1
Running head: Invertebrate losses from beach disposal
2
3
Beach disposal of fine sediments leads to losses of invertebrate prey
4
5
Lisa M. Manning1,3, Charles H. Peterson1,*, Melanie J. Bishop2
1
6
University of North Carolina at Chapel Hill, Institute of Marine Sciences, Morehead City,
7
North Carolina 28557 USA
2
Department of Biological Sciences, Macquarie University, New South Wales 2109 Australia
22
3
Present address: National Oceanic and Atmospheric Administration, 1315 East-West Highway,
23
SSMC3, Silver Spring, MD 20910 USA.
24
*
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Email: cpeters@email.unc.edu
1
25
ABSTRACT: Despite increasing use of beach filling with dredged materials to protect
26
coastal property and public beach amenity from erosion, our understanding of how this
27
practice impacts sandy beach ecosystems is poor. We coupled field monitoring of two
28
successive beach disposal events with manipulative mesocosm experiments to assess the
29
mechanisms and extent of ecological impacts of fine sediment disposal. Frequent sampling at
30
replicate disposal and control sites on Topsail Island, North Carolina, revealed that following
31
each of two beach disposals of dredge spoils, each a year apart, beach granulometry was
32
transformed from medium to fine sands and turbidity plumes exceeding state water quality
33
limits by up to 12 times were created in the surf zone. Where disposal occurred before annual
34
invertebrate recruitment to the beach, it caused press responses to the mole crab Emerita
35
talpoida (negatively impacted) and the spionid polychaete Scolelepis squamata (positively
36
impacted). Where disposal followed recruitment, it acted as a pulse disturbance, suffocating
37
new recruits (Donax variabilis and haustoriid amphipods). Impacts of both press and pulse
38
disturbances lasted almost a year, overall depressing invertebrate prey abundances. In
39
mesocosm experiments, turbidity plumes of the magnitude experienced in the field slowed
40
growth of clams and modified habitat choice by predatory fin-fishes. Hence, while the
41
rapidly-eroding dredge spoil provided no lasting storm protection for beachfront
42
development, it negatively impacted both invertebrate prey resources and predator foraging
43
behavior. These negative ecological consequences to key foraging grounds without obvious
44
public benefit suggest that the practice of beach disposal of dredge spoils be disallowed.
45
46
KEY WORDS: beach nourishment • coastal erosion • fine sediment disposal • sandy beach •
47
sea-level rise • soft sediments • surf fish • turbidity.
48
49
2
50
51
INTRODUCTION
Global climate change is already having widespread impacts on ecological systems (e.g.,
52
Parmesan 2006), creating urgent need for adaptive management interventions to sustain their
53
functions and services (Staudinger et al. 2012). Sandy beaches are among the most threatened
54
ecosystems (Schlacher et al. 2007, Defeo et al. 2009, Dugan et al. 2010), not only as a
55
consequence of the direct effects of rising sea levels and enhanced frequency of intense storms
56
on sandy beaches constrained by coastal development (IPCC 2007), but also because of
57
interactions between these manifestations of climate change and management responses intended
58
to counteract these and other causes of erosion of oceanfront property (Dugan et al. 2008,
59
Schlacher et al. 2012). Sea walls and other hard structures have historically been installed on
60
beaches to protect beachfront development but have been of variable success, in some instances
61
exacerbating beach erosion (Pilkey & Wright 1988), and in many cases causing measurable
62
ecological impacts to beach ecosystems (e.g., Dugan et al. 2008, Jamarillo et al. 2012).
63
Consequently, beach nourishment, whereby sediments from elsewhere are added to beaches to
64
counteract erosion, has become the favored management response (Valverde et al. 1999). Beach
65
nourishment is now practiced routinely along ocean beaches around the world and on beaches of
66
large inland lakes (e.g., Basco 1999, Hanson et al. 2002, Cooke et al. 2012).
67
It is often assumed that because sandy beaches are physically dynamic, the organisms that
68
live in this environment are pre-adapted to handle the stresses associated with sediment
69
deposition. Early studies of impacts of beach nourishment on abundances of intertidal
70
invertebrates (Hayden & Dolan 1974, Gorzelany & Nelson 1987) showed only short-term
71
impacts lasting for just a few months. Flaws in the design of most studies monitoring ecological
72
impacts of nourishment have, however, severely limited the inferences that can be made from
73
many of these (Peterson & Bishop 2005, Speybroek et al. 2006). In a review of largely
74
unpublished reports, Nelson (1988) concluded that where beach nourishment projects use
3
75
sediments that are substantially finer than native beach sands, suppressions in invertebrate
76
abundances may persist through time with the magnitude of impact increasing with the
77
proportionate contribution of fine materials in the beach fill. Subsequent field studies where
78
beach fill was finer than native sediments (Rakocinski et al. 1996, Peterson et al. 2000) and
79
reviews (Hackney et al. 1996, Greene 2002) have re-enforced this view. It has also become clear
80
that substantially augmenting the coarse fraction of sediments and shell hash on the beach can
81
have similarly large impacts on the dominant infaunal invertebrates, especially burrowing bivalve
82
molluscs (McLachlan 1996, Peterson et al. 2006).
83
Despite increasing acknowledgement of ecological impacts of beach nourishment, our
84
mechanistic understanding of how these arise remains limited (Peterson and Bishop 2005,
85
Speybroek et al. 2006). Sandy beach invertebrates may be directly impacted by beach filling
86
through suffocation after burial under deposited sediments (e.g., Schlacher et al. 2012).
87
Alternatively, or additionally, impacts may arise through changes to aspects of the physical
88
environment, such as sediment grain size and turbidity (Peterson et al. 2006, Manning et al.
89
2013). Frequently, there is a large degree of mismatch between the sedimentology of native
90
beaches and fill sediments. Whether the fill sediments are finer or coarser than native beach
91
sediments may influence the duration and magnitude of biological impacts directly, by affecting
92
the ability of organisms to burrow and feed (Manning et al. 2013, Van Tomm et al. 2013), and
93
indirectly by influencing the lifetime of fill placement (Warwick 2013).
94
Our lack of mechanistic understanding of impacts of beach filling on sandy beaches is
95
troubling given the importance of beaches as nesting sites for threatened and endangered sea
96
turtles, and as foraging grounds for birds, fishes, and crabs (Hubbard & Duggan 2003,
97
McClenachan et al. 2006, Manning et al. 2013). Studies making clever use of human
98
interventions to sandy beaches, such as the harvest of clams (Defeo & de Alvara 1995,
99
Brazeiro & Defeo 1999) and the removal of wrack (Dugan et al. 2003), have shown that
4
100
despite the physically dynamic environment of sandy beaches, biological interactions also
101
play a role in organizing communities on ocean beaches. Hence, impacts of beach filling on
102
sandy beach invertebrates may reflect not only changes in the physical environment but also
103
changing interactions among species.
104
As the demand for interventional adaptation of coasts to climate change grows
105
(Staudinger et al. 2012), a better understanding of the mechanisms by which sandy beach
106
ecosystems respond to and recover from disturbance, including human interventions, is
107
imperative. Here, we couple field monitoring of two successive spoil disposal events with
108
manipulative mesocosm experiments to assess the mechanisms and extent of ecological
109
impacts to the sandy beach ecosystem of filling with fine sediments. The fill events, separated
110
by a year, utilized sediments from maintenance dredging of a navigation channel, justified by
111
the dual purpose of spoil disposal as well as nourishment. Frequent sampling of control and
112
disposal sites before, during, and after each of the fill events assessed: (1) the period over
113
which the fill augmented sediment volume on the beach; (2) the extent, magnitude and
114
duration of impacts to sediment grain size on the beach, and turbidity in the surf zone; and (3)
115
impacts on the abundances and body sizes of dominant beach invertebrates. Mesocosm
116
experiments assessed whether enhancement of turbidity, such as caused during and shortly
117
after spoil deposition, affected individual growth rate of the biomass dominant, the bean clam
118
Donax variabilis, or habitat choice by two common surf fishes. We hypothesized that in the
119
event that fine sediments are rapidly eroded from the beach, impacts to fauna would be short-
120
lived.
121
122
MATERIALS AND METHODS
123
Study sites and monitoring design
5
124
We assessed physical and biological impacts to the intertidal and shallow subtidal beach
125
of repeated spoil disposals on Topsail Island, North Carolina, USA (see Appendix 1, available in
126
MEPS supplementary material). Topsail Island is a low-lying, transgressive barrier island with
127
an average width of 280 m. The island is extensively developed despite regular over-wash by
128
storms (Cleary & Pilkey 1996). Protection of the island’s beaches and coastal development
129
involves regular redistribution of sand from the low to the high beach by bulldozing as well as
130
beach filling with sediments from outside sources. Spoil from maintenance dredging of
131
navigation channels in New River Inlet and connecting portions of the Intracoastal Waterway
132
serves as a major source of sediment for beach filling, which has occurred for years, even before
133
our study.
134
Our study considered two beach filling events. During the first, in late April – early June
135
1999, 94,996 m3 of dredge spoil was distributed across 350 m of beach at the northern end of the
136
island. During the second, in April 2000, a further 49,066 m3 was distributed across
137
approximately the same area. In October 1998, six months prior to the first filling event, we
138
established two sites within the area designated for filling by dredge spoils (Topsail Reef I [D1]
139
and II [D2]) and two control sites (Topsail Dunes [C1] and Roger’s Bay [C2]: Appendix 1) to the
140
south of the fill areas. It was not possible to intersperse disturbed and control sites because
141
dredge disposal covered a continuous stretch of shoreline at the north end of Topsail Island and
142
potential reference barrier islands to the north differed in geomorphology. Nevertheless, analysis
143
of samples collected prior to the disposal events indicated that benthic macrofauna did not
144
display a naturally confounding north-south gradient in abundance. All study sites were exposed
145
to the open ocean and experienced a mean tidal range of about 1 m.
146
To assess effects of each beach disposal event on sediment grain size and sorting,
147
turbidity, and benthic macrofaunal abundances, we conducted monthly (warm-season) to
148
bimonthly (cold-season) sampling at each of the four study sites between October 1998 and
6
149
September 2000. This design spanned times from before the first disturbance event, between
150
the two events, and after the second. Within each site and for each sampling date, we sampled
151
on morning low tides under calm ocean conditions along three transects extending from the
152
toe of the dune to a water depth of 1 m. The three transects were each 40 m apart, with the
153
position of the first randomly determined on each sampling date. This arrangement ensured
154
that we did not repeatedly sample the exact same patches of habitat across time, and that our
155
samples on any given date were spatially independent (patches of Donax spp. and Emerita
156
talpoida can be up to 15 m in diameter; L. Manning, pers. obs.). Along each transect, we
157
stratified our sampling of sediments and fauna across five tidal zones (see McLachlan 1980,
158
Peterson et al. 2006): (1) the supra-tidal, extending from the toe of the dune to the high-tide
159
drift line; (2) the high intertidal, seaward of the drift line, and spanning the area where sand
160
dries during low tide; (3) the mid-intertidal zone which remains wet during low tide even
161
after gravitational water loss; (4) the low intertidal, or swash zone, the area of final run-up
162
and -down of waves at low tide; and (5) the shallow subtidal which extended from the lower
163
margin of the swash zone to 1 m depth into the surf zone at low tide.
164
165
Field sampling
166
To monitor changes in the topography of the beach as a result of filling, we measured
167
beach profiles before (4 April 1999) and immediately after (25 June 1999) the 1999 sediment
168
disposal, and a year later, before (5 April 2000) and after (15 May 2000) the 2000 disposal event.
169
Within two hours of each low tide, a single profile was produced for each of the four field sites,
170
measured from a fixed stake behind the primary dune and extending from the top of the supra-
171
tidal to 1 m deep in the surf zone. Vertical measurements were made using a Topcon AT-2
172
Autolevel and a 7.5 m telescoping rod, and horizontal measurements with a measuring tape (each
7
173
to the nearest cm). Profiles were used to calculate percent changes in beach sediment volume
174
from before to after each fill event and between the two before times.
175
To test the hypotheses that beach disposal would decrease median sediment grain size
176
and reduce variation in particle size, a 4.8 cm diameter x 10 cm deep core of sediment was
177
collected from the middle of each zone on each transect, on every sampling date. In the
178
laboratory, each sediment core was rinsed with de-ionized water to remove all salt, silts and
179
clays, and dried to constant weight at 90° C. Dried samples were sieved on a 2 mm screen to
180
remove gravels, then split to a weight of 30-70 g and passed through a nested series of nine
181
sieves, of mesh sizes 1414, 1000, 707, 500, 354, 250, 177, 125 and 89 µm (Folk 1980). The
182
weight of each fraction (one gravel and ten sand) was recorded to the nearest 0.01 g, and
183
median grain size (in µm) and sorting (the inverse of the spread of the grain size distribution)
184
were calculated for each sample (Folk 1980).
185
To assess impacts of beach disposal on the turbidity of the adjacent surf zone, on
186
sampling dates between November 1998 and May 2000 we collected two 23-ml surface water
187
samples from the bottom of each transect, in water 1-2 m deep. The turbidity of each water
188
sample was measured in the field using a portable Orbeco-Hellige turbidity meter. On 28
189
May 1999 and on 5 April 2000, during active pumping of sediment slurry onto the beach, we
190
also took 3 replicate surface water samples from 1 and 2 m deep water at distances 20 m up-
191
current of, and 20 m and 70 m down-current of the discharge point. The two water depths
192
were chosen to represent the swash and surf zones, respectively.
193
Samples for assessing the impact of beach disposal to benthic macrofauna were collected
194
using a hand-corer of 10.1-cm internal diameter x 20 cm depth. Initially (October 1998 – June
195
1999) we collected four cores from each zone of each transect, but this was later increased to
196
eight (between July 1999 – September 2000) due to the very low densities of many taxa,
197
especially amphipods. Sample contents were pooled within zone and sieved over a 1 mm mesh,
8
198
with faunal densities standardized according to the total area cored. On each date, we also
199
measured the width of each zone so that we could weight zone-specific densities of fauna by
200
zone widths to obtain an unbiased estimate of total density within a 1 m wide transect spanning
201
the beach width (see Peterson et al. 2006). Sieved samples were fixed in 10% buffered formalin
202
and invertebrates were enumerated by species in the laboratory. Lengths of all bivalves (longest
203
anterior-posterior dimension) and mole crabs (anterior-posterior length of carapace) were
204
measured with vernier calipers to the nearest 0.01 cm.
205
206
207
Clam growth under elevated turbidity
To test the hypothesis that elevated turbidity can affect the growth of filter-feeding
208
beach invertebrates, we conducted a 19-day mesocosm experiment. We compared increase in
209
weight and shell length at low and high densities of the surf clam Donax variabilis between
210
conditions of ambient and elevated turbidity. The experiment was conducted in twelve 2.1-m
211
long, 1.2m-wide,0.6 m deep, outdoor wave tanks situated on the northern shore of Bogue
212
Sound, North Carolina. Each tank was partially filled with beach sand, which was sloped to
213
create a deeper subtidal basin and an intertidal swash zone. At the subtidal end of each tank,
214
waves were produced at a frequency of 2 min-1 by the continuous flow of unfiltered ~35 ppt
215
seawater into an asymmetric trough that periodically overbalanced, rotated, and emptied
216
(photo in Manning et al. 2013).
217
Six of the tanks, randomly assigned to the control treatment, received unfiltered
218
seawater directly from Bogue Sound. These were maintained at a turbidity of 10.2 to 35.6
219
NTU, which represents the natural range in turbidity at control sites (see Results). The
220
remaining six, assigned to the experimental treatment, were maintained at a daytime turbidity
221
of 80 NTU, well within the range recorded at disposal sites during beach filling. The
222
enhanced turbidity of the experimental treatment was achieved by passing Bogue Sound
9
223
water through a 1400 L header-tank where it manually received pulverized, inorganic kaolin
224
clay (Albion Kaolin Company, Georgia) every 30 min. To replicate diurnal cycles of
225
turbidity on Topsail Island, clay addition to the head-tank was discontinued for 8-10 night-
226
time hours. Clay was kept in suspension within the header-tank by bubbling air from its
227
bottom. During the experiment the water temperature within the tanks ranged from 24º to 27º
228
C and did not vary among treatments.
229
Three replicate tanks of each turbidity treatment were randomly assigned to the low
230
(222 m-2) and three to the high (444 m-2) clam density. Individually marked clams (20, low
231
density; or 40, high density), of mean (+SE) length of 0.83 + 0.01 mm and weight of 0.206 +
232
0.006 g were placed in a 0.5-cm plastic mesh basket measuring 30 cm x 30 cm with 15 cm
233
depth and a single basket of the designated treatment was buried in each tank so that its top
234
edges were flush with the surface of the sediment. At the start of the experiment and after 19
235
days, the anterior-posterior length of each clam was measured to the nearest 0.1 mm using
236
calipers and hand dried, and weighed to the nearest 0.1 mg.
237
238
239
Response of surf fishes to turbidity
To test whether elevated turbidity may affect the local distribution of two locally
240
common surf fishes, juvenile summer flounder (Paralichthys dentatus) and Florida pompano
241
(Trachinotus carolinus), we conducted behavioral experiments. A 4.2 m-long, 1.2 m-wide, x
242
0.6 m-deep outdoor tank was divided into three 1.4 m-long zones (1-3) using string
243
suspended above the surface. Water flowed into the tank at each end: zone 1 received
244
unfiltered seawater, while zone 3 received either unfiltered (control) or high-turbidity (100 –
245
120 NTUs; experimental) seawater from the previously described header tank. Water drained
246
out through a 9 cm high stand-pipe in zone 3. An identical, but non-functional standpipe was
247
placed in zone 1 to remove any confounding effect of structure on experimental treatments.
10
248
Each fish was observed under two sets of experimental conditions, the first without
249
(control) and the second with (experimental) enhancement of turbidity. For each control run,
250
seawater-flow was shut off and a single fish was introduced to the middle zone (2) and
251
observed after 1 min of acclimation. An observer positioned motionless alongside zone 2
252
recorded the total time out of 15 min for summer flounder (7-12 cm in total length; n = 8) and
253
5 min for Florida pompano (5-9 cm in total length; n = 26) that each fish spent in zones 1 and
254
3. Observation interval varied by fish species to reflect differences in mobility. After each
255
fish was observed under control conditions, it was observed under experimental conditions
256
during which seawater flowed into zones 1 and 3.
257
258
Statistical analyses
259
We used separate univariate permutational analyses of variance (PERMANOVAs,
260
Anderson 2005) to test for interacting effects of treatment and time on: (1) median grain size; (2)
261
sediment sorting; (3) turbidity; and (4) density of dominant macrofaunal taxa by species.
262
Although PERMANOVAs assume that data are independent, they do not require that data be
263
normally distributed (Anderson 2005). Hence, in many instances PERMANOVAs are more
264
suitable for analyzing patchy macrofaunal data than analyses of variance. The PERMANOVAs
265
each used Euclidean distances between untransformed data. All but the PERMANOVAs of
266
turbidity had four factors: Treatment (2 levels: Control, Disposal); Site (4 levels: C1, C2, D1, D2,
267
nested within Treatment); Year (2 levels: 1, Oct 98 - Sept 99; 2, Oct 99 – Sept 00) and Month (7-
268
9 levels, depending on the analysis). Month was orthogonal to Year because this sandy beach
269
system displays strong seasonality in its ecology, which in the absence of disturbance, causes
270
predictable patterns of difference in variables among months (Peterson et al. 2006). The
271
PERMANOVA of turbidity replaced the two factors Year and Month with the single factor Time
272
because there was insufficient sampling in 2000 to allow contrasts across years that held season
11
273
constant. Separate analyses of sediment granulometry were done for each of the five elevation
274
zones on the beach, as sediment properties may be expected to naturally vary across this gradient.
275
Analyses on turbidity and fauna used transect values as replicates (see Field Sampling for a
276
description of how faunal densities per 1-m wide transect were calculated) as our interest was in
277
the total reduction of prey abundance for shorebird and surf fish predators. Analyses were run
278
using unrestricted permutation of raw data and, where significant effects were seen at α = 0.05,
279
they were followed by pair-wise a posteriori tests to identify sources of significant differences.
280
Kolmogorov-Smirnov tests assessed differences in the size frequency distributions of
281
Donax variabilis and Emerita talpoida between control and disposal sites during September of
282
each year, when abundances had peaked.
283
Three-way PERMANOVA analyses, with three factors, Turbidity (2 levels: control,
284
enhanced), Density (2 levels: low, high) and Tank (3 levels: nested within Turbidity and
285
Density), tested for density-dependent effects of turbidity on the proportionate shell growth
286
and weight gain of D. variabilis during the mesocosm experiment. Statistical procedures were
287
as for the four-way PERMANOVAs described above.
288
To test for behavioral responses of Florida pompano and summer flounder to
289
turbidity, we first used two-tailed, paired t-tests to confirm that there was no difference in the
290
time fish spent in zones 1 and 3 of the tank in the absence of turbidity. We then used two-
291
tailed paired t-tests to assess differences in time spent at the two ends when clear water was
292
added to zone 1, and turbid water to zone 3.
293
294
RESULTS
295
Field sampling
296
297
At the two disposal sites, there was a 53-56% increase in the sediment volume of the
beach from shortly before to immediately after the 1999 fill event (Fig. 1). By contrast, over the
12
298
same period, the control beaches experienced an 8-11% decrease in sediment volume. By a year
299
after the fill event, the sediments on disposal beaches had returned to within 2% of their starting
300
volumes from before the 1999 disposal event, and the control beaches to within 8%. During the
301
second, smaller fill event of 2000, the disposal sites exhibited a 32-36% increase in sediment
302
volume, while the control beaches changed less than 6% (Fig. 1).
303
The 1999 fill event resulted in a large reduction in the median grain size and an increase
304
in sediment sorting at the disposal sites immediately after sediment addition (PERMANOVA,
305
significant Treatment x Year x Month interaction, see Appendix 2 available in MEPS
306
supplementary material; Fig. 2). Over this period, small increases in the median grain size and a
307
decrease in sediment sorting were evident on control beaches (Fig. 2). These differential changes
308
at disposal sites relative to controls occurred in four of the five zones sampled, with the exception
309
being zone 5, the surf zone, where no significant change in median grain size was detected
310
(Appendix 2; Fig. 2). Within most zones (1, 3-4), the significant differences in granulometry
311
between control and disposal sites did not persist past September 1999, three months after beach
312
filling. Immediately following the 2000 fill event, analogous differences in granulometry
313
between disposal and control treatments were significant only within zone 4 (swash).
314
Nevertheless, at the final sampling date in September 2000, the median sediment grain size was
315
finer at disposal than control sites in zones 1-3, the highest three on the beach (Fig. 2).
316
Active pumping and deposition of dredge spoil was occurring on two of the dates (28
317
May 1999 and 5 April 2000) when turbidity samples were collected. On both occasions, the
318
average surf-zone turbidity on the disposal beaches was significantly elevated as compared to
319
control beaches (1999: 170 vs. 10 NTUs; 2000: 27 vs. 7 NTUs, Fig. 3; PEMANOVA, significant
320
Time x Treatment interaction, see Appendix 3 available in MEPS supplementary material).
321
Additional water samples systematically collected in the swash and surf zones indicated turbidity
322
levels around 300 NTUs near the outlet pipe, and smaller but substantial elevations of turbidity to
13
323
levels of 70-110 NTUs at distances of 20 and 70 m away in the down-current direction. Turbidity
324
increases on disposal beaches were smaller on 5 April 2000 because the long-shore current on
325
that date happened to flow from the discharge pipe away from the sampling transects. In 1999,
326
turbidity levels remained significantly elevated at disposal sites as compared to controls for about
327
two months after pumping had ceased. On only one sampling date (3 May 1999), when side-cast
328
dredging was being conducted in New River Inlet, was turbidity elevated at control relative to
329
disposal beaches (Fig. 3).
330
Twenty-three macrobenthic invertebrate species from 4 phyla (Annelida, Nemertea,
331
Mollusca, Arthropoda) were collected and identified from core sampling. Two species (the
332
bivalve mollusc Donax variabilis and the small polychaete Scolelepis squamata) constituted 89%
333
of the total number of organisms collected. The next most abundant taxa were the haustoriid
334
amphipods, Haustorius sp. (accounting for 3% of all organisms), Amphiporeia virginiana (3%)
335
and Parahaustorius longimerus (2%) as well as the mole crab, Emerita talpoida (2%). Most taxa
336
occurred in all five elevation zones, although distributions were not constant across zones. Donax
337
(D. variabilis and D. parvula) and E. talpoida were most abundant in the low intertidal (zone 3)
338
and swash zone (zone 4). The haustoriid amphipods and S. squamata were most abundant in the
339
surf zone (zone 5).
340
Impacts of beach filling on faunal abundances varied among the six most abundant taxa
341
and between the two fill events (PERMANOVA, significant Treatment x Year x Month,
342
Treatment x Month, and Treatment x Year interactions; see Appendix 4 available in MEPS
343
supplementary material; Fig. 4). In 1999, during which disposal occurred prior to the recruitment
344
of most sandy beach invertebrates (examine controls relative to period of first disposal in Fig. 4),
345
the disturbance had no detectable impact on the abundance of D. variabilis, S. squamata, or
346
Haustorius sp., but resulted in reduced abundances of E. talpoida, P. longimerus and A.
347
virginiana at disposal sites relative to controls. Among these, E. talpoida was particularly
14
348
affected, with peak abundance at disposal sites four-fold lower than at control sites. Over the
349
winter months (December to February), during which all taxa displayed a large seasonal
350
reduction in abundance, densities converged between disposal and control treatments. Following
351
the 2000 disposal event, reduced abundances of the amphipods P. longimerus and A. virginiana
352
were again seen at disposal as compared to control sites. E. talpoida, by contrast, did not
353
detectably differ between disposal and control sites following the 2000 fill event, but instead, D.
354
variabilis and Haustorius sp., which had begun recruiting to beaches prior to the 2000 disposal
355
event (examine controls relative to period of second disposal in Fig. 4), displayed reduced
356
abundances at disposal relative to control sites for several months following the disturbance. The
357
polychaete, S. squamata, which also had commenced recruitment by the time of the 2000
358
disposal, peaked in July at three times the abundance at disposal as compared to control sites. By
359
fall convergence in the abundance of taxa between control and disposal sites was evident in most
360
instances.
361
Following each disposal event, differences in the size frequency distributions of D.
362
variabilis were apparent (Fig. 5; Kolmogorov Smirnov tests: 1999, D = 0.60, df =1, p < 0.001;
363
2000, D = 0.45 df = 1, p < 0.001). At disposal sites, D. variabis >15 mm in shell length were
364
notably absent, and the median shell lengths were consequently lower (1999: 4.8 mm; 2000: 4.2
365
mm) than at control sites (1999: 7.0 mm; 2000: 4.5 mm). Differences in the size frequency
366
distributions of E. talpoida between disposal and control sites were smaller than for D. variabilis
367
(Fig. 5; Kolmogorov Smirnov tests: 1999, D = 0.15, df =1, p = 0.04; 2000, D = 0.26, df = 1, p =
368
0.03). In 1999, during which very few E. talpoida were recorded from disposal sites, the median
369
carapace length of mole crabs was 3.3 mm at control sites but 4.4 mm at disposal sites. In 2000,
370
when much greater recruitment of E. talpoida to disposal sites was recorded there was little
371
difference in median carapace length between the treatments (disposal: 42 mm, control 41 mm).
372
15
373
374
Clam growth under elevated turbidity
Elevated turbidity significantly reduced the proportionate shell growth and weight
375
gain of clams at the low but not at the high clam density (PERMANOVA, significant
376
Turbidity x Density interaction; see Appendix 5 available in MEPS supplementary material).
377
Clams at high density exhibited a mean (± 1 SE) proportionate growth in shell length of 0.49
378
± 0.01 mm, irrespective of turbidity, whereas clams at low density exhibited a proportionate
379
shell growth of 0.49 ± 0.02 mm under ambient turbidity and 0.35 ± 0.01 mm under enhanced
380
turbidity. The proportionate gain in weight was 1.61 ± 0.10 among clams deployed at high
381
density in both turbid and control tanks, 1.59 ± 0.13 in control tanks with a low clam density
382
versus 1.11 ± 0.08 in turbid tanks with a low clam density.
383
In tanks with a high clam density the turbidity dropped by 64.4 ±11.4 NTUs between
384
the addition of clay and 25 min afterwards. In the tanks of lower clam density the drop in
385
turbidity was, at 35.7 ±5.2 NTUs, significantly less (PERMANOVA, p < 0.05). Thus
386
turbidity levels were drawn down more rapidly in tanks with more clams filtering the
387
relatively confined waters.
388
389
390
Response of surf fishes to turbidity
In the absence of a turbidity difference between the two ends of experimental
391
mesocosms (zones 1 and 3), both the Florida pompano and the summer flounder spent similar
392
amounts of time at each (paired t-test; pompano: t = 0.67, df = 25, p = 0.509; flounder: t =
393
1.17, df = 7, p = 0.280; Fig. 6). When, however, clear water was added to one end (zone 1)
394
and turbid water to the other (zone 3), the Florida pompano spent significantly more time (by
395
almost 5 fold) in the clear water of zone 1 than the turbid water of zone 3 (paired t-test; t =
396
14.53, df = 50, p <0.001), while the summer flounder spent more time (by over 16 fold) in the
397
turbid zone 3 (paired t-test; t = -6.86, df = 7, p < 0.001; Fig. 6).
16
398
399
DISCUSSION
400
To improve mechanistic understanding of impacts to sandy beach ecosystems of
401
beach disposal of fine sediments, our study coupled field monitoring of changes to biotic and
402
abiotic variables following two successive dredge spoil disposals on Topsail Island, North
403
Carolina with manipulative mesocosm experiments directly assessing biological impacts of
404
enhanced turbidity. We predicted that where dredge spoil was rapidly eroded from the
405
nourished beach following deposition, impacts to sediment-dwelling invertebrates would be
406
short-lived. To the contrary, we found sizeable impacts of dredge spoil disposal on several
407
invertebrate taxa that extended nearly an entire year. Furthermore, our mesocosm
408
experiments revealed effects of turbidity plumes, of the magnitude generated by the
409
nourishment events, on the growth clams and the habitat choice of surf-fish predators.
410
As expected, the predominantly fine sediments added to Topsail Island during
411
disposal of dredge spoil were rapidly eroded from the beach. Beach profiling indicated that
412
essentially the entire volume of sediment added to the beach in May 1999 was eroded away
413
by April 2000. Despite rapid sediment erosion, a significant reduction, relative to controls, in
414
mean sediment grain size was apparent at all elevations of the nourished beaches across the
415
entire summer season following each spring nourishment event, during which faunal
416
abundances typically peak. Complete convergence of sediment properties between control
417
and nourished beaches did not occur until March 2000, following the first nourishment event,
418
and had not occurred by October 2000, the final sampling date following the second
419
nourishment event.
420
Although the deposition of dredge spoil on north Topsail Island had only fleeting
421
benefits in nourishing the beach and hence protecting coastal property against storm erosion
422
and property loss, it induced large and statistically significant impacts on abundances of all
17
423
six numerically dominant infaunal invertebrates on the sandy beach. The magnitude of these
424
differences between disposal and control sites varied among taxa and between years – and not
425
uniformly corresponding to the two-fold larger volume of spoil deposited in 1999 than in
426
2000. Of the five taxa whose abundances responded negatively to disposal, one, the mole
427
crab Emerita talpoida, was significantly affected only in 1999, and two, the bean clam Donax
428
variabilis and the amphipod Haustorius, were detectably impacted only in 2000, whereas the
429
remaining two, the amphipods Parahaustorius longimerus and Amphiporeia virginiana,
430
displayed impacts in both years. The one species positively affected by disposal, the spionid
431
polychaete Scolelepis squamata, exhibited a stronger response in 2000 than in 1999.
432
At least four mechanisms may have explained the invertebrate responses to spoil
433
disposal. First, mortality may have been induced through burial, crushing, and suffocation, by
434
the sediments as well as by the bulldozers used to spread them (e.g., Peterson 1985, Thrush et
435
al. 2003). Second, the turbidity generated during deposition of the slurry of dredge spoils
436
may have clogged gills and palps of filter-feeding invertebrates, also leading to mortality
437
(Reilly and Bellis 1983). Third, environmental changes, such as shifts in water clarity or
438
sedimentology, may have altered predation rates by surf fishes or shorebirds on beach
439
invertebrates. Fourth, switching the granulometry from the native medium sands to fine sands
440
may have influenced behavioral habitat selection of dispersing larvae of Donax, Emerita, or
441
Scolelepis, or juveniles of the direct-developing haustoriid amphipods. The first three
442
mechanisms, each of which involves post-settlement mortality, require that the disposal event
443
succeed the seasonal recruitment pulses that elevate infaunal abundances from near zero in
444
winter to high numbers in the warm season. The fourth, in contrast, requires that disposal of
445
incompatible sediments precede or overlap with the seasonal recruitment period.
446
447
Our frequent sampling on a monthly basis during the warm season provided the
ability to infer the timing of the disposal relative to the major annual recruitment pulse for
18
448
each abundant taxon and hence to disentangle the likely mechanisms by which each was
449
impacted. In each year, disposal preceded the initiation of the seasonal increase of the
450
Emerita talpoida population via recruitment of planktonic propagules. This suggests that the
451
observed suppressions of Emerita abundance were caused by habitat selection by planktonic
452
megalopal larvae, which in central North Carolina exhibit their major settlement in June-July
453
(Diaz 1980). Emerita talpoida is known to be suppressed in abundance by finer sediments
454
(Hayden & Dolan 1974). The much greater suppression of Emerita recruitment we observed
455
in 1999 compared to 2000 was consistent with the two-fold difference between years in
456
quantity of fine sediment deposition.
457
Recruitment of Donax variabilis and all three haustoriid amphipod taxa occurred after
458
the 1999 disposal but midway through the 2000 disposal process. The 1999 event did not
459
detectably influence recruitment of Donax or Haustorius sp., and the 2000 event caused a
460
reduction in but not a complete failure of spring recruitment of these taxa. This response
461
implies that disposal suppressed recruitment success of new recruits without much influence
462
on adults of these taxa. Small Donax burrow through sediments much more slowly than
463
larger ones (Nel et al. 2001), implying less resilience to burial and resistance to suffocation
464
under sediments among recent recruits and partial immunity among adults to sedimentation at
465
levels applied on Topsail Island.
466
Both of the amphipods P. longimerus and A. virginiana displayed virtual recruitment
467
failures in each year, independent of whether disposal occurred midway through or after the
468
recruitment season. These recruitment failures at disposal sites occurred despite sustained
469
populations at control sites. The near absence at disposal sites of these direct-developing
470
amphipod taxa with limited dispersal capacity may reflect an impact of disposal on adults,
471
needed nearby to seed the population recovery. Silt and clay are known to inhibit the
472
burrowing capabilities of P. longimerus, particularly when coupled with cold winter
19
473
temperatures (Maurer et al. 1981). Hence, when buried by disposal sediments, typically
474
deposited on beaches in winter or early spring, even adult amphipods of this species may
475
suffer particularly high mortality rates because they are unable to escape toxic levels of
476
ammonium and sulfides, or move effectively into more oxygenated sediments (Maurer et al.
477
1985). A. virginiana, which usually inhabits the top 0-2.5 cm of sediments (Croker and
478
Hatfield 1980), may also be more susceptible to death by burial than many other species of
479
amphipod. Previous research has shown that disturbed, populations of A. virginiana take
480
several years to recover (Jamarillo et al. 1987). Here, the time interval between disturbances
481
(a year) was shorter than the required recovery time observed by Jaramillo and colleagues.
482
We hypothesize that the only taxon to exhibit an enhancement of abundance on the
483
disposal sites, Scolelepis squamata, was responding behaviorally to the increase in finer
484
sediments. In laboratory experiments, this broadly distributed opportunistic polychaete
485
displayed a preference for medium-fine sediment (Van Tomme et al. 2013). As required for
486
behavioral habitat selection to explain the Scolelepis patterns, both disposal events at Topsail
487
Island preceded the time period when Scolelepis density exhibited its largest seasonal
488
increase at disposal sites. Finer sediments are presumably richer in organic matter that may
489
provide a food supplement for this spionid polychaete and thereby reduce starvation
490
mortality. Scolelepis is likely to practice mostly deposit feeding in its turbulent, shallow
491
subtidal habitat (zone 5) on the sandy beach. Alternatively, the enhancement of Scolelepis
492
may reflect decreased foraging efficiency by predatory fish like pompano in the more turbid
493
environment (Manning et al. 2013) or decreased inter-specific competition with other
494
invertebrates.
495
Impacts to fauna, where occurring, persisted throughout the warm season of normally
496
high densities with recovery not occurring until the next major recruitment event. Annual
497
spring-time repetition of the spoil disposal prevented recovery of those species that recruit via
20
498
dispersing larvae, while new sediment deposition and lack of rapid recovery of locally
499
breeding adults probably combined to retard recovery of direct-developing taxa. The net
500
consequence of suppressing populations of Donax, Emerita, and three haustoriid amphipods,
501
while enhancing abundance of the spionid polychaete, Scolelepis, was a large reduction in
502
integrated warm-season biomass of these invertebrates, because Donax and Emerita are much
503
larger than polychaetes. Because benthic macroinvertebrates of sandy beaches deliver the
504
valued ecosystem services of providing important, accessible, and dense food resources for
505
crabs, juvenile surf fishes, and resident, migrating, and breeding shorebirds (McLachlan &
506
Brown 2006), spoil disposal has the potential to cause losses of prey subsidies to higher
507
trophic levels. Loss of these prey resources from ocean beaches can have large cascading
508
impacts on shorebirds (Dugan et al. 2003, Peterson et al. 2006) and fishes (Lasiak 1986,
509
Hackney et al. 1996).
510
In addition to altering granulometry and abundances of beach invertebrates, the
511
disposal of dredge spoils onto Topsail Island increased the turbidity of the surf zone, both
512
during active pumping of the sediment slurry onto the beach and during subsequent wave-
513
induced erosion of fine particles from the disposal sites. The enhancement of turbidity in the
514
surf zone of fill sites exceeded the allowable North Carolina standard of 25 NTUs by as much
515
as a factor of 12 (Fig. 3). Through experiments in wave-tank mesocosms, we demonstrated
516
that these turbidity levels were sufficient to suppress individual growth of suspension-feeding
517
Donax variabilis, provided that clams are held at low enough density to avoid the
518
experimental artifact of artificially dense clams rapidly filtering down the turbidity treatment.
519
Field sampling revealed that both Donax and Emerita had smaller body lengths at disposal
520
sites than on control beaches during almost all of the two-year sampling period. Although
521
these size differences mostly reflected demographic differences in age distributions, stunting
522
of growth via periodic exposure to elevated turbidity presumably also contributed to
21
523
maintaining the smaller body sizes of these biomass dominants. This process of growth
524
reduction represents an additional mechanism by which dredge spoil disposal reduced prey
525
biomass available for predators of Donax and Emerita, especially critical for migrating red
526
knots, dunlin, and sanderlings, and for resident juvenile pompano.
527
Besides its negative impact on individual growth of infaunal prey, elevated turbidity
528
had direct effects on visually orienting predators. In mesocosm experiments, pompano
529
avoided turbid water, whereas summer flounder preferred it over clearer conditions. Wilber et
530
al. (2003) similarly demonstrated avoidance of turbidity generated from beach nourishment
531
by the visually orienting bluefish and attraction to turbid fill sites by the northern kingfish in
532
New Jersey. Avoidance of turbidity makes sense for visually orienting predatory fishes such
533
as pompano (Manning & Lindquist 2003, Manning et al. 2013) and diving seabirds such as
534
terns (Cyrus & Blaber 1987) because of impaired ability to detect prey. In contrast, flounders
535
are ambush predators which, due to their camouflage by sand, do not need to be able to see
536
more than a few centimeters in order to capture approaching prey. They may gain more in
537
fitness by becoming further hidden from their prey than they lose through reduction of their
538
own visual acuity.
539
As rising sea levels and an increasing frequency of violent storms from global climate
540
change enhance risk of damage to oceanfront property, public support for beach nourishment
541
appears to be growing (Peterson & Bishop 2005). Our study suggests that beach nourishment
542
projects utilizing fine sediments from dredge spoil have trivial public benefit, yet significant
543
negative ecological impacts. Following placement on Topsail Island, the fine-grained
544
sediments obtained from maintenance dredging of navigation channels eroded away so
545
rapidly that they did not bolster the volume of protective beach sands over the high storm-risk
546
seasons. Instead, the spring-time nourishment with fine sediments depressed invertebrate
547
populations for almost a whole year such that their populations had barely recovered by the
22
548
next annual disposal event, thereby generating a cumulative impact from successive
549
nourishment events. Hence, such beach disposal of dredge spoil represents governmentally
550
sanctioned habitat degradation without requirements for mitigation, which is required in the
551
U.S. for even accidental injury to other highly productive marine benthic habitats (e.g.,
552
Fonseca et al. 2000). A way forward may lie in instead revisiting a former use of dredge
553
spoils to construct islands in sounds and estuaries. Estuarine islands have experienced
554
dramatic erosion, of as much as 50% of average island area in the Chesapeake Bay, for
555
example (Erwin et al. 2007), resulting in loss of vegetation and sedimentary habitats for bird
556
roosting and foraging. Beneficial uses of fine dredge spoils are sorely needed along with
557
prohibition of spoil disposal on ocean beaches.
558
559
Acknowledgements. We thank C. Tallent, T. Riley, J. Grabowski, S. Powers and H.C.
560
Summerson for their assistance with field and laboratory work. O. Defeo provided helpful
561
comments on an earlier version of this manuscript. M. Kenworthy offered useful insight into fish
562
ecology and behavior. This work was supported in part by a doctoral fellowship from the
563
University of North Carolina, a North Carolina Fisheries Resource Grant, and North Carolina Sea
564
Grant mini-grants. M. Bishop was supported by the Brian Robinson Fellowship and the
565
Macquarie University Outside Studies Program during preparation of the manuscript.
566
567
568
569
570
571
LITERATURE CITED
Anderson MJ (2005) Permutational multivariate analysis of variance. Department of
Statistics, University of Auckland, Auckland, NZ
Basco DR (1999) Overview of Beach Engineering in the United States of America. Old
Dominion University, Virginia, USA
23
572
Brazeiro A, Defeo O (1999) Effects of harvesting and density-dependence on the
573
demography of sandy beach populations: the yellow clam Mesodesma mactroides of
574
Uraguay. Mar Ecol Prog Ser 182:127-135
575
Cleary WJ, Pilkey OH (1996) Cape Lookout to Cape Fear, North
576
Carolina, regional overview. In: Cleary WJ (ed) Environmental coastal geology: Cape
577
Lookout to Cape Fear. Carolina Geological Society, Wilmington, North Carolina, USA,
578
p 87-128
579
580
581
582
583
584
585
Cooke BC, Jones AR, Goodwin ID, Bishop MJ (2012) Nourishment practices on Australian
sandy beaches: a review. J Env Manag 113:319-327
Croker RA, Hatfield EB (1980) Space partitioning and interactions in an intertidal sandburrowing amphipod guild. Mar Biol 61:79-88
Cyrus DP, Blaber SJM (1987) The influence of turbidity on juvenile marine fishes
in estuaries. J Exp Mar Biol Ecol 109:71-91
Defeo O, de Alava A (1995) Effects of human activities on long-term trends in sandy beach
586
populations: the Edge clam Donax hanleyanus in Uraguay. Mar Ecol Prog Ser 123:73-82
587
Defeo O, McLachlan A (2005) Patterns, processes and regulatory mechanisms in sandy beach
588
589
590
591
592
593
macrofauna: a multiscale analysis. Mar Ecol Prog Ser 295:1-20
Defeo O, McLachlan A, Schoeman DS, Schlacher TA, Dugan J, Jones A, Lastra M, Scapini F
(2009) Threats to sandy beach ecosystems: a review. Est Coast Shelf Sci 81:1-12
Diaz H (1980) The mole crab Emerita talpoida (Say): a case of changing life history pattern.
Ecol Monogr 50:437-456
Dugan JE, Hubbard DM, McCrary MD, Pierson MO (2003) The response of macrofauna
594
communities and shorebirds to macrophyte wrack subsidies on exposed sandy beaches of
595
southern California. Est Coast Shelf Sci 58:25-40
24
596
597
598
Dugan JF, Hubbard DM, Rodil IF, Revell DL, Schroeter S (2008) Ecological effects of
coastal armoring on sandy beaches. Mar Biol 29:160-170
Dugan JE, Defeo O, Jaramillo E, Jones AR, Lastra M, Nel R, Peterson CH, Scapini F,
599
Schlacher T, Schoeman DS (2010) Give beach ecosystems their day in the sun. Science
600
329:1146
601
602
603
604
605
606
607
608
609
Erwin M, Miller J, Reese JG (2007) Poplar Island environmental restoration project:
Challenges in waterbird restoration on an island in Chesapeake Bay. Ecol Rest 25:256-262
Folk RL (1980) Petrology of sedimentary rocks. Hemphill Publishing Company, Austin,
Texas, USA
Fonseca MS, Julius BE, Kenworthy WJ (2000) Integrating biology and economics in seagrass
restoration: how much is enough and why? Ecol Eng 15:227-237
Gorzelany JF, Nelson WG (1987) The effects of beach replenishment on the benthos of a subtropical Florida beach. Mar Env Res 21:75-94
Greene K (2002) Beach nourishment: a review of the biological and physical impacts. Atlantic
610
States Marine Fisheries Commission, ASMFC Habitat Management Series no.7,
611
Washington, DC, USA
612
Hanson H, Brampton A, Capobianco M, Dette HH, Hamm L, Laustrup C, Lechuga A,
613
Spanhoff R (2002) Beach nourishment projects, practices, and objectives - a European
614
overview. Coast Eng 47:81-111
615
616
617
618
619
620
Hayden B, Dolan R (1974) Impacts of beach nourishment on distribution of Emerita talpoida, the
common mole crab. J Waterw Harb Coast Eng Div 100:123-132
Hubbard DM, Dugan JE (2003) Shorebird use of an exposed sandy beach in southern
California. Est Coast Shelf Sci 58:41-54
IPCC (2007) Summary for Policymakers. In: Parry ML, Canziani OF, Palutikof JP, van der
Linden PJ, Hanson CE (eds) Climate change 2007: impacts, adaptation and vulnerability.
25
621
Contribution of Working Group II to the Fourth Assessment Report of the
622
Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK,
623
p 7-22
624
625
626
Jaramillo E, Croker RA, Hatfield EB (1987) Long-term structure, disturbance, and recolonization
of macroinfauna in a New Hampshire sand beach. Can J Zool 65:3024-3031
Jaramillo E, Dugan JE, Hubbard DM, Melnick D, Manzano M, Duarte C, Campos C, Sanchex R
627
(2012) Ecological implications of extreme events: footprints of the 2010 earthquake along
628
the Chilean coast. PLoS ONE 7:e35348
629
630
631
Lasiak TA (1986) Juveniles, food and the surf zone habitat: implications for teleost
nursery areas. S Afr J Zool 21:51-56
Leewis L, van Bodegom PM, Rozema J, Janssen GM (2012). Does beach nourishment have
632
long-term effects on intertidal macro invertebrate species abundance? Est Coast Shelf
633
Sci.112:172-181
634
Manning LM, Peterson CH, Fegley SR (2013) Degradation of surf fish foraging habitat driven by
635
sedimentological modifications caused by beach nourishment. Bull Mar Sci 89:83-106
636
Maurer D, Keck RT, Tinsman JC, Leathem WA (1981) Vertical migration and mortality of
637
638
benthos in dredged material: Part II—Crustacea. Mar Env Res 5:301-317
Maurer D, Church TM, Wethe C, Lord C (1985) Marine benthos in relation to pore water
639
chemistry and sediment geochemistry of simulated dredged material. Int Rev gesamten
640
Hydrobiol Hydrogr 70:369-377
641
642
643
644
McClenachan L, Jackson JB, Newman MJ (2006) Conservation implications of historic sea turtle
nesting beach loss. Front Ecol Env 4:290-296
McLachlan A (1980) Intertidal zonation of macrofauna and stratification of meiofauna on highenergy sandy beaches in the Eastern Cape, South Africa. Trans R Soc S Afr 44:213-223
26
645
646
647
648
649
650
McLachlan A (1996) Physical factors in benthic ecology: Effects of changing particle size on
beach fauna. Mar Ecol Prog Ser 131:205-217
McLachlan A, Brown AC (2006) The ecology of sandy shores. Academic Press, Burlington,
MA, USA
Nel R, McLachlan A, Winter DP (2001) The effect of grain size on the burrowing of two Donax
species. J Exp Mar Biol Ecol 265:219-238
651
Nelson WG (1988) An overview of the effects of beach nourishment on the sand beach fauna.
652
In: Tait LS (ed) Beach Preservation Technology 88. Problems and advancements in
653
beach nourishment. Florida Shore and Beach Preservation Association Inc, Tallahassee,
654
Florida, USA, p 295-310
655
656
657
658
659
660
661
662
663
Parmesan C (2006) Ecological and evolutionary responses to recent climate change. Ann Rev
Ecol Evol Syst 37:637-669
Peterson CH (1985) Patterns of lagoonal bivalve mortality after heavy sedimentation and their
paleoecological significance. Paleobiology 11:139-153
Peterson CH, Bishop MJ (2005) Assessing the environmental impacts of beach nourishment.
BioScience 55:887-896
Peterson CH, Hickerson DHM, Johnson GG (2000) Short-term consequences of nourishment and
bulldozing on the dominant large invertebrates of a sandy beach. J Coast Res 16:368-378
Peterson CH, Bishop MJ, Johnson GA, D’Anna LM, Manning LM (2006) Exploiting beach
664
filling as an unaffordable experiment: benthic intertidal impacts propagating upwards to
665
shorebirds. J Exp Mar Biol Ecol 338: 205-221
666
Pilkey OH, Wright HL (1988) Seawalls versus beaches. J Coastal Res 4:41-64
667
Rakocinski CF, Heard RW, LeCroy SE, McLelland JA, Simons T (1996) Responses by
668
macrobenthic assemblages to extensive beach restoration at Perdido Key, Florida, USA.
669
J Coastal Res 12:326-353
27
670
Reilly F, Bellis FJ (1983) The ecological impact of beach nourishment with dredged
671
materials on the intertidal zone at Bogue Banks, North Carolina. US Army Corps of
672
Engineers, CERC Misc. Report 83-3. Fort Belvoir, Virginia, USA, 74 pp
673
674
675
676
677
Schlacher TA, Dugan J, Schoeman DS, Lastra M, Jones A, Scapini F, McLachlan A, Defeo O
(2007) Sandy beaches at the brink. Divers Dist 13:556-560
Schlacher TA, Noreiga R, Jones A, Dye T (2012) The effects of beach nourishment on benthic
invertebrates in eastern Australia: impacts and variable recovery. Sci Total Env 435:411-417
Speybroek J, Bonte D, Courtens W, Gheskiere T, Grootaert P, Maelfait JP, Mathys M, Provoost
678
S, Sabbe K, Stienen EWN, Van Lancker V, Vincx M, Degraer S (2006) Beach nourishment:
679
an ecologically sound coastal defense alternative? a review. Aquat Conserv: Mar Freshw
680
Ecosyst 16:419-435
681
Staudinger MD, Grimm NB, Staudt A, Carter SL, Chapin FS, Kareiva P, Ruckelshaus M, Stein
682
A (2012) Impacts of Climate Change on Biodiversity, Ecosystems, and Ecosystem Services:
683
Technical Input to the 2013 National Climate Assessment. Cooperative Report to the 2013
684
National Climate Assessment, 296 pp
685
Thrush SF, Hewitt JE, Norkko A, Kummings VJ, Funnell GA (2003) Macrobenthic recovery
686
processes following catastrophic sedimentation on estuarine sandlfats. Ecol Appl
687
13:1433-1455
688
689
690
Valverde HR, Trembanis AC, Pilkey OH (1999) Summary of beach nourishment episodes on the
U.S. East coast barrier islands. J Coastal Res 15:1100-1118
Van Tomme J, Vanden Eede S, Speybroek J, Degraer S, Vincx M (2013) Macrofaunal
691
sediment selectivity considerations for beach nourishment programmes. Mar Env Res
692
84:10-16
693
694
Warwick JA (2013) Dispersal of fine sediment in nearshore coastal waters. J Coast Res
29:579-596
28
695
Wilber DH, Clarke DG, Ray GL, Burlas M (2003) Response of surf zone fish to beach
696
nourishment operations on the northern coast of New Jersey, USA. Mar Ecol Prog Ser
697
250:231-246
29
698
FIGURE LEGENDS
699
Fig. 1. Elevation profiles of pooled disposal (D1, D2; grey) and control (C1, C2; black) sites
700
immediately before (entire lines) and after (broken lines) the 1999 and 2000 sediment
701
disposal events.
702
Fig. 2. Mean (± 1 SE) grain size and sorting of sediments at pooled control (black, filled
703
symbols) and disposal (grey, open symbols) sites, before, during (grey shaded areas), and
704
after the two disposal events. Sorting is an index of the tightness of the grain size distribution
705
(the inverse of variance). Significant (at α = 0.05) a posteriori tests for significant Year x
706
Month x Treatment interactions (PERMANOVA; Appendix B) are denoted with asterisks (*).
707
Definitions of zones are as described in the Materials and Methods, with Zone 1 in the
708
supratidal and Zone 5, the surf zone. n = 6.
709
Fig. 3. Mean (± 1 SE) turbidity in the surf zone of pooled disposal (black, filled symbols) and
710
control (grey, open symbols) sites, before, after and during (grey shaded areas) the two
711
disposal events. Significant (at α = 0.05) a posteriori tests for significant Year x Month x
712
Treatment interactions (PERMANOVA; Appendix C) are denoted with asterisks (*). n = 6.
713
Fig. 4. Mean (± 1 SE) abundances per transect of each of the numerically dominant taxa of
714
beach macroinfauna at pooled control (C; black, filled symbols) and disposal beaches (D;
715
grey open symbols), before and after (grey shaded areas) disposal events. Significant (at α =
716
0.05) a posteriori tests for significant Year x Month x Treatment interactions
717
(PERMANOVA; Appendix D) are denoted with asterisks (*). Parahaustorius longimerus
718
and Amphiporeia virginiana displayed main effects of Treatment (disposal vs. control). n = 6.
719
Fig. 5. Size frequency histograms of Donax variabilis shell height and Emerita talpoida
720
carapace width at control (black bars) and disposal (grey bar) sites during September
721
recruitment peaks of each year.
30
722
Fig. 6. Mean (+ SE) time (in seconds) spent by (a) Florida pompano and (b) summer
723
flounder at two ends (zone 1, 3) of an experimental mesocosm in the absence of turbidity (-T)
724
and following the addition of clear water to zone 1 and turbid water to zone 3 (+T). The total
725
duration of the experiment was 5 min (=300 sec) for pompano and 15 min (=900 sec) for
726
flounder. n = 26 pompano and 8 flounder.
31
Year 1: 1999
Year 2: 2000
A. D1
8
Before (5 April 2000)
After (15 May 2000)
Before (4 April 1999)
After (25 June 1999)
6
4
2
0
Elevation (m above mean low water springs)
8
B. D2
6
4
2
0
8
C. C1
6
4
2
0
8
D. C2
6
4
2
0
0
50
100
150
0
50
100
150
Distance (m)
FIG. 1
32
MEDIAN GRAIN SIZE
1200
SEDIMENT SORTING
A. Zone 1
1000
Control
Disposal
* *
*
800
* **
500
400
*
0
0
B. Zone 2
1200
1000
*
*
800
*
*
* *
400
*
**
*
500
0
0
C. Zone 3
Grain size (m)
1200
1000
*
*
800
*
*
500
0
0
D. Zone 4
1000
*
**
*
800
*
*
500
400
*
0
1200
*
*
400
1200
*
0
E. Zone 5
1000
*
*
800
500
400
01-Sep-00
01-Jun-00
01-Mar-00
01-Dec-99
01-Sep-99
01-Jun-99
01-Mar-99
01-Dec-98
01-Sep-00
01-Jun-00
01-Mar-00
01-Dec-99
01-Sep-99
01-Jun-99
01-Mar-99
0
01-Dec-98
0
Date
FIG. 2
33
FIG. 3
01-May-00
*
01-Feb-00
01-Nov-99
*
*
01-Aug-99
01-May-99
50
01-Feb-99
01-Nov-98
Turibidty (NTU)
200
Control
Disposal
150
100
*
*
*
0
Date
34
FIG. 4
E. Parahaustorius longimerus
60
C>D
*
20
20
40
10
10
20
0
0
01-Oct-00
0
*
01-Jul-00
*
01-Apr-00
*
01-Jan-00
0
01-Oct-99
*
01-Jul-99
0
*
01-Apr-99
*
*
01-Jan-99
30
*
600
01-Oct-98
0
B. Emerita talpoida
01-Oct-00
*
01-Jul-00
D. Haustorius sp.
5
01-Apr-00
10
01-Jan-00
*
01-Oct-99
*
01-Jul-99
40
01-Apr-99
* *
15
01-Jan-99
A. Donax variabilis
01-Oct-98
20
01-Oct-00
Control
Disposal
01-Jul-00
01-Apr-00
01-Jan-00
01-Oct-99
01-Jul-99
*
01-Apr-99
30
01-Jan-99
01-Oct-98
*
Total abundance (1000 animals per 1 m wide transect)
60
C. Scolelepis squamata
*
400
*
*
200
*
F. Amphiporeia virginiana
C>D
*
Date
35
A. Donax variabilis
B. Emerita talpoida
September 1999
20
25
CONTROL
CONTROL
DISPOSAL
DISPOSAL
20
15
15
Frequency (no. of observations)
10
10
5
5
0
0
September 2000
20
25
20
15
15
10
10
5
5
0
0
0
5
10
15
20
0
Length (mm)
FIG. 5
5
10
15
20
0
5
10 15 20 25 30 0
5
10 15 20 25 30
Length (mm)
36
250
A. Florida pompano
*
Zone 1
Zone 3
200
150
100
Time (sec)
50
0
800
B. Summer flounder
*
600
400
200
0
-T
+T
Treatment
FIG. 6
37