer 177:332-345.
Lasker, R. l 975. Field criteria for survival of anchovy larvae: the relation between
inshore chlorophyll maximum layers and successful first feeding. Fishery Bull.,
73(3):453-462.
Lasker, R., and MacCall, A.. 1983. New ideas on the fluctuations of the clupeoid
stocks off California. Proceedings of the Joint Oceanographic Assembly 1982
General symposia, p. 110-120.
Laws E. A 1997. El Nino and the Peruvian anchovy fishery. Sausalito, CA
University Science Books. 80 pp.
151
Le Clus, F. and Kruger, I. 1982. Time and space distribution of temperature,
salinity, plankton, and fish eggs off South West Africa in 1980/81 a preliminary
data report. Cohn sciet. Pap. mt. Commn SE. Ati. Fish. 9(2):121-145.
Le Clus, F. 1990. Impact and implications of large-scale environmental anomalies
on the spatial distribution of spawning of the Namibian sardine and anchovy
populations. S. Afr. J. Mar. Sci. 9:141-159.
Le Clus, F. 1991. Hydrographic features related to sardine and anchovy spawning
in the northern Benguela system, comparing three environmental regimes. S. Afr. J.
Mar. Sci. (Rogge Bay); Vol.10, p. 103-124
Lenarz, W., H., Schwing F.B., Ventrsca D. A., Chaves F., and Graham, W. M
1995. Explorations of El Nino events and associated biological population
dynamics off central California. CALCOFI REP ., vol. 36, p. 106-119
Liuch-Belda, D., Crawford, R. J. M., Kawasaki, T., MaCall, A.D., Parish, R. H.,
Schewartsloze, R.A., and Smith, P. E. 1989. World-wide fluctuations of sardine
and anchovy stocks: the regime problem. S. Afr. J. Mar. Sci. 8:195-205.
Liuch-Belda, D., Liuch-Cota, Hemandez-Vazquez D. B., and Salinas-zavala, C.A.
1992. Sardine population expansion in eastern boundary systems of the pacific
ocean as related to sea surface temperature. In Benguela Trophic Functioning.
Payne, A. I. L. Brink, K. H. Mann, K. H. And R. Hilborn (Eds). S Afr. J. Mar. Sci.
12 :147-155.
Liuch-Belda, D., S. Hernandeez-Vazquez and R.A. Schwartzloze. 1991a. A
hypothetical model for the fluctuation of the California sardine population
(Sardinops sagax caerulea). In long Term Variability of Pelagic Fish Populations
and Their Environment, T. Kawasaki, S. Tanaka, Y. Toba and A. Taniguchi, eds.
Proc. Tnt. Symp., Sendai, Japan. 14-18 Nov. 1989. Oxford: Pergamon Press, p. 293300.
Liuch-Belda, D., Luch-Cota, D.B., Hernandeez-Vazquez S, Schwartzloze, R.A, and
Sahinas-Zvala, C.A. 1991b. Sardine and anchovy spawning as related to
temperature and upwelling in the California Current System. Calif Coop. Oceanic
Fish Invest. Rep. 32:105-111.
Lo, N. C. H., Green Ruiz, Y. A., Cervante, M. J., Moser, H. G., and Lynn, R. 3.
1996. Egg production and spawning biomass of Pacific sardine (Sardinops sagax)
in 1994, determined by the daily egg production. CALCOFI REP., vol. 37, p. 160174
152
MacCall, A. D.
1986. Changes in the Biomass of the California Current
Ecosystem. In Kenneth Sherman and Lewis M. Alexander (eds.), Variability and
Management of Large Marine Ecosystems.Boulder: Westview-AAAS Selected
Symposium 99, p. 33-54
MacCall, Alec D. 1990. Dynamic geography of marine fish populations. Wash. Sea
Grant Program; 153 pp.
Mackenzi, B. R., and Leggett W. C. 1991. Quantifying the contribution of smallscale turbulence to the encounter rates between larval fish and their zooplankton.
prey: Effects of wind and tide. Mar. Ecol. Prog. Ser.. Oldendorf [MAR. ECOL.
PROG. SER.], vol. 73, no. 2-3, p. 149-160
Matsuda, H , T. Wada, Y. Takeuchi, and Y. Matsumiy 1991. Alternative models for
species replacement of Pelagic fishes. Res. Popul. Ecol. 33, 41-56.
May, R. M. 1973. Stability and complexity in model ecosystems. Princeton
University Press. Princeton, New Jersey 235 pp.
Minobe, S. 1999. Resonance in bidecadal and pentadecadal climate oscillations
over the North Pacific: Role in climatic regime shifts. Geophys. Res. Lett., vol. 26,
no. 7, p. 855-858
Moser, H. G., Charter, P. E., Smith, D. A., Ambrose, S. R., Charter, C.A., Myer,
Sandknop, E. M., and Watson, W. 1993. Distributional atlas of fish larvae and eggs
in the California Current region: taxa with l000or more total larvae, 1951 through
1984. Calif Coop. Oceanic Fish. Invest. Atlas 31, 233 pp.
Muck, P., Zafara, de Moreno, A, Aranda, C. 1989. The seasonal dynamics of sea
surface temperature and its impact on anchoveta distribution off Peru. THE
Peruvian upwelling ecosystem: Dynamics and interactions., 1989, p. 33-44,
ICLARM conference proceedings. Manila no. 18.
Munroe, T. and M. Nizinski,
1997. Clupeidae. Herrings, sardines. In K.E.
Carpenter and V. Niem (eds.) FAO Identification Guide for Fishery Purposes. The
Western Central Pacific.
Murphy G. I., 1988. Characteristics of Clupeoids. In Fish population dynamics : the
implications for management edited by Gulland, J. A.. Wiley, Chichester [England]
422 p.
Nakai, Z,., and Hottori, S. 1962. Quantitative distribution of eggs and larvae of the
Japanese sardine by year 1949 through 1951. Bull. Tokai Reg. Fsh. Res. Lab.
(9):23-60
153
Nakai, Z. 1962. Studies relevant to mechanisms underlying the fluctuation in the
catch of the Japanese sardine, Sardinops melanosticta (Tmm. et Schieg.) Jap. J.
Icth.,9(1-6): 1-113
Nakata, K., Hada, A., and Matsukawa, Y. 1994. Variations in food abundance for
Japanese sardine larvae related to the Kuroshio meander. Fish. Oceanog. vol. 3,(1),
p. 3 9-49
Nakata, K., Zenitani, H., and Inagake, D. 1996. Did offshore shift of the main
spawning ground of Japanese sardine affect the availability of food for the larvae?
Proceedings of international workshop: survival strategies in early life stages of
marine resources A.A Balkema publishers, Broohfield, VT (USA), pp. 115-124
Nesterov, A.A, 1993. Changes in the distribution and abundance of the Southeast
Pacific as factors affecting long-period variations of pelagic fish stocks. Fishery
investigations in the Atlantic and south pacific oceans, Atlantniro , Kaliningrad
(Russia), p. 124-138, SB Nauch. TR. Atlantniro
Nesterov, A.A, 1996. Historical and recent changes in the abundance of pelagic fish
in the southeast pacific. Fishery biology research by atlantniro in 1994-1995.
Kaliningrad (Russia), 1996, pp. 156-166
Noakes, D.J.; Beamish, R.J.; Klyashtorin, L.; Mcfarlane, G.A. On the coherence of
salmon abundance trends and environmental factors. North Pacific Anadromous
Fish Commission Bulletin; Issue No.1, p. 454-463
Ogawa, Y., and Nakahara, T. 1979. Interrelationships between pelagic fishes and
plankton in the coastal fishing ground of the southwestern Japan Sea. Mar. Ecol.
(Prog. Ser.), 1(2), 115-122
Omani Annual Fisheries statistics 1995. Ministry of Agriculture and Fisheries.
Sultanate of Oman.
O'Tool, M. J. 1977. Investigations into important fish larvae in the South East
Atlantic in relation to the hydrological environment. Ph.D. thesis. University of
Cape Town: pagination discontinuous, 8 sections+7 supporting papers.
Owens, N. J. P, Law, C. S., Brkill, P.R., Mantoura, R.F., Woodward, E.M.S.,
Bellan, I.E., Aiken, J., I-lowland, R.J.M., Liewellyn, C.A., 1993. Size fractionated
primary production and nitrogen assimilation in the northwestern Indian Ocean.
Deep-Sea Res. 40,697-7 10.
Parrish, R. H., Bakun, A.. Husby, D.M. and Nelson, C.S. 1983. Comparative
climatology of selected environmental processes in relation to eastern boundary
current pelagic fish reproduction. In Proceedings of the Expert Consultation to
154
Examine Changes in Abundance and Species Composition of Neritic Fish
Resources. San Jose. Costa Rica, April 1983. Sharp, G.D. and Csirke J. (Eds).
F.A.O. Fish. Rep. 291(3):731-777.
Puccia, C. J. & R. Levins. 1985. Qualitative Modeling of Complex Systems. An
Introduction to Loop Analysis and Time Averaging. Harvard Univ. Press. 259 pp.
Radovich, J. 1982. The collapse of the California sardine fishery: what have we
learned? Calif. Coop. Oceanic Fish. Invest. Rep. 23:56-78
Regner, S.; Gacic, M. 1974. The fluctuation of sardine catch along the eastern
Adriatic coast and solar activity. Acta Adriat.
ll,pp. 1-15
,
Inst. Oceanogr. Ribar, vol. 15, no.
Roemmich, D., and McGowan, J. 1995. Climatic warming and the decline of
zooplankton in the California Current. [Science (Wash.)], vol. 267, no. 5202, p.
1324-1326
Rosa, H., Jr. and T. Laevastu.
1960. Comparison of biological and ecological
characteristics of sadines and related species -a preliminary study. p. 523-552. In H.
Rosa, Jr. and G. Murphy (eds.) Proc. World Scientific Meeting on the Biology of
Sardines and Related Species, 14-21 September 1959, Rome, Italy. Vol. 2, Species
Synopses.
Rothschild, B. J. and Osborn, T. R. 1988. Small-scale turbulence and plankton
contact rates. J. Plankton Res., vol. 10, no. 3, pp. 465-474
Rothschild, B. J. 1995. Fishstock fluctuations as indicators of multidecadal
fluctuation in the biological productivity of the ocean. In Beamish, R. J., Climate
change and northern fish populations. Can. Spec. Pubi. Fish. Aquat. Sci. 121., p.
201-209
Russell, B.C. and W. Houston. 1989. Offshore fishes of the Arafura Sea. The
Beagle, Records of the Northern Territory Museum of Arts and Sciences. 6(1):6984.
Saji, N.H., Goswami, B.N., Vinayachandran, P,N., Yamagata, T., 1999, A dipole
mode in the tropical Indian Ocean, Nature 401, p. 3 60-363.
Santander, H. and DE Castillo, 0.S. 1977. Ichthyoplankton from the Peruvian
coast. In: Proceedings of the symposium on warm water zooplankton, S.Z Qasim,
editor, National Institute of Oceanography. p. 105-123.
155
Schumacher, R. W. 1999. Studies of Marine Macro- and Microorganisms. Part 1.
The Chemistry of Ascidians from the Family Didemnidae. Part II. Metabolites from
Marine Sediment Derived Actinomycetes. Dissertation Abstracts International Part
B: Science and Engineering [Diss. Abst. Tnt. Pt. B - Sci. & Eng.}, vol. 59, no. 12,
p. 6330
Shannon, L.J. 1998. Modelling environmental effects on the early life history of the
South African anchovy and sardine: a comparative approach. In Benguela
Dynamics:Impacts of Variability on Shelf-Sea Environments and their Living
Resources. Pillar S.C., Moloney, C.L., Payne, A.I.L. and Shillington F.A. (Eds).
S.Afr.J.mar.Sci. 19: 291-304.
Shannon. L.V. , Boyd, A. J., Brundrit, G. B., and Tauton-Clark, J. 1986. On the
existence of an El Nino- type phenomenon in the Benguela system. J. Mar. Res.
44(3):495-520.
Shantov, V. P., and Vasilkove V. P. 1982. Long term population fluctuations of
north Pacific sardines. I. Population dynamics of the eastern Sardine, Sardinops
sagax melanosticta, and California sardine, Sardinops sagax caerulea, in the
Twentieth Century.
Shelton, P. A., and Hutchings, L. 1982. Ocean stability and anchovy spawning in
the southern Benguela Current region. Fish. Bull, vol. 88, no. 2, pp. 323-338
Shelton, P. A. and Armstrong, M. J. 1983. Variations in parent stock and
recruitment of sardine and anchovy population in the Southern Benguela system, in
Sharp, G.D. and Csirke, J. (Eds), Proceedings of the Expert Consultation to
examine changes in abundance and species composition of neritic fish resources,
FAO Fish. Rep., 291,3, 1113-32.
Sheperd, J. and Cushing, D.H. 1980. A mechanism for density dependent survival
of larval fish as basis of stock recruitment relationship. J. Cons. Tnt. Mer 39:160167.
Sheperd, 3. G. 1982. A versatile new stock and recruitment relationship for
fisheries and construction of sustainable yield curves. 3. Cons.
mt.
Explore. Mer
40:67-75.
Sheppard, C. 2000. Seas at the Millennium; an environmental evaluation. Volume
II Regional chapters the Indian Ocean to the Pacific 920 pp.
Schwartzlose, R.A, Aiheit, 3., Bakun., Hagen, E., Kawasaki, T., Liuch-belda, D.,
Liuch-Cota, S.E., Maccall, A.D., Matsuura, Y., Nevarez, M.O., Parrish, R.H., Roy,
C., Serra, R., Shust, K.V., Ward, N.M, and J.Z. Zuzunagea. 1999. Worldwid large-
156
scale fluctuations of sardine and anchovy populations. S.Afr. J, Mar. Sci. 21:289347.
Siddeek, M.S.M, Al-Habsi, H. N., Al-Jufaili, S. M. And Al-Ghafri, I. N. 1994.
Spawning cycle, recruitment patterns and maturity length of Indian oils sardine at
Al-Azaiba, the gulf of Oman. The Third Asian Fisheries Forum. Asian Fisheries
Society, Manila, Philipines, pp 484-487.
Sinclair, M. 1988. Marine populations: an essay on population regulation and
speciation. University of Washington Press. 252 pp.
Smith, P.E. 1995. A warm decade in the Southern California Bight. Calif. Coop.
Oceanic Fish. Invest. Rep. 36:120-126.
Smith, P.E ., and Moser, H. G. 1988. CaICOFI time series: An overview of fishes.
REP. CCOFI., vol. 29, p. 66-78
Souter A., and Issacs J. D., 1969. History of fish populations inferred from fish
scales in anaerobic sediments off california. Calif. Coop. Oceanic Fish. Invest. Rep.
13 pp. 36-70.
Souter, A., and Issacs, J. D. 1974. Abundance of pelagic fish during the 19th and
centuries as recorded in anaerobic sediment off California. Fishery Bulletin,
72, 257-273
20th
Stander, G. H. 1963. The sardine of South West Africa (sardinops ocellata).
Temperature: its annual cycles and relation to wind and spawning. Investi Rep.
Mar. Res. Lab. S.W.Afr. p 9:57.
Stander, G. H. 1964. The Benguela Current off South West Africa. Invest Rep.mar.
Res. Lab. S.W. Afr. 12:43pp+Plates 5-8 1.
Subrahanyan, R., 1959. Studies on the phytoplankton of the west coast of India.
Part 1, "Quantitative and qualitative fluctuation of the total phytoplankton crop,
zooplankton crop and their inter-relation, with remarks on the magnitude of the
standing crop and production of matte and their relationship to fish landings." Proc.
Indian Acad. Sci., 50B:113-187
Suvatti, C. 1981. Fishes of Thailand. Royal Institute of Thailand, Bangkok. 379 p.
Taunton-Clark, J. and Shannon, L. V. 1988 -. Annual and interannual variability in
the South-east Atlantic during the 20th century. S. Afr. J. Mar. Sci. 6: 97-106.
Tomczak, M. and Godfrey, J. S. 1994. Regional Oceanography: An Introduction.
Oxford, England ; New York: Pergamon, 422 p.
157
Tomosada, A. 1988. Long term variation of sardine catch and temperature.
Reproductive biology of Japanese sardine and anchovy in relation to their
population dynamics. Bull. Jpn. Soc. Fish. Oceanogr. 51: 51-54
Tyler, A. V., and Gallucci, V. F. 1980. Dynamics of fishes stocks. In Fisheries
management, R. T. Lackey and L. A. Nielsen, eds. John Wiley and Sons, Ltd., p.
111-147.
Ueber, E. and MacCall, A. 1990. "The Rise and Fall of the California Sardine
Empire." In Michael H. Glantz ed. Climate Variability, Climate Change and
Fisheries. Cambridge, Cambridge U.P.,: p. 3 1-47.
Van der lingen, C.D. 1999. The feeding ecology of, and carbon nitrogen budgets
for, sardine Sardinops sagax in the Southern Benguela upwelling ecosystem. Ph.D.
thesis, University of Cape Town, Cape Twon, South Afria. (vii)+ 199 pp.
Wada, T., and Kashiwai, M. 1991. Changes in growth and feeding ground of
Japanese sardine with fluctuation in stock abundance. In Long-term variability in
the pelagic fishh populations and their environment. T. Kawasaki, S. Tanaka, Y.
.
Toba and A. Taniguchi (eds). Oxford: Pergamon Press, p. 181-190.
Wada, T; Matsubara, T; Matsumiya, Y; Koizumi, N. 1992. Influence of
environment on stock fluctuations of Japanese sardine, Sardinops melanostictus.
Climate change and northern fish populations.,1995, pp. 387-394, Can. Spec. Pubi.
Fish.. 121
Walsh J. J., Whitledge, T. E., Esaias, W. E., Smith, R. L., Huntsman, S. A.,
Santander, H., and DeMendiola, B. R. 1980. The spawing habitat of the Peruvian
anchovy, Engraulis ringens. Deep-sea Research. 27, 1-28
Walsh J. J., Wirick, C.D., Dieterle, D.A., and Tingle A. G. 198l. Environmental
Constraints on Larval Fish Survival in the Bering Sea. The early life history of fish:
Recent studies, pp 24-27, Rapp. P.-V. Reun.Ciem, vol. 178
Ware, D.M, and Tsukayama, I. 1981. A possible recruitment model for the
Peruvian anchovy. Cooperative investigation on the anchovy and its ecosystem
between Peru and Canada., pp. 55-61, Bol. Inst. Mar. Peru.
Watanabe, Y., Butler J. L., and Mori, T. 1988. Growth of Pacific saury, Cololabis
saira, in the northeastern and northwestern Pacific Ocean. Fish. Bull., vol. 86, no.
3, p. 489-498
158
Watanabe, Y., Zenitani, H., Kimura, R.1995. Population decline of the Japanese
sardine Sardinops melanostictus owing to recruitment failures Can. J. Fish. Aquat.
Sci.!J. Can. Sci.., vol. 52, no. 8, p. 1609-1616
Watanabe, Y., Zenitani, H., Kimura, R. 1996. Causes of population decline in the
Japanese sardine (Sardinops
melanostictus): Overfishing or early mortality?
Proceedings of an international workshop: Survival strategies in early life stages of
marine., A.A. Balkema Alkema publishers, Brookfield, VT (USA), p. 83-94
Watanabe, Y., Zenitani, H., Kimura, R. 1996. Offshore expansion of spawning of
the Japanese sardine, Sardinops melanostictus, and its implication for egg and
larval survival Can. J. Fish. Aquat. Sci., vol. 53,no. 1, p. 55-61
Watanabe, Y and Wada, T. 1997. Expansion and contraction of migration range of
the Japanese sardine in the northwestern Pacific. Memoirs of the Faculty of
Fisheries, Hokkaido University, vol. 44, no. 1, p. 49-52
Watanabe, Y and Kuroki, T. 1997. Asymptotic growth trajectories of larval sardine
(Sardinops melanostictus) in the coastal waters off western Japan. Marine Biology
[Mar. Biol.], vol. 127, no. 3, p. 369-378
Watanabe, Y., Oozeki, Y. and kitagawa, D. 1997. Larval parameters determining
preschooling juvenile production of Pacific saury (Cololabis saira) in the
northwestern Pacific. Can. J. Fish. Aquat. Sci. vol. 54, no. 5, p. 1067-1076
Watanabe, Y, and Saito, H. 1998. Feeding and growth of early juvenile Japanese
sardines in the Pacific waters off central Japan. J. Fish Biol., vol. 52, no. 3, p. 519533
Watanabe, Y., Zenitan, H., Kimura. R., sato, C., Okumura, Y., Sugisaki, H., and
Oozeki, Y. 1998. Naupliar copepod concentrations in the spawning grounds of
Japanese sardine, Sardinops melanostictus, along the Kuroshio Current. Fish.
Oceanogr., vol. 7, no. 2, p. 101-109
Whitehead, P.J.P.
1985. FAQ species catalogue. Vol. 7. Clupeoid fishes of the
world (suborder Clupeoidei). An annotated and illustrated catalogue of the
herrings, sardines, pilchards, sprats, shads, anchovies and wolf-herrings. Part 1.
Chirocentridae, Clupeidae and Pristigasteridae. FAO fisheries synopsis. Rome
[FAQ Fish. Synop.], vol. 7, pt.1, no. 125, FAO/LTNDP, Rome (Italy), 303 pp
Wolf, P. 1992. Recovery of the pacific sardine and the California sardine fishery.
CALCOFI REP ., Vol. 33,
159
Wroblewski , J. S., and Rielmian, J. G. 1987. The non-linear response of plankton
to wind mixing events- implications for survival of larval northern anchovy. J.
Plankt. Res. 9(1):103-123
Yasuda, I., Sugisaki, FL, Watanabe, Y., Minobe, S. S., and Oozeki, Y.
1999.
Interdecadal variations in Japanese sardine and ocean/climate. Fish. Oceanogr, vol.
8,no. "pp. 18-24
Zuta, S., Tsukayama, I., and Villanueva, R. 1983. El ambiente marino y las
fluctaciones de las principles problacies pelagicas de la costa peruana. In
Proceedings of the Expert Consultation to examine changes in abundance and
species composition of neritic fish resources, FAO Fish. Rep., 291 (2): 17-36
160
APPENDICES
161
Appendix A. Summary of the Sardines and Anchovies Cases Studied.
Category
California
Japan
Peru
South Africa
Species
California sardine
(Sardinops caerulea)
Pacific sardine
(Sardinops sagax)
Japanese sardine
(Sardinops melanostictus)
Peruvian sardine
(Sardinops sagax)
South African PiJ chard
(Sardinops ocellatus)
warm
warm
warm
cool
spawning season January - June
February to May
September-March
August March
spawning area
south of Kyushu
Peruvian coast
Spawning
temperature___________________________
Punta Eugenia
Southern California Bight
San Francisco Bay
-
Cape Aguihas area
north of 22 S
around Walvis Bay 22 to 24 °S)
north of Walvis Bay
(
Off Chili only
Distribution at
low abundance
Punta Eugenia
South and east of Kyushu
Southern California Bight small coastal areas western
San Francisco Bay
and southern Japan
north 27 OS
Distribution at
high abundance
Pacific coast
British Colombia
Coastal range of south
Sakalina, Japanese sea coast
of the Korea Peninsula
Surrounding seas of the
Japanese island
Along the coast of
Peru and
Chili as far as 40 °S
north of Cape Town and south
of Walvis Bay
along the Namibian coast
Major currents
California Current
Kuroshio Current
Oyashio Current
Humboldt Current
Peru Coastal Current
Peru Undercurrent
Peru Countercurrent
Benguela Current
Appendix A continued
Category
California
Japan
Peru
South Africa
Species
Northern anchovy
(Engraulis
Japanese Anchovy
(Engraulis japonicus)
Peruvian anchoveta
(Engraulis ringens)
South African Anchovy
(Engraulis Capensis)
Spawning
cool
cool
cool
warm
temperature_______________________
Spawning
season
Spawning
area
Distribution at
low abundance
December- April
March-December
September and January in October and February
Southern California
Bight
Southern California
Bight
Scattered on the Pacific Ocean
and Japan Sea side
Ise and Mikawa Bays, Segami
Bay off Bojo Peninsula
Peruvian coast
Distribution at
high abundance
entire Pacific coast
The entire Peruvian
coast
Major currents
California Current
East China, Western Japan,
Seto Inland Sea, Coastal area
of central to southern Japan on
the Pacific coast
Kuroshio Current
Tsushima current
In the South cooler
and in deeper cooler
between the Cunene River
and Mercury Island
Cape Agulhas area
areas on'y
Humboldt Current
Peru Coastal Current
Peru Undercurrent
Peru Countercurrent
north of Cape Town and
south of Walvis Bay
along the Namibian coast
Benguela Current
IiJ
Appendix B. Terminology
Community matrix
In any given model system interactions between the different variables in
the system can be represented by coefficients in a square matrix, called the
community matrix, which has one row and column for each variable.
Complementary subsystem
All the n-k variables that are not included in the open path, denoted as n-k,
where k is the number of variables involved in the open path and n is the
total number of variable in the model.
Complementary feedback
The feedback of the complementary subsystem, denoted as Fflk0m, where
F is feedback.
Conditionally stable system
A system that is achieves stability by satisfying one condition or more.
Conjunct loops
Loops that have at least one variable in common.
Disjunct loops (m)
Loops that have no variables in common.
Feedback (F)
The overall sign of the interaction between variables involved in a path.
165
Negative feedback
Occurs when an increase/decrease in one variable (the initiator) sets in
motion processes that lead to decrease/increase in that same variable.
Overall Feedback
(Fe)
The feedback of the whole the system.
Open path
The series of links that connect variables i to j without crossing any variable
twice.
Path
A path is the interaction between two or more variables. A path is denoted
by (P) and a path Pij is read as path i from j.
Path length
Determined by the number of different variables in the path.
Positive feedback
Occurs when an increase/decrease in the initiator variable causes other
variables to change in such a way as to produce further increase/decrease in
the initiator.
Stable system
A system that, when perturbed from a stable equilibrium, will return to that
equilibrium. If the system fail to return to the equilibrium, the system is
described as an unstable system.
166
Table of predictions
A table constructed to show how each variable will change (increase,
decrease, or no change) as a result of changes in itself or any other variables
in the system.
Appendix C: Different calculations involved in analysis of a loop system.
a. The system:
e
FR
b. The community matrix
a5
0
aSFR
0
a4
aAFR
aFJ,
aFJ,4
aF1)R
c. Defining loops and their length
Loops of length one
Loops of length two
(-ass)
(-aSFRaFRS)
(-a)
(-aFpa,sJR)
(-aFRFR)
167
d. Conjunct and disjunct loops in the system
Conjunct
Disjunct
(-ass) and (-asFRaFRs)
(-ass) and (-ayR) and (-a)
(-a) and (-aFaR)
(-ass) and (-aFjaR)
(-aFRYR) and (-asFRaFRs)
(-a) and (-ayRaF)
(-aFRYR) and (-aFayR)
(-aSFRaFRS) and (-aFRsasFR)
e. Feedback
Fl
(ass) + (a) + (aFR)
2
F2 =
in1
addition of all loops of length 2 and subtraction
products of disjunct loops length 1(m).
F2 = (aAFR
3
F3 =
FP4) + (F .a) a55 aM
of
all combinations
a aFJR
a,4 aFJ
of pairwise
168
= addition of all loops of length 3 and substation of all combinations of pairwise
products of length 1 with ioops of length 2 and addition of all combinations of
triplet products of disjunct loops of length 3.
F3 =
None
(ass )(aF
a4J7),)
(a )(aSFR a)J,) + (a )(a )(aFR)
(-1)(-1) (-1)(--1) + (-1)(-1)(l) = 1 1-1 = 3
f. Defining the complementary feedback for the different paths.
Path
k
P11
1
Path length
Complementary
(k-i)
0
feedback (Ffl.k)
F2
3
2
(aFg)(-aF)
FO
P13
2
1
asFa
P21
3
2
FO
1
0
(-1)
F2
2
1
2
1
2
1
Fl
1
0
(-ass)
F2
(1)
P12
(-aF)(asFR)
(-1)
(aR)(-aFRs)
P22
(1)
Fl
(-ass)
(aAYR)
P31
(-apRs)
P32
(-aFR)
AFRFR
Fl
(-a)
(-ass)(-a)
(1)
g. Predicting change:
Lc
(-a)
(-asFR)(aFpA)
P23
v (At;
Fl
YFcomP
A n-k
169
Change in the equilibrium abundance of S (1) due to an increase in its own growth
rate (c) is
SS
2
-
+ (l)(aR ap.)
+ (1)
Ac
=+
Similarly change in the equilibrium abundance of A (1) due to an increase in the S
growth rate (c) is:
Afp3F
AS
Ac
F',
Finally, change in the equilibrium abundance of FR (3) due to an increase in the S
growth rate (c) is:
4112
APR
FIS'1
+(aF)(aM) +--
Ac
The same procedure is followed for the rest of the variables. The following table
shows the results of the prediction in the different variables.
Variables
Sardine
Anchovy
Sardine
Anchovy
Food Resource
Food Resource
'I,
1.
+
-
-
+
-
-
+
+
+
170
Appendix ft Tables of predictions and their polynomial coefficients for
different Basic Sets.
D.3.1. Basic Set 1 (a, b, and c) tables of predictions and their polynomial
coefficients (Feedback at different levels). The interaction between sardines and
anchovies is represented in the first colunm. Interactions between other variables
are fixed as indicated in the diagram at the top of each column. Zero overall
feedback indicates that the model fails to respond to any changes and therefor no
table of predictions can be obtained. The models that are shaded out are equivalent
to the ones directly above because of symmetry. Hence, due to symmetry in the
relationship between sardines and anchovies, under subsets (a) and (c), the
interactions for amensalism, commensalism and predator-prey are only analyzed
once. Note that the columns and rows of the matrix tables represent Fishing,
Sardine biomass, Anchovy biomass, and Food Recourse, respectively.
171
D.3.1 continued
a
b
Table of
prediction
interaction
2
+ + + +
- +
0
no-interaction
-
+
0
+
0
0
+
+ +
amensalim
+
o
0
+ +
0
+
0
+
0
0
+
0
0
+
-5
+
0
- 2
0
+
-1
-i
2
0
+
Polynomial
coefficient
+
7
+
0
Table of
prediction
Polynomial
coefficient
-5
-7
-2
-1
-1
0
-2
-5
-2
-1
commensalism
+ + + +
- +
- - +
0
+
0
0
-s
9
+
0
+ +
-2
6
0
0
- +
- - +
+
+
0
0
+
-1
+
0
± +
-2
+
0
0
-6
-2
+
0
0
+
-1
172
D.3.1 continued
/\
a
Interaction
Table of
prediction
ro
cO/O
-+
Table of
prediction
Polynomial
coefficient
++
+
predator-prey
b
I
-8
+
I
0
0
+
0
0
+
I
-1
_8J
0
L±
r+ +
p
0
1+ +
ro1
21
competition
I
1-61
I3
I
L-']
I
-
L+
0
+
ol
0
+
+
0
0
+0
+
0
0
+J
31
Lij
0+
+
I
I
0
+±
101
-61
I
01
0
r++++i r41
01
+1
0
I
+
0
L-ij
I
--+
I+
I
1-61
1-21
I
0
0 0 +
_________ __________
mutualism
rii
51
-+
[-21
0
Polynomial
coefficient
0
+
0
0
+
[-11
1-51
1-61
1-21
I
[iJ
[ii
1
I
I
I
1-41
L-11
-7
-4
-2
173
D.3.1 continued
Table of
prediction
Interaction
Polynomial
coefficient
0
2
4
no-interaction
1
amensalim
commensalism
4-
+0
1
1
0
4
0
+
0
-+
0
0 +
1
1
+
0
+ +
1
4
0
+
0
+
0
0
+
1
1
174
D.3.1 continued
C
Table of
prediction
Interaction
Polynomial
coefficient
0
3
predator-prey
5
p
+ + + +
+
competition
- +
0
0
+00+
mutualism
4
+o
+0
+ + +
+
0
0
2
+
1
-1
-2
5
-3
-1
175
Appendices D.3.2, P.3.3, and P.3.4. Tables of predictions and their polynomial
coefficients (Feedback at different levels) for the models in Basic Set 2, Subset
A, B, and C groups (a, b, and c) as indicated. The interaction between sardines
and anchovies is represented in the first column. Interactions between other
variables are fixed as indicated in the diagram at the top of each column. Zero
overall feedback indicates that the model fails to respond to any changes and
therefor no table of predictions can be obtained. The models which are shaded are
equivalent to the ones directly above because of symmetry. Under Subsets a and c,
the interactions for amensalism, commensalism and predator-prey are only
analyzed once. Due to symmetry in the relationship between sardines and
anchovies. Note that the columns and rows of the matrix tables represent Fishing,
Sardine biomass, Anchovy biomass, Sardine Food Recourse, and Anchovy Food
Resource, respectively.
N
''
Ew
c000Nm_ N\Oootm
11111
111111
111111
Ø4I
______________I'
Ii
I
++++ ++++ ++++ +++
I
I
0+I++ +++++ ++l
+l++I +l++l +I++I +1++I +I+
++I l+++I 1+0+11+ ++l 1+ ++I
+1 I++ +010+ +1 l++ +1 I++ +1
I
Ew
I
,.-
- -
-
Ilil
'
+I+++
I
\
I
I
++l++++l++
cnt
+l++l +I++l
l+++i 1+
+1
+++1 ++
++l 1+
+1 I++
________________
I
I
II
E
.2
C
E
C.)
C
CO
I
I
I
(a)
I
U)
I
'
I
C
I
c
I
I
0
A
177
D.3.2 continued
a
0
b
t
Table of
prediction
Interaction
± + + + +1
± -I
predator-prey
±
±
t
± ± ±1
±
Table of
prediction
Polynomial
coefficient
r-41
-13
H-
- 16
I-
0
I-ill
4j
+ ± ± ±1 r-21
71
I -10
±±
±I
1-91
± ± ± ±1
[1
-1--- *
±±+
±
I- ± ±
1±
L± +
+
±
I
I
L-ii
±1
r-61
-II1-15
1-14
I
±11
I
-9
±
+II-4l
± +] Lu
I
±
I
I -8
± ± ±1 I_31
+
0
±
+
+
]
I
0
±
[-ii
[-ii
+1
-I
-31
-
±
+0±10]
-61
1
0
L3
[_]
+
±
± ± + ± ± [51
Iu
-±-±
±+
I
I
± -, I-iiJ
±
±
I
S
±-±
mutualism
+
0
-11
Hi
[-i]
I
L±±-++]
I
I
I
± ± +1
1+
L
I
I
I
I__8I
0
1
r-1
-
1
±
±
I
I
0
I- ± - ± -1
H±
competition
+
Polynomial
coefficient
I
±±][iJ
4
t
±
+±
I
I
±
±±±
±±
1
1-31
L-i]
178
D.3.2 continued
0
t
Polynomial
coefficient
Table of
prediction
Interaction
± ± ± ±
± - +
no-interaction
- ± ± ±
± ±
±
0
±
-5
±
-2
0_
-1
±
±
+ ± ± ±
±
+ - ±
± + ±
4
±
+
-1
± - ±
- ± - ±
±
commensalism
-2
-1
0
0+0
-
-5
-6
0
±
amensalim
±
+
±
-2
±
±
5
-3
-7
-2
_ij
179
D.3.2 continued
t
±
predator-prey
4
Polynomial
coefficient
Table of
prediction
I riteraction
0
±
±
-2
-6
+
-6
-2
± ± ±
±
o
±
±
±
0
0
± +
±
8
-1
0
0
competition
- 2
-4
-2
-1
o
mutualism
± ± ± ±
-4
± ±
-6
-4
-2
8
±
± ±
+
± ± ±
± ±
-1
a)
5.
3
F
g
g
++l 1+
+010+ ++l 1+
"0Il++ CI
C++I+ o++l+Lè\Y
I
I
0lI++
I
+++ 0+1++g
+ +
0+
+
+
CI I++0l
C++ + 0+
+
/
+000+ +000+
+000+
____________
J
+ 0
0+
+
I
+
+
\-_
I_
+
0
0+
1+0011+101 I+CoII+o-..1
+
+
0+ +
+
++
I
+
++
+
o+l++ O+1+Oool+I 001+0 C+I+0
+0+0+ +000+ +000+ +000+ +000+
H
go
/
:f
181
D.3.3 continued
tt
b
Table of
prediction
Interaction
±0
+
±
predator-prey
4
+0
0
0
-13
-18
-12
0
0+
i
0
Polynomial
coefficient
-±-±
-3
+
±
±
Table of
prediction
Polynomial
coefficient
+
0
±-±±
0
-+
0
o
[-21
1-81
-12 I
1-91
1-31
[ij
o
± ± + ± +1 r-21
+±-+
.
--±-
--±±
-±
[±
0
Jo
±
I-
competition
0
+
oJ
0
-
0
0
o
I
oj
±]
1±
±±
[+0 0
0
01
0(
+j
l
±j
l_o
I
I-3
[_ij
[01
1-71
J_
-10
-8
1-91
I -12
l°i
-4 I
L'J
± ± + ±1 r-41
±
I
±
01
-10
[_i
L o
r ± ± ± +1 -17
I- ±
± 0
-18
H
0
0 1 [-1
J_5
I
I
mutualism
+
0
l
oI
I
±
1±
±+
01
01
[±o 0 o+j
-12
1-121
1-31
L'J
00
0
E ci)
o
I
I
++
I
I
I
I
-
111111
+
I
I
I
0
+
0
I
I
I______________
+1+01
______________
r-i
++I
+
E
-
a'
E
E
0
0
I
I
I
I
+
+
+
4
I
+
+
I________
0+10+ ++t0+ ++10+
+00 01+00 +1+0+ +1+0+
+00+0
l+l +1+01
111111
=E
ci,
El
I
+01 ++
+0+ I'0
II
+
+
-ri
2
o
\+l+0l
.2
(I)
01
183
D.3.3 continued
0
Table of
prediction
Polynomial
coefficient
± + 0 ± 0
o ± - 0
-2
Interaction
predator-prey
4
±
±
0
±
0
±
0
± -
0
±
0
+ + 0 ± 0
0 ± - 0
- - ± 0 ±
S
+
0
±
+ -
0
0
0
0
-13
-9
-1
-2
-13
_
-1
0
-1
-
competition
-7
-3
-1
mutualism
± ± ± + ±
-4
± - - ±
0
- 13
0
±
0
+
0
± ± -
0
+
+
0
-- 13
-1
184
D.3.3 continued
C
C
N®
Table of
prediction
Interaction
±
0
±
0
±-±
--±0
-±
0
o
no-interaction
Polynomial
coefficient
o
±
0
0
0
0
0
±
-7
-6
-2
0
-1
amensalim
-5
-6
-2
-1
commensaflsm
±
o
4
±
±
0
±
-2
0
±
0
±±
0
0
±
0
-2
-'
185
D.3.3 continued
00
0
Interaction
±
predator-prey
0
0
±
±-± 00
-7
-2
+
-1
--±0
0±± 0
1_
0
± ±
± ±
+
mutualism
-
±
0
-±
competition
Polynomial
coefficient
-1
Table of
prediction
0
0
0
0
± ±
-±
0
-2
0
+
-1
0
+ ± ± ±
- -± ± -±
±
+
-±±
0
- 1
0
0
-3
0
0
0
±
1:
-2
-1
'0
/
0Nm
0I
E
i
1
I
<
+4LL.
(JJ
I
I
I
I
III
I
II
I
I
I
Q-o
I
I
I
i'1i111
I
+occ+
-I
I
II
+0o+ +c
Ii
+1+00 +1
too
+0I++
_I_______
0+
00
+oI++ +0
I
Ew
o
C
oox
(
E\SJ/
ItijItI
1
1
Q-o
I
+000+ +000+
+00+0 +00+0
I
I
I
+00+0
+1+00
00 ++00
+1 I++ +0t++
0++
C
0
I
+1 l++
C
E
E
col
ci)
01
C
cn
C
0
El1
1
C.)
/
\
x
o/X
\
E nj
i
I
I
1
I
I
I
I
_(liI
+oc+ +oo+
+o+
+oc+0
II
_______
I
0
WI
El
I
I
I
I
I
II
I
I
+c+
++o
+1
E
+
I
1
++ 100
l++
+
+000+
O+OOC+?CC
++Ic0 +I+cc ++tc
+Ic++ ++I+++i+++i ++
____
+
+coo+
ONCm_
000
Qfl+flQ)
E a
+
+000+
+00+0
05NNOfl
I
001
00
C
+
o
(q)
00
I-0+
+I0++
-
c
.2
(U
0
I
188
D.3.4 continued
0
0'
Polynomial
coefficient
Table of
prediction
Interaction
0
0
-2
-8
no-interaction
-2
-1
amensalim
±
± ± ±
1
2
- - +
0
0
o
+
0
0
±
±
0
0
±
0
-2
0
0
0
+
-i
+
+ + +
± - 0 0
commensalism
4
0
0
+
0
0
0
0
±
0
0
0
0
+
0
±
+
-1
- 6
-11
-2
189
D.3.4 continued
00
tt
Polynomial
coefficient
Table of
prediction
Interaction
0
-3
predator-prey
10
-8
4
-2
-1
± ± ± ± ±
competition
---±00
-±- 00
±
+
mutualism
0
0
0
0
2
0
±
0
-6
-2
0
+
-1
+ ± + + +
-2
± - - +
0
0
0
0
-12
+
0
0
±
0
.±
0
0
0
+
-6
-2
190
Appendix E: Omani Fishers Responses to the Interview Questions by Region.
1 .The ages of the fishermen and how long they have been in the sardine/anchqyy
fishery.
Al-Batina
More than 50 years old, more than 20 years in sardine and anchovy fishing.
Muscat
More than 50 years old, more than 20 years in sardine and anchovy fishing.
Al-Sharqiya
More than 50 years old, more than 30 years in fishing.
Al-Wusta
32 years old and more than 15 years experience in the sardine and anchovies
fishing.
Dhofar
More than 50 years old, more than 20 years in sardine and anchovy fishing.
2. The sardine/anchovy seasons and whether the seasons are related to cold versus
hot months or to the monsoon.
Al-Batina
Sardines appear mostly in September to April, with peak abundance in December
and January. Anchovies, in contrast, have no particular season.
Muscat
Sardines are available all the year around. They disappear temporarily in summer
time. The peak for sardine abundance is during the colder months. Anchovies,
however, come and go without warning.
Al-Sharqiya
Fishers here hardly ever target for sardines or anchovies at any time. But sardines
are observed more during cold months. Anchovies have no particular season.
191
Al-Wusta
Sardines are always around. Anchovies, however come, and go and have no
particular season.
Dhofar
The sardine season is defined. Anchovies on the other hand have no definite
season. Sardines are vulnerable to the fishers' gear all year around except during
the monsoon (June-early September in some areas).
3. Whether the sardines/anchovies caught together in the same gear.
Al-Batina
Sardines and anchovies are rarely caught with the same gear.
Muscat
They can be caught together when both are in high abundance along the shore. This
rarely happens, however.
A1-Sharqiya
No.
Al-Wusta
No. The fishermen never target sardines and anchovies together. If anchovies are
spotted, different gears will be used than if sardines are spotted.
Dhofar
No. The fishermen never target sardines and anchovies together. If anchovies are
spotted, different gears will be used than if sardines are spotted.
4. The ype of gear used for the sardines and anchovies and the gear types
pfe4.
Al-Batina
Beach seine is the most common gear for both fish. Purse seine can be used if the
fishers can afford it. Beach seine is preferred by most sardine and anchovy fishers
because it is cheap and easier to operate. There are few fishers that target sardines
and anchovies using purse seines. Nets used for sardines have bigger mesh size
than the ones used for anchovies.
192
Muscat
Gilinet, beach seine, and purse seine can be used for both fishes. However the mesh
size used for anchovies is smaller than the one used for sardines. Beach seine is
mostly used for sardine and anchovies. Only a few or the well established, full-time
sardine and anchovy fishers use purse seines.
A1-Sharqiya
If there is any fishery for sardines or anchovies it is temporary. Beach seine with
smaller mesh size can be used for anchovies while with bigger mesh size for
sardines.
Al-Wusta
Beach seine, gilinet, and purse are the common gears for sardines and anchovies.
Which gear is used depends on the school size spotted. Purse seines are usually
used for bigger schools.
Dhofar
Gilinets, beach seines, purse seines, and cast nets. Which gear is used depends on
the school size spotted. Purse seines are usually used for bigger schools. Smaller
mesh size nets are used for anchovies, while bigger mesh size nets are used for
sardines.
5. The distance along the shore and offshore that the sardine/anchovy fishery
extends.
Al-Batina
4 to 6 meters deep offshore is where the gear can be operated. Along the shore the
fishery is limited within the village fishing area
Muscat
3 to
5
meters deep offshore is where the gear can be operated. Along the shore the
fishery is limited within the village fishing area
Al-Sharqiya
There are no specialist sardine or anchovy fishers in this region. But most of the
catch is obtained from near-shore in waters 4-5 meters deep.
193
Al-Wusta
4 to 6 meters deep offshore around the island and toward the offshore waters.
Dhofar
3 to 5 meters deep, only offshore. Along the shore the fishermen can only go up to
the boundary with the next fishing village.
6. The use and price of sardines and anchovies.
Al-Batina
Both types of fish are dried and used as fertilizer, food for cattle, and for human
consumption when sardines or anchovies are fresh. Anchovies are worth twice as
much as sardines.
Muscat
Both types of fish are dried and used as fertilizer, food for cattle, and for human
consumption.
Due to urbanization, most fishermen will sell the catch fresh to fish processing
companies nearby. Anchovies are worth more than sardines.
Al-Sharqiya
All the catch is used as bait only. Fishers are not allowed to use the catch for any
other purpose. Fishers might complain to the mayor of the fishing village if
anybody used the fish for other purposes.
Al-Wusta
Because Masira Island is very far from the primary markets, drying is the only way
to sell the catch. Only during the cold season can the fish be transported fresh to the
Muscat and Dubai Markets. The fish can keep its freshness if it is well maintained.
The price per truckload of sardine or anchovy has to be high for the fishers to enter
the fishery.
Dhofar
Most of the catch is used locally as food for cattle and as fertilizer after it is dried.
Anchovies are less expensive than sardines.
7. Whether sardines or anchovies are preferred by the fishers and why.
194
Al-Batina
Anchovies are preferred due to their high price and demand. They are more favored
over the sardines because they are more often used for human consumption.
Because the sardines are more abundant fishers will target sardines.
Muscat
Some fishermen are indifferent between the two and will fish either sardines or
anchovies. Other fishermen will target the most expensive one according to the
market. The well established, full-time sardine and anchovy fishers target both
fishes. In general anchovies are worth twice as much as sardines.
Al-Sharqiya
Sardines ar preferred because they are more available and make good fish bait.
Al-Wusta
The fishers on the island prefer sardines to anchovies. Sardines are believed to be
excellent bait for different fishes. Anchovies are not targeted except if demanded
by the Muscat or Al-Batina markets. Because Masira Island is very far from the
Muscat markets, the Masira fishers require a buyer before they target the fish and
the price also should be high. Sardines and anchovies could be sold by the fishers
for up to 130-150 Omani Rials per ton (wet).
Dhofar
Fishers prefer sardines over anchovies because there is greater demand for sardines.
8. Periods and duration when sardines/anchovies disappear and whether the two
fish disappear together or alternately.
Al-Batina
Anchovies used to be more abundant than the sardines a long time ago. Sardines at
that time did not have a particular season. Things changed and sardines became
more abundant. The two fish alternately disappear. There are short periods when
sardines disappear; for a maximum of two months during summer. Anchovies
cannot be understood. They come and go without warning.
195
Muscat
Anchovies are rarely found in this region and they have no particular season. They
can disappear for up to 3-5 years before they return in higher abundance.
Anchovies stayed for two month before they disappeared again. Anchovies in the
50s were more abundant than sardines and at that time the sardines had no defined
season. Since then sardines have never disappeared from the local waters except for
short periods in the summer (June-July). The fishermen believe that sardines and
anchovies do not exist together in high abundance.
AI-Sharqiya
Sardines never disappear from the area and fishers avoid catching sardines to keep
them available. The fishers believe that sardines are an important food resource for
the tunas that are available in their waters. Anchovies come and go from the region.
The fishers did not know whether the two alternate because we do not target either
fish.
Al-Wusta
The sardines never disappear from around the island. Anchovies, however, are rare
but they are rarely targeted and the fishers were not interested in them. The fishers
remember that anchovies used to be more abundant than sardines a long time ago,
sometime during the 70s. They believe that anchovies and sardines are never found
together.
Dhofar
Anchovies used to be more abundant than sardines a long time ago. Anchovies then
disappeared and sardines became dominant. Sardine and anchovies do not show up
at the same time.
9. Whether the sardine/anchovyjlisappearance is cop1etely or whether they can be
found in identified areas.
196
Al-Batina
The fishers in this region believe anchovies goes to deeper waters where they can
not be caught by the traditional gears. Sardines are always available along the
shore, but the fishermen do not always target them.
Muscat
Anchovies never disappear completely from the Omani waters. They always can be
found in Al-Batina region. Fishermen do not know where the sardine can be found
when they disappear for long time.
Al-Sharqiya
Fishers in this region know that sardines and anchovies are targeted in the Muscat
and Al-Batina regions all the year around. In Sharqiya, however, sardines are never
targeted directly, but the fishers never fail to get what they need as bait.
Al-Wusta
Sardines never disappear from this region. Anchovies are believed to always be
abundant along the Al-Batina coast.
Dhofar
Sardines and anchovies never disappear completely from the Dhofar region. They
are present all year and are believed to be in deep waters where they cannot be
reached by the traditional fishing gear. The fishers in this region believe that
anchovies are always available in the Al-Batina region.
10. How the fishers respond to increases and decreases in sardines and anchovies in
terms of purchasing gear. Also, the fishers' practices of gear sharing.
Al-Batina
Some fishermen buy more fishing gear and increase their effort when the fish
become more abundant. The majority, however, do not. Regardless of whether the
fishers buy more gear, they have to stay together to reduce the operating coast.
Extra nets, when not used, are covered on the beach until the next fishery.
197
Muscat
The fishermen in this region are satisfied with what they make from these two
fisheries, so they do not increase their effort by buying more fishing gear when the
fish are more abundant. The gears are always kept on beach when not being used.
A1-Sharqiya
No measures are taken to develop the fishery when the sardines or anchovies are
more abundant. The goal is to preserve the sardines or anchovies as food sources
for the large pelagic fish species.
Al-Wusta
Masira is a small island. The fishers there work together and do not buy more gear
for sardines or anchovies when these fish are more abundant.
Dhofar
The sardine and anchovy fishery in the Dhofar region is localized. Increasing the
frequency of fishing is enough to satisfy the local market demand.
11. How the fishers respond to increases and decreases in the sardines and
anchovies: whether by increasing their fishing frequency or by hiringe new
employees to increase their fishing activities. In addition, what happens to the extra
gear and employees (if any) when the stocks decrease.
Al-Batina
The majority of the fishermen increase the frequency of fishing by hiring more
fishers.
When sardines or anchovies disappear the hired extra employees are laid off. Most
of these employees are part timers and have other part-time employment are not
affected by the laying off.
Muscat
If they have to, the fishermen prefer to hire more fishermen and help than buying
new gears. Because it is easier to get rid off the extra fishermen whenever it is
necessary.
198
A1-Sharqiya
No change.
Al-Wusta
Fishers increase the frequency of fishing with the available number of fishermen.
They might stop targeting other fish stock and concentrate on sardines or anchovies
depending on the demand.
Dhofar
Fishers increase frequency of fishing with the available number of fishermen. They
might stop targeting other fish stock and concentrate on sardines or anchovies
depending on the demand.
12. When the sardines/anchovies retuin after a long disappearance, which of them
tends to come back in higher numbers.
Muscat
Anchovies, but for short period of time
A1-.Sharqiya
Do not know the answer
A1-.Wusta
Anchovies, but for a short time.
Dhofar
Anchovies, but for a short time.
13. The sardines/anchovies expansion whether it is along the shore or offshore.
Al-Batina
Along the shore and offshore.
Muscat
Along the shore and offshore.
Al-Sharqiya
It should be along the shore because it is found in Muscat and Al-Batinah
199
Al-Wusta
around the island and offshore
Dhofar
along the shore and offshore.
14. The periods when the fishermen stop fishing sardines/anchovies even though
they are available: why and for how long.
Al-Batina
Anchovies are caught whenever they are abundant. Sardines are function of market
demands and the availability of other valuable fish.
Muscat
Sardines and anchovies prices are always function of the market for most
fishermen. Oniy few fishers stay in the sardine fishery even when the market is not
demanding. These fishers sell their catch to companies, which use the sardine as
bait.
Al-Sharqiya
No sardines fishery going on in this region at all. Sardines are caught whenever
more bait is needed only
Al-.Wusta
Most of the fishers stop fishing sardines and anchovies unless there is a demand for
it. They stop fishing as well in sunmier time (Jun to August) due to rough
environmental conditions.
Dhofar
During monsoon some villages are forced to stop fishing because the sea condition
is rough. June to early September.
15. The types of fish caught by the fishers other besides sardines/anchovies.
Al-Batina
Yes. Fishermen will go for any other fish that are more valuable and demanded by
the market. Including Kingfish, tuna, miscellaneous of ground fishes.
Muscat
Yes. Fishermen will go for any other fish that are more valuable and demanded by
the market. Including Kingfish, tuna, miscellaneous of ground fishes.
Al-Sharqiya
Yes. This region is famous for it large pelagic fisheries especially Tunas.
Al-Wusta
Fishers go for any other valuable fish if exist. Including Kingfish, tuna,
miscellaneous of ground fishes.
Dhofar
Fishers go for any other valuable fish if exist. Including Kingfish, tuna,
miscellaneous of ground fish, lobster, and abalone.
16. Whether the fishers continue fishing for sardines/anchovies when other fish are
available or fishers decrease their activities so they can catch other fish as well.
Al-Batina
Normally fishing activities depend on the market demand and supply. Because
other fish are more expensive than sardines, fishers reduce their activities on
sardines and target other fish. Anchovies on the other hand are always targeted
whenever they are available.
Muscat
All year around fishermen will split the time to day and night time fishing. Daytime
will be for sardines and anchovies, easy to spot. Nighttime fishing will be for other
fish species.
Al-Sharqiya
Sardines and anchovies never targeted as a fishery by itself in any circumstances.
Al-Wusta
Sardine and anchovy fishing is a function of market demand. The fishers will target
the sardines or anchovies when the Market demands for them. Otherwise fishers
target other fish.
201
Dhofar
Sardines and anchovies are targeted whenever they are available because there is
always a demand for it by the farmers.
17. Whether the fishermen noticed any changes in the oceanographic conditions
(water temperatures, currents, or appearance of rare species) that they think might
be related with changes in sardine and anchovy abundance.
Al-Batina
No particular change in the sea condition that they can remember. However, when
anchovies used to be abundant fishermen used to have more date trees. So they will
split their fishing time between the two income sources. When anchovies
disappeared the date trees dried. Fishermen realized that the salinity of the
underground water increased which basically kills the trees.
Muscat
Kingfish and Tunas are the main fish species that are observed with the anchovies
and sardines. The fishers do not remember anything about changes in oceanic
condition However, they do remember that when anchovies were available the sea
was more productive.
A1-Sharqiya
No particular change remembered except that the sea used to bring more tunas and
large pelagic long time ago than now.
Al-Wusta
Kingfish and Tunas are the main fish species that are observed with the anchovies
and sardines. The fishers do not remember anything about changes in oceanic
condition
However, they do remember that when anchovies were available the sea was more
productive.
202
Dhofar
No particular change in the sea condition that they can remember. However, when
anchovies used to be abundant fishermen used to have more fish to catch beside
anchovies.
Download