Document 11340950

advertisement
..I'
Proc~edillg.r-Fir~
Eff~cls ort Rar~ and End.v.g~r~d
S~cie.r and Habitats Conf~r~IIC~,NOII.1J-16, 1995.
Coew d'Al~M, Idaho.
to lAWF, 1997. Print~d ill U.S.A.
1 23
Effects of the Hatchery Complex Fires on Northern Spotted Owls in
the Eastern Washington Cascades
William L. Gaines,Robert A. Strand, and SusanD. Piper
u.s. ForestService.600 Sherbourne.Leavenworth.WA98826
Tel. (509) 782.-1413; Fax (509) 548-5817
Abstract. During the summer of 1994the Hatchery
Complex fIres burned 17,603ha in the eastCascadesof
Washington. ThesefIres affectedthree habitatreserves
and six octivity centersfor the northernspottedowl (Strix
occidentalis caurina). The availability of spotted owl
habitatwithin a 2.9km radiusof theseactivitycenterswas
reducedby the direct effectsof the fire, (averagehabitat
loss=31%, range 8%-45%)but wasalso significantly reducedbydelayedtreemortalityandinsectcausedmortality
(averagehabitat10ss=55%,
range10%-85%).Fewerspotted owls occuppiedand reproducedat thesesitesthan in
previous years. Spottedowl habitat locatedin riparian
areasor ona benchwassomewhatmorelikely to remainas
fire refugia thanhabitatlocatedon mid-upperslopes.This
was especiallytrue on southaspects.The availability of
spottedowl habitatundercurrentand inherentfIre regimes
was very dynamic acrossthe eastCascadeslandscape.
Appropriate management
strategiesmay includestrategically located low density fue) areascreatedto protect
adjacentspottedowl habitat
that 'There is a very low probabilitythat any(spottedowl
habitatreserve)createdin theEastCascades
subregionwill
avoid catastrophicwildf"Ire over a significant JX>rtionof
its landscapeover the next century." Thus the management dilema,while fire suppressionmay have increased
the amount of suitable spotted owl habitat across the
lanscape,it has also resultedin a greaterrisk of habitat
loss due to catastrophicfIres (Agee and Edmunds 1992,
Buchananet al. 1995). Managementof theselandscapes
must considerthe short-termhabitatneedsof the spotted
owl and the risk associatedwith standreplacementfires.
Important in the managementof these landscapesis
understandinghow fire affectsspottedowls and theirhabitat. Our objectiveswere to quantify,as muchas possible,
the effectsof the HatcheryComplexfires on spottedowl
habitat.occupancyrates,and reproduction. In addition,
the future risk to additional owl sitesand JX>ssible
managementstrategieswill re presented.
Study Area
Introduction
Firesare a naturaleventwithin the eastsideCascades
ecosystems(Agee 1994),however,flre suppressionand
logging have contributedto higher fuel loadingsthat can
lead to higher fIre intensity than would have inherently
occurred(Agee and Edmunds1992). The fIres of 1994
were a dramatic exampleof this. Areasin which the inherentdisturbanceregime would havebeenlow to moderateintensityand high frequencyfifes,burnedasmoder~
ate to high intensity stand replacementfIres. Severalof
thesefIreswerepartof theHatcheryComplexthatincluded
the Rat, Hatchery,Eightmile, BlackjackI and Blackjack
II fires. In total thesefires burnedabout17,603ha.
These fIres burned within threeareasthat have been
identified as Late-SuccesionalReserves(USFS 1994and
1995). Theseareashave management
objectivesto provide habitat for late- successionalassociatedspecies,including the Threatened northern spotted owl (Strix
occidentaliscaurina). Agee and Edmunds(1992)stated
This studywasconductedon theLeavenworthRanger
District. WenatcheeNational Forest, located on the east
sideof the Cascadesof Washingtonstate(Figure 1). Elevationsrangefrom 650 metersto 1300meters. The study
area is composedof ponderosapine (Pinus ponderosa)
plant associationsat lower elevations,and Douglas-fIr
(Pseudotsuga
menziesil)
andgrandfir (Abiesgrandis)plant
associationsat the mid-elevations.
Disturbances,suchas fIre, have beenshownto havea
significantinfluence on the vegetationpatternsand processesof eastCascadelandscapes(Gastet al. 1991,Agee
1994,Johnsonet al. 1994,Harrod et al. 1996). Prior to
fIre suppression,
fire occurredat relatively frequentintervals within east~
forests(AgeeandEdmunds1992).
For example,within the JX>nderosa
pine and dry Douglas
fIr plant seriesfire occurredat intervalsof 10 to 25 years
resulting in a foreststructure largely compnsedof open
park-like ponderosapine forests(Agee 1991 and 199.4).
Fires in theseforestswere usuallyof low to moderateffitensity.
---
.
124
Gaines.W1 Sb'and.R.A..andPipeJ',
S.D.
--""'"
;J -,.o\~(
-"", ,
"
;_:~-
-1
T~0
.1-
t
~-
e
!
'""~
--u-
0
-
~..
-_.
20-
~-"""~
Figure 1. Vicinity mapof the Ratand HatcheryFires.
-Effects of the Hatchery Complex Fires on Nothern Spotted Owls -125
Fire suppression and logging have signficantly altered
these forests. With the advent of effective fire supression
technologies, fIre frequency within these forests was significantly reduced (Agee and Edmunds 1992). In addition, logging practices until about 1950 often focused on
removing the most fire tolerant species,the large ponderosa pine (Wellner 1984). The end result is that eastside
forests are now composed of more shadetolerant tree species, are less fIre tolerant, are at higher densities, and are
more prone to large scale high intensity fIres (Agee and
Edmunds 1992, Camp 1995).
Data Analysis
A linear correlation analysis (Zar 1984) was completed
to determine if there was a relationship between site status and the degree to which habitat was affected by the
fIres within a 0.8 kIn and 2.9 kIn radius. A Chi-Square
Goodness of Fit test (Zar 1984) was used to determine if
Metbods
there was a significant difference between habitat burned
vs habitat available at several lOCationson the burned landscape.
Habitat Inventory
Inventories of spotted owl habitat were completed
from aerial photo-interpretation followed by field verification. These data were then loaded into MOSS geographic information system for habitat analyses. Habitat
inventories were completed prior to the fITe,immediately
post-fIre, and one year after the fITesoccurred in order to
measure habitat changes over time. Suitable habitat was
defined as having at least a 60% canopy closure presence
of numerous snagsand several down logs, and two or more
canopy layers. The habitat mapping resolution was 0.8
ber of young. Surveys and site monitoring were accomplished through a cooperative effort between the U.S.
Forest Service, National Council on Air and Stream Improvement, and Natapoc Resources.
Persistence a/Spotted Owl Nest Sites
Camp (1995) presents a probability model for the development of late-successional fIre refugia under an in.
herent disturbance regime for the eastern Cascade mountains. Because spotted owl nesting habitat has been shown
to be associated with these late- successional conditions
(Buchanan et aI. 1993, Buchanan et aI. 1995) this model
was used to show the potential for a fIre under an inberent disturbance regime, at the 28 known nest sites on the
Leavenworth Ranger District.
M.
Once habitat maps were developed anadditional analysis was condocted to describe landscape features for habitat that was within the fITe perimeter but remained suitable following the fITe. This analysis was completed for
all habitat within a 2.9 km radius around three of the six
fITe affected activity centers. A spotted owl activity center is the location in which there was a resident single,
pair or nest site (USFS 1991). These three spotted owl
activity centers were selected because they were distributed across the burned landscape, they were the most extensively affected, and there was no overlap in habitat
within the 2.9 km radius. This information was developed to determine if habitat at particular landscapelocations burned at proportions equal to, greater than or less
than was available within the three activity centers. Habitat was classified into one of the following categories:
north aspect (>270 to 89 degrees), south aspect (>90 to
270 degrees), riparian, v~ey (>300m wide), bench «10%
slope and >6 ha), and mid-upper slope.
Spotted Owl Inventories
The Region 6 spotted owl survey protocol was followed (USFS 1991). During 1995, spotted owl activity
centers were surveyed regardless of how extensive the
habitat was changed as a result of the fIres. Sites were
monitored to determine if spotted owls were present, and
if so, their status: single, pair, or reproductive pair. If
ed
be reprod uctIve,
. adA:
spotted owls were determm
to
wtional monitoring was conducted to determine the num-
.
.
Results and D&ussion
Habitat Availability
The availability of suitable spotted owl habitat within
a 2.9 kIn radius of the six affected activity centers prior
to the fIre, immediately following the fIre and one year
post-fIre are shown in Table 1. The average amount of
habitat loss that occurred within a 2.9 kIn radius was 31 :0
and ranged from 8% to 45%. The average amount ofhabltat lost within a 2.9 kIn radius as of one year after the fIre
was 55% and ranged from 10% to 85%.
...
These data show that a considerable reductIon m habltat occurred as a result of the direct effects of the fIre,
however, additional habitat reductions occurred well after the fIres were out. This occurred because trees darnaged but not directly killed by the fires eventually died.
Table 1. Habitatavailabilitypre-fire,IX>st-flre,
andoneyearIX>stfire within a2.9 km radius of sponooowl activity centers(n=6),
HatcheryComplexFires.
SileNo.
I
2
3
4
5
6
~-fire
(Ita)
691
493
512
604
618
377
Post-fire
(Ita)
463
454
399
344
420
208
1 Yr Post-fire
(Ita)
342
446
114
3~;
171
.
126
Gaines,W L., Sb'and,R.A., andPiper, S.D.
In addition, some b"eeSthat survived the flTeShave been
killed by increasedinsectactivity. It is expectedthatthis
nend will continue as insectpopulationsincrease,however the extent is not yetknown.
The resultsof the analysisto detennineif suitablehabitat at various locations on the landscapeburned in proportion to its availability are shown in Figures 2 and 3.
Habitatlocated on north aspectsand within riparian areas
burnedless than available, thoughnot statisticallydifferent(p>O.O5). Habitat on north slopeslocatedon a bench
burned at proportions equal to it's availability. Habitat
on north slopeson the mid-upperpart of the slopeburned
at levels slightly greaterthan available, but not statistically different than expected(p>O.O5).On southslopes
habitat located on a benchburnedat levels that were significantly less than expected(p<O.O5). Habitat located
on southslopesand onthe mid-upperportion of the slope
burned at proportions greaterthan available, but not at
levels significantly different than expected(p>O.O5).
SiteStatus
The resultsof the spottedowl surveys at the six flIe
affectedactivity centersare shownin Table 2. Four of
the six siteswere not occupiedduring 1995. At one of
thesesitesthe flfe ovenook the activity centervery rapidly and at extremelyhighintensity. It is likely that these
owls did not survive the flfe, and their octivity centeris
likely no longercapableof supportingan owl pair. Habitat at an additional site wasreducedto levels that make it
unlikely that it could supportspottedowls. This sitewas
not occupiedin 1995.Fewerof theseactivity centerswere
occupied in 1995 than at any time during the previous
four years. In addition, only one site was reproductive,
also the lowestlevel comparedto the previous four years.
Site Statusand Habitat Availability
There was no correlation betweenthe habitat available within a 2.9 km radiusone yearpost-fire and the site
status(alpha=O.O5,
p>O.O5).Therewas, however,a cor-
r.
00
so
4
3
1~%Bumed I
10 % Available!
2
I
Riparian
VaJley
Bench
Mid-Upper
Landscape Feature
Figure 2. Proportion of spon~ owl habitat that burned vs that available at various locations on north astx:Cts,Hatchery Complex ftreS.~
.
-Effects of the Hatchery Complex Fires on Nothern Spotted Owls -127
30
25
20
15
I~% Burned!
l.o %Available!
I
5
Riparian
Valley
Bench
Mid-Upper
Landscape Feature
Figure 3. Proportionof SJX>tted
owl habitatthat burnedvs that availableat variouslocationson southaspects,
HatcheryComplexfires.
Table 2. Statusof SJX>tted
owl activity centersfor four yearsprior
to the Hatchery Complex fires and oneyearfollowing.
Stanu
1991
1992
1993
1994
1995
However, reproduction during 1994 was also quite low,
and as shown in Table 3, 1995 was overall a low year for
reproduction. Reproduction,.at leastone yearpost-f]1'ewas .
not reduced much below preVIously reported low years.
No~mt
5~
Single
Pair
Reproductive 5~
No. sites
4
25%
25%
5~
4
2Mo
2Mo
60'10
5
17%
33%
17%
33%
66%
6
6
17%
17%
Spotted Owl Nest Sites and Inherent Fire Regime
A total of 28 nest sites on the Leavenworth Ranger
District were evaluated using the fife refugia model developed by Camp (1995). The results of this evaluation
are shown in Figure 4. Two (7%) of the nest sites were in
relation between site status and the amount of habitat reroaming after the fire within a 0.8 mile radius (alpha=O.05,
Table 3. Comparisonof II young/site at spotted owl activity
centersaffactedand not by the HatcheryComplexfires.
p<O.?5). As the availability o~ habitat increased so did
the sIte status (no presence bemg low status and repro-
Sile Data
ductive pair being the highest status).
Affcctedby fires
CN
young/site)
Reproduction
Data from all owl sites on the Leavenworth Ranger
District were used to compare reproduction at burned vs
unburned sites. These data are shown in Table 3. The
1995 season (one year post burn) was lower than any of
the previous four years at the sites affected by the fifes.
No. sites
1991
1992
1993
1994
1995-
0.6
1
0.8
0.3
0.2
5
5
6
6
6
Not affected
fy fires
CN
young/site)
0.9
1.3
.03
1.8
0.3
No. sites
1315
15
14
17
~
.;.:4":;,
128
Gaines,W.L., SU'and,
R.A., andPiper, S.D.
locations wheretherewas only a 2% probability of them
remaining as fire refugia underan inherent fire regime.
Fourteen(50%) of the sites occurredat a position on the
landscapein which therewas a 10% probability of them
remaining as fire refugia under an inherentdismrbance
regime. Eleven (39%) of the nestsitesoccurredat locations in which there was a 19% probability, and 1 (4%)
was at a location in which therewas a 51% probability of
them remaining as fire refugia. This assessment
exarnplifies the dynamic nature of spottedowl habitat
acrossa landscapein which fire played its inherentrole.
This is especiallytrue on forests on the eastside of the
Cascadeswhere relatively dry conditions resultedin frequentfifes (Agee 1994).
Conclusions
Becausedatafrom only one seasonhasbeencollected
up to this point, conclusionsmustbe maevery cautiously.
Additionalmonitoring of thesesiteswill occurin succeeding yearsand datamay supponthe fmdings madein this
initial evaluationor it may provide new insights. However, at this point the following observationsabout the
effectsof the HatcheryComplexfires on spottedowls and
their habitathavebeenmade.
The fifes reducedthe availability of spottedowl habitat around six activity centers. The initial direct reduction causedby the fife was not an accuratereflection of
the total affectson habitatavailability. Fire damagedtrees
continuedto die,andincreasedinsectactivity contributed
to the total reductionin spotted owl habitat within the
burnedlandscape.
Habitatlocatedwithin riparian areasand on a bench
is somewhatmore likely to remainas suitable following
a fife vs habitat locatedon the mid-upper slopes,especially on southslopes.This information shouldbe useful
in the developmentof managementplans for habitat reserves.
In the six fife affectedactivity centerstherewasa decreasein the numberof reproductivepairs andan increase
in thenumberof sitesnotoccupied.Direct mortality likely
occurredat one of thesitesasa result of rapid and intense
fife.
S
1mNo. of nest sites
0 % of nest sites
2%
10%
19%
51%
Probability of Becoming fire Refugia
Figure 4. Persistence of sjX>ttedowl nest sites (n=28) using the Fire Refugia model (Camp 1995). Hatchery Complex fires.~
.
..'""'i.-_;_,,~Si;;A!':';'
-Effects
of the Hatchery
Complex
The availability
of spotted owl habitat across the eastem Cascades
landscape
appears to have been very dynarnic
under
inherent
disturbances
(Camp
1995).
This
is
an
sb"ategies
important for consideration
habitat
reserves.
when developing
The flIe refugia
management
model
d
C
eveloped
by
~dy
s~ould
(1995)
be usef~.m
m the hIghest
.,..
...ow
amp
and
on Nothern
u.s.
from
strategies
of sustammg
habitat
reserves
ters (Agee
fuel
fire
fuel
loading
1992).
areas could
of adjacent
climax
and
may be neccessary
and Edmunds
density
tection
of
Wellner,
owl habl-
ponderosa
to protect
owl
habitat
pro-
and restoration
of
may
imp~t
Servlce,
. P 0 rtl an d, Or .
1984. History
proiX>Sed
nonhem
SIXI~
and status of silviculblral
Douglas-fir
manage-
and grand flI forest types.
glas-flI and grand flI forest types. Wa. State Univ. Coop. Ext.,
low
dual objectives:
that
for surveymg
Pages 3-10 in D.M. Baumgarter aro R. Mitchell, Eds. SilviculblIal management strategies for pests of the interior Dou-
cen-
located
C.A.
F
ment in the interior
within
activity
Strategically
accomplish
spotted
tree density
-129
1991. Gwdelines
activities
ls .orest
USDA
tal
Management
Owls
of the northern spotted owL USDA Forest Service, Portland.
U.s.Or.Forest Service.
thIS
that resu~t
spotted
Spotted
Forest Service. 1994. Record of Decision for the managementoflate-successional
associated species within the range
.management
mlonnauon
developl.n~
probabIlity
Flres
Pullman,
Zar.
J.H.
Wa.
1984. ~iostatistical
Englewood
Analysis.
Prentice-Hall,
Cliffs, N.J.
pine.
References
Agee.
J.K. 1994.
~systems
GI'R-320.
Fire
IIxi
weather
disturbances
in terresrrial
of the Eastern Cascades. Gen. Tech. Rep. PNWUSDA Forest Service, PNW Station. Portland, Or.
52pages.
Agee. J.K.,1Ixi R.L. Edmunds. 1992. Forest protection guidelines
fur the northern S}X)tted owL Pages 181-244 in U.S. Dept. of
Jntaior,
R~very
plan for the northern
spotted owl-fmal
draft. Vol. 2 U.S. Govt. Print. Off.. Washington. D.C.
Agee. J. K. 1991. Fire histOry of Doug!as- fir forests in the Pacific
Northwest. Gen. Tech. Rep. PNW -GTR-285. USDA Forest
Service, PNW Research Station. Portland, Or.
Budlanan. J.B.. L.L. Irwin. and E.L. McCutchen. 1995. WithinItaI¥i nest site selection by SIXltted owls in the eastern Washington Cascades. J. Wildt Manage. 59(2):301-310.
Bucl1anan, J.B., L.L. Irwin, and E.L. McCutchen. 1993. Char~teristics
~,
of spotted owl nest trees in the Wenatchee National
Forest. J. Ra~
Res. 27(1):1-7,
A.E. 1995. Predicting late-successional
physiography
and topography.
~
fire refugia from
Dissertation.
Seattle, W L
Gist. W.R.. D. W. Scott, C. Schmitt, D.Clemens,
S. Howes, C.G.
Johnson. Jr., 8I¥i others. 1991. Blue Mountains
report:
new perspectives
Univ. Wa.,
in forest heaTh. USDA
forest health
Forest Ser-
vice.
Harrod,
R.J., W.L.
Gaines,
Biodiversity
in the Blue
R.I. Taylor, and others. 1996.
Mountains.
In R. Iaindl and T.
Quigley. Eds. Search for a solution -sustaining
the land,
people and ~nomyofthe
Blue Mountains, a synthesis of our
knowledge.
USDA
Forest Service, Blue Mountains
Resources Institute, LaGrande, Or.
Johnson. C.G.. R.R. Clausnitzer, P.J.Mcl1ringer.
1994. Biotic
Gen. Tech.
aJ¥i abiotic
Rep.
guidelines
Eastan
PNW -GTR-322.
for the Wenatchee
Washington
andC.P .Oliver.
processes of eastside ecosystems.
USDA
PNW Research Station. 50 pages.
U.s. Forest Service. 1995. Late-Successional
IIxi
Natural
Forest Service,
Reserve: Standards
National
Forest in the
Cascades IIxi Yakima Provinces. USDA
Forest Service, Wenatchee
National
Forest, Wenatchee, Wa.
.
Inc.,
Proceedings: Ar~~nference
on
T10 NA L ASSOC IAn 0 N
...
Washington
Foundation for
the Environment
r"
D~:D ARE
,
,
Proceedings: First Conference on
Fire Effects on Rare and Endangered
Species and Habitats
Coeur d' Alene, Idaho
November, 1995
Dr. Jason M. Greenlee
Editor
Conference
organizer:
ASSOC
IAn
0.
OF
lAND ARE
Co-sponsors:
International Association of Wildland Fire
Wildlife Forever
Washington Foundation for the Environment
Dr. Jon Keeley
Colorado Natural Heritage Program
The Nature ConserVancy
U.S. Fish and Wildlife Service. NIFC
Abundant Ufe Seed Foundation
Montana Prescribed Fire Services, Inc.
Environmental & Botanical Consultants
USDA Forest Service (Reg-G)
Proceedings courtesy of:
~~
&
.Dedicated
Ildhfe
Forever
Washington Foundation
for the Environment to preserving and enhancing our state's
environmental heritage by supporting educational
and Innovative projects
Download