9/24/11 Partnering with Beaver in RestoraDon Design John Stella, Ph.D. State University of New York, Syracuse Environmental Science and Forestry (SUNY-­‐ESF) September 19 -­‐ 21, 2011 A workshop offered through the Intermountain Center for River RestoraDon and RehabilitaDon at Utah State University Key Concepts Life history niche Worldwide distribuDon Beaver diet and food preference The colony, the lodge, and the cache Site establishment and habitat successional cycles •  Beaver as ecosystem engineer vs. nuisance •
•
•
•
•
John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design Beaver Life History & Biology Photo by Michael S. Quinton, Na5onal Geographic Society John Stella, Ph.D. Anna Harrison, M.S. State University of New York, Syracuse Environmental Science and Forestry (SUNY-­‐ESF) Life History Niche •  Largest rodent in North America (up to 90 lbs!) •  Ubiquitous within N. hemisphere temperate ecosystems •  Physiographic se]ng ranges from boreal to aridlands •  Habitat generalist; highly adaptable •  Common habitat ingredients: water + wood –  Northern tundra and treeline range boundary: wood limitaDon –  Southern range desert boundary: perennial streamflow and/or wood limitaDon 1 9/24/11 2
2
POLLOCK ET AL.
POLLOCK ET AL.
Photo: C. Demers, SUNY-­‐ESF Worldwide distribuDon of beaver •  Castor canadensis (N. America) •  Historically, 60–400 million pre-­‐
European sedlement (Seton 1929) •  Currently, 6-­‐12 million (Naiman et al. 1988), but esDmates are crude •  SpaDal distribuDon approaches its historical range •  C. fiber (Eurasian beaver) •  More limited current distribuDon, but expanding back to parts of its historical range. A
A
B
B
Pollock MM, Heim M and Werner D. 2003. Hydrologic and geomorphic effects of beaver dams and their influence on fishes. FIGURE 1. Estimated current and historic distribution of beaver in North America (A) and Eurasia (B). Isolated
populations
in peninsular
Florida
and Southern
California
are not
In Eurasia,
hatching
delineates
FIGURE 1. Estimated
current
and historic
distribution
of beaver
in shown.
North America
(A)cross
and Eurasia
(B).
Isolated
current distribution.
In North
America,
the current
and historic
are approximately
coincidental.
populations
in peninsular
Florida
and Southern
California
are notdistributions
shown. In Eurasia,
cross hatching
delineates
(Based on
Jenkins 1979;
HalleyAmerica,
and Rosell
MacDonald
et al.
1995 ).
current
distribution.
In North
the2002;
current
and historic
distributions
are approximately coincidental.
(Based on Jenkins 1979; Halley and Rosell 2002; MacDonald et al. 1995 ).
Woody Food ConsideraDons •  Maximizes energy intake with low costs •  Easy digesDbility; short gut retenDon Dme •  Avoid bad-­‐tasDng secondary compounds •  Willows, aspen most commonly preferred; conifers avoided Beaver Diet: “A choosy generalist herbivore” herbaceous plants, incl. aquaDc •  Spring/Summer: and riparian forbs, grasses, grains and row crops •  Fall/Winter: tubers, bark and cambium of cached woody plants •  Woody plants comprise 86% of winter diet; 16% of summer diet (Roberts and Arner 1984) •  Number of woody species consumed range from 3 at northern range limit to >30 in southern region (Aleksiuk 1970, Hill 1982, Novak 1987) The Colony Photo by Michael S. Quinton, Na5onal Geographic Society •  Colony unit = 6−8 related individuals •  Avg. liders = 2−5 kits •  Young stay with parents at least 2 years •  Adults (>2 yrs) disperse to establish new lodge •  Territories marked with scent mounds •  Home ranges tend to follow shorelines Photo by Anna M. Harrison John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 2 9/24/11 The Lodge and Food Cache •  AcDve lodges indicated by fresh food cache in fall •  AcDve lodges spaced at least 0.5−1 km apart •  Colony saturaDon densiDes vary with landscape and region •  Max. density ranges 0.5−5 colonies/km2 acDve lodges (Adirondack Mtns., NY) lodge Photo by Anna M. Harrison cache (Hill 1976, Novak 1987, Baker and Hill 2003) LocaDon, locaDon, locaDon….. •  Bank dens vs. aquaDc lodges •  Food caches can be submerged or exposed Bank den (Colorado Natural Heritage Program) Mid-­‐stream lodge in Hinsdale County, CO (Colorado Natural Heritage Program) AquaDc Habitat is CriDcal to their Success •  Beaver more agile in water than on land; maximize Dme in the water •  Ponds provide cover from predators and foraging pathways •  Lodge includes underwater entrance, nest area above water Photo by Anna M. Harrison Dams •  Created to impound water around lodge •  Dam locaDon cued by running water •  Dams constructed of wood and available debris (e.g., plasDc, metal) •  Where palatable species are rare, conifers are used more in dams, with hardwoods saved for the food cache (Barnes and Mallik 1996) Mid-­‐lake lodge John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 3 9/24/11 World’s largest beaver dam Dam/Pond Complexes Images courtesy of EcoinformaDcs, Inc. •  Found in Alberta, Canada (2007) using Google Earth •  850 m; longer than Hoover Dam! •  MulDple dams create safe transportaDon corridors to connect large ponds •  Dams complexes grow over Dme, allowing beaver more access to food sources Photo: G.S. Haulton Photo by Anna M. Harrison Even some unlikely places… In what ecosystems do we find them? •  Lakes •  Rivers and streams •  Abandoned channels on floodplains •  Wetlands •  Estuaries •  Glacier outwash streams Pierre Côté Mendenhall Glacier, AK (Photo Bob Armstrong) California Academy of Sciences Beaver Dam Creek, Long Island, NY John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 4 9/24/11 Beaver impacts: landscape conversion •  Beaver change landscape from terrestrial to aquaDc •  Most landscape change occurs in first 20 years Beaver impacts: increase wetland area •  Increased landscape diversity –  Wright et al. 2002 •  Waterfowl habitat •  Increased amphibian habitat –  Karraker and Gibbs 2009 Photo by Anna M. Harrison 18 Pond New Pond Beaver impacts: Possible beaver pond succession post abandonment Beaver impacts: forest structure •  Removal of understory and canopy trees •  Open up canopy to understory/unpalatable species Stream Meadow Old Pond Naiman 1988 Photo by Anna M. Harrison John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 5 9/24/11 A local (to NY setate) example Ecosystem ngineer that foiled our research crew Ecosystem Engineer, or Nuisance? •  Unintended beaver damage includes Dmber, crops, ornamental plants, and even buildings •  Common management concerns include culvert blockage, road flooding, dam removal and beaver control structures Photo: Adirondack Ecological Center due to a road culvert dam… then removed… 23 Photo: Adirondack Ecological Center John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design Photo: Adirondack Ecological Center 6 9/24/11 Ecosystem engineer Acknowledgements
•  Stacy McNulty, Adirondack
Ecological Center
•  Field support: Joshua
Cousins, Bill Dunker
•  SUNY-ESF McIntire Stennis
Grant Program
For more informaDon: stella@esf.edu Photo: Adirondack Ecological Center … and now for<fied with a ‘beaver deceiver’ References Aleksiuk, M. 1970b. The seasonal food regime of arcDc beaver. Ecology 51:264– 70. Barnes, D. M., and A. U. Mallik. 1997. Habitat factors influencing beaver dam establishment in a northern Ontario watershed. Journal of Wildlife Management 61:1371–77. Baker, B. W., and E. P. Hill. 2003. Beaver (Castor canadensis). Pages 288-­‐310 in G. A. Feldhamer, B. C. Thompson, and J. A. Chapman, editors. Wild Mammals of North America: Biology, Management, and ConservaDon. Second EdiDon. The Johns Hopkins University Press, BalDmore, Maryland, USA. Hill, E. P. 1976. Control methods for nuisance beaver in the southeastern United States. Pages 85–98 in Proceedings of the seventh vertebrate pest control conference. Karraker, N.E., and J.P. Gibbs. 2009. Amphibian producDon in forested landscapes in relaDon to wetland hydroperiod: a case study of vernal pools and beaver ponds. Biological ConservaDon 142:2293-­‐2302. Naiman, R. J., C. A. Johnston, and J. C. Kelley. 1988. AlteraDon of North American streams by beaver. Bioscience 38:753–761. Novak, M. 1987. Beaver. Pages 283–312 in
M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch, eds. Wild furbearer management and conservaDon in North America. Ontario Trappers AssociaDon and Ontario Ministry of Natural Resources. Roberts, T. H., and D. H. Arner. 1984. Food habits of beaver in east-­‐central Mississippi. Journal of Wildlife Management 48:1414–19. Seton, E. T. 1929. Lives of game animals, Vol. 4, Part 2, Rodents, etc. Doubleday, Doran, Garden City, NY. Wright, J. P., C. G. Jones, and A. S. Flecker. 2002. An ecosystem engineer, the beaver, increases species richness at the landscape scale. Oecologia 132:96-­‐101. John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 26
Landscape Influences on Long-­‐Term Beaver Occupancy John Stella, Ph.D. Anna Harrison, M.S. State University of New York, Syracuse Environmental Science and Forestry (SUNY-­‐ESF) 7 Location of Dams in
Watersheds
•
Key features for dam building (Novak, 1987): 1.  Stream gradient 2.  Cross secDonal area •
•
Pond will persist unDl dam abandonment or failure Pond succession depends on beaver acDvity and landscape characterisDcs, especially hydrology (Naiman et al. 1988) 9/24/11 Factor #1: stream gradient <10% (however, this varies greatly by region) Amount of surveyed habitat currently or formerly
used (%)
The Ecosystem Engineer’s Design SpecificaDons Beaver prefer to dam small, low-gradient streams
with unconfined valleys, but they can also dam
both large and high-gradient streams. Retzer et
al. (1956) studied 365 reaches in 61 streams
throughout Colorado to determine the physical
factors determining beaver pond location. Beaver built dams on 82% of all the low-gradient (1–
3%) streams surveyed, 73% of reaches with 4–6%
gradients, and 61% of reaches with 7–9% gradients (Figure 2). Use of streams with a slope greater
than 9% dropped off dramatically. On streams with
gradients greater than 15%, just one active dam
was found on a 16% stream slope and one abandoned dam was found on a 21% stream slope.
unconfined (>4 channel widths) valleys, and almost all of them were in watersheds less than 15
km2. In this study, 16% was the steepest gradient
where a beaver pond was found. Similarly, Suzuki
and McComb (1998) studied 170 beaver dams in
the Drift Creek basin, Oregon and consistent with
these results found that more than 90% were on
stream gradients of less than 6%. Beier and Barrett
(1987) also found that geomorphic and hydrologic
conditions were the best predictors of dam-site
suitability, with gradient, stream depth, and stream
width the most important factors. They found that
biological factors, such as food availability, did not
provide additional explanatory power as to the
location of dam sites.
The maximum size of streams that beaver
can dam is not well documented, and it will certainly vary from region to region, depending on
hydrologic conditions. In Washington, historic
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1-3
4-6
7-9
10-12
13-15
16-18
19-21
21
FIGURE 2. Beaver–dam frequency based on stream gradient from a survey of 356 stream reaches in the
mountainous regions of Colorado, showing that beaver strongly prefer to dam low-gradient streams (adapted from
Retzer et al. 1956).
Photo: Anna M. Harrison Photo: Anna M. Harrison Factor #2: woody cover Beaver dam presence inversely related to proporDon of open canopy area and stream gradient. Curtis & Jensen (2004), beaver dam activity along roadsides in NY state.
(Journal of Wildlife Management 68:278-287)
John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design PopulaDon trends based on aerial photos and aquaDc patch creaDon 1990
PATCH CREATIONBY BEAVERS
August
10
•  Beavers first select sites that create the -c
8largest ponds 0
•  Most landscape 6 -i
change occurs during C,)
the 20 years a{er 40t0
iniDal occupancy AQUATIC
I
decades (Fig. 1). The 1948 pond
idly during the first decade aft
increase in average pond-site ar
** **I
significant differences among th
|/I
T
1948, 1961, and 1972 age classe
T
1
i
pond-site area increases in all
0)
and small, ranging from 0.1 to
The combination of high rates
V
/
C
er initial pond area, and rapid g
0
1948, and 1961 pond cohorts th
V
2
,
,
ential (Fig. 2). As of 1986, the
75% of the total number of pon
total area impounded. The estab
0
NEW
10
20
30
40
50
was the primary cause of increa
site area prior to 1961 (70% of
Approximate Pond Site Age (year)
(Johnston and Naiman 1990 BioScience 38:753–761) FIG. 1. Averagepond-sitearea,by pond cohort.* = 1940 rest being due to the enlargem
pond cohort (y = 1.27 ln(x) + 3.72), A = 1948 pond cohort The proportions were reversed
(y= 0.63 ln(x) + 3.85), E= 1961 pond cohort(y = 0.18 ln(x) ponds constructed by beaver du
+ 3.16), V = 1972 pond cohort (y = 0.14 ln(x) + 1.62). * cades of
occupancy have the g
and ** indicatesignificanteffectsof pond-siteage on average
pond area (Kruskal-Wallisone-wayANOVA) at .05 and .01 landscape.
significancelevels.
Pond establishment relat
beaver popula
Although beaver were conti
RESULTS
ponds throughout the time per
Patch characteristics
significant differences among ti
All sites that had been impounded by beaver during of new-pond creation (Table 2)
the 46-yr period were still clearly distinguishable on
between 1940 and 1961, when
the 1986 photos. Although some areas had been briefly created at an average rate of 25
***
0
8 9/24/11 MoDvaDng QuesDon: What landscape factors maintain suitable habitat for beaver over the long-­‐term? Test system: Adirondack Mtns, NY State PotenDal site influences on long-­‐term occupancy Maintenance Cost Expected RelaDonships Dam Volume Beaver Occupancy Dura<on Number of Dams Cross Valley Slope PosiDve NegaDve Resource Quality & Quan<ty Hardwood basal area Landscape Capacity So{wood basal area Down Valley Slope Forage Area Pond Area Total basal area (Harrison &. Stella 2010) Not for distribu5on s1
•
•
30-­‐year beaver occupancy dataset (HunDngton Wildlife Forest, NY State) 14 sites straDfied by occupancy rate: 1.  Located on stream reach or wetland 2.  Open-­‐canopy wetlands 3.  Evidence of dam 4.  One acDve colony only 5.  Sites are discrete land areas Sites varied
in occupancy
duration from
3% to 100%
of years
surveyed
(Contact J. Stella, stella@esf.edu) John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design (Harrison & Stella 2010) Percent
39 s2
s3
s4
s5
s6
s7
s8
s9
s10 s11 s12 s13 s14
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Total
(Harrison & Stella 2010) Not for distribu5on 38 (Please contact J. Stella, stella@esf.edu) Photo: Anna M. Harrison 30 25 19 14 13 13 12 10
9
6
5
3
100.0 83.3 63.3 46.7 43.3 43.3 40.0 33.3 30.0 20.0 16.7 10.0
Not for distribu5on 1
3.3
1
3.3
40
(Please contact J. Stella, stella@esf.edu) 9 9/24/11 Field Sampling 3 •  Forage Area •  Pond Area •  Dams and Dam Volume •  Upland Forest VegetaDon –  5-­‐m wide belt transects –  Full inventory of tree community Landscape capacity was an important predictor of occupancy duraDon 9
12 4
(Harrison & Stella 2010) Not for distribu5on (Harrison &. Stella 2010) Not for distribu5on 41 (Contact J. Stella, stella@esf.edu) 42 (Please contact J. Stella, stella@esf.edu) Best logistic regression models
predicting occupancy duration included
all three categories of predictors
Food quanDty and quality were also important predictors Variables in Model with Odds Ratios
Landscape
Capacity
QAIC QAIC
Rank
c
diff
Not for distribu5on (Harrison & Stella 2010) (Please contact J. Stella, stella@esf.edu) John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 43 w
i
Dam
Vol
HW BA
SW Total
BA
BA
AUC
1
26.2
0.0
0.13
0.774
2
27.2
1.01
0.08
0.772
3
27.2
1.03
0.08
0.770
4
27.4
1.27
0.07
0.744
5
28.1
1.91
0.05
0.778
Null
Food Quantity
and Quality
Maintenance Cost
Pond Forage
Slope
Slope Dams
Area
Area
x-valley
50.7 24.91 <0.01 -----
0.259
-0.004 0.175
(1.30)
(1.00)
(1.19)
0.148
0.189
(1.16)
(1.21)
0.167
0.196
(1.18)
(1.22)
0.360
-0.005
(1.43)
(1.00)
0.315
(1.37)
Not for distribu5on -0.005
0.061
(0.99)
(1.06)
(Harrison & Stella 2010) 44
(Please contact J. Stella, stella@esf.edu) 10 9/24/11 Occupancy Model Summary Maintenance Cost Dam VVolume
olume Dam Number of Dams Cross Cross V
Valley alley SSlope lope Beaver Beaver Occupancy Occupancy 30 year dataset Pond Area PosiDve NegaDve Resource Quality & Quan<ty Total basal area Landscape Capacity Down Valley Slope Highly Significant Hardwood Hardwood asal aarea rea Hardwood bbasal Forage Forage Area
Area So{wood basal area (Colorado Natural Heritage Program) What predicts beaver occupancy in a landscape with limited food and structural wood? (Harrison & Stella 2010) Not for distribu5on Not for distribu5on (Please contact J. Stella, stella@esf.edu) (Please contact J. Stella, stella@esf.edu) References Acknowledgements
•  Stacy McNulty, Dr. Jacqueline
Frair, Dr. Jake Bendix
•  Drs. Martin Dovciak, Ruth
Yanai, Steve Stehman, and
Carol Johnston
•  Adirondack Ecological Center
•  Field support: Joshua
Cousins, Bill Dunker, Michael
Silberman & James Johnson
CurDs, P. D. and P. G. Jensen. 2004. Habitat features affecDng beaver occupancy along roadsides in New York state. Journal of Wildlife Management 68:278-­‐287. Harrison, A.M., J.C. Stella. 2010. Engineering the forest ecosystem: impacts on woody vegetaDon structure and composiDon by beaver, a central place forager. MeeDng of the Ecological Society of America, Pidsburgh, PA, August 2010. Johnston, C. A. and R. J. Naiman. 1990. AquaDc patch creaDon in relaDon to beaver populaDon trends. Ecology 71:1617-­‐1621. Naiman, R. J., C. A. Johnston, and J. C. Kelley. 1988. AlteraDon of North American streams by beaver. BioScience 38:753–761. Pollock, M. M., M. Heim, and D. Werner. 2003. Hydrologic and geomorphic effects of beaver dams and their influence on fishes. American Fisheries Society Symposium. 213-­‐233. Funding:
•  SUNY-ESF McIntire Stennis
Grant Program
Retzer, J. L., H. M. Swope, J. D. Remington, and W. H. Rutherford. 1956. Suitability of physical factors for beaver management in the Rocky Mountains of Colorado. State of Colorado, Department of Game and Fish Technical BulleDn No. 2, Denver. For more informaDon: stella@esf.edu 47
John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 11 9/24/11 Riparian vegetaDon diversity & complexity Riparian Ecology and Feedbacks of Beaver John Stella, Ph.D. Anna Harrison, M.S. State University of New York, Syracuse Environmental Science and Forestry (SUNY-­‐ESF) Photograph by Kit Cheryl Reynolds Vegetation Mosaic, Merced R.
Herbaceous cover
Cottonwood forest
Mixed riparian forest
Valley oak forest
High-­‐connecDvity corridors for dispersal Longitudinal Complexity: Linear string of alternaDng reach types (constrained vs. open) Photo: www.douglascountywa.net S. Badger River, Eastern Washington Image: Naiman et al. 2005, Riparia John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 12 9/24/11 Lateral zonaDon of plant communiDes (with geomorphic surfaces) High habitat complexity (e.g., acDve & abandoned channels, wetlands, sloughs, springs, ridges & swales) Image: Naiman et al. 2005, Riparia Image: Naiman et al. 2005, Riparia Functional Adaptations of Riparian Plants
(sensu Naiman et al. 2005)
•  Invader - Produces large numbers of wind
and water dispersed propagules that
colonize alluvial substrates
•  Endurer - Resprouts after breakage or
burial after floods, and herbivory
Sitka spruce
Mayfly
Willow
•
Resister - Withstands prolonged flooding,
moderate fires and/or disease
•
Avoider - Lacks adaptations to specific
disturbance types. Life cycle often
completed between disturbance events.
(E.g., annual plants, stream insects)
John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design Adaptations of wetland plants
to anoxic environments
Aerenchyma
(hollow tissue for conducting
air to roots)
Lenticels
(stem pores to allow
direct gas exchange)
Images: Naiman et al. 2005, Riparia 13 9/24/11 Willows and Poplars in Southwestern Riparian Zones Keeton, W. S., C. E. Kra[, and D. R. Warren. 2007. Riparian Forest Influence on Stream Habitats (Adirondacks, NY) •  Sexual reproducDon Dming coincides with regular spring floods •  Seedlings need to establish in high-­‐flow years •  But vigorous sprouters •  Highly palatable to beaver Photo: John C. Stella Live and dead biomass increase with forest age Keeton, Kra{, and Warren, 2007. Ecological ApplicaDons 17:852-­‐868. Keeton, W. S., C. E. Kra[, and D. R. Warren. 2007. Forest structure influences in-­‐stream habitat •  Stream LWD correlates posiDvely with forest basal area. Keeton, W. S., C. E. Kra[, and D. R. Warren. 2007. •  Number of debris dams and pools was correlated with number of large logs in the system. Keeton, W. S., C. E. Kra[, and D. R. Warren. 2007. John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 14 9/24/11 A large terrestrial footprint for an aquaDc species! Ecological Feedbacks of Beaver •  Shallower water table •  Increased groundwater moisture •  Forest species composiDon and size distribuDon •  MulD-­‐stemmed growth •  Woody species regeneraDon Photo: www.mar5nezbeavers.org Photo: Anna M. Harrison 69 Beaver forage footprint is well-­‐predicted from pond area A Classic ‘Central Place’ Forager •  Schoener (1979) predicted that animals forage to maximize net energy intake per unit Dme •  Because provisioning costs increase with distance, animals should forage more selecDvely farther from their central place (i.e., the lodge). •   Food size;  Provisioning Dme Nicola Plowes, ASU (Harrison & Stella 2010) Not for distribu5on (Contact J. Stella, stella@esf.edu) John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design Not for distribu5on (Please contact J. Stella, stella@esf.edu) 15 9/24/11 SelecDvity Generalized Central Place foraging model Most Preferred Op
P p or
tun
i sD
MoDvaDng QuesDon: How do forest impacts by beaver vary with distance from the pond? c Non-­‐preferred Distance from the impoundment (Harrison & Stella 2010) Not for distribu5on Not for distribu5on (Please contact J. Stella, stella@esf.edu) (Please contact J. Stella, stella@esf.edu) General Forage Preferences (but high system specificity) Species selecDvity in a Northeastern forest •  Forage preferences depend on what is available Selec<vity = cut / cut + live (Gallant et al. 2004, Raffel 2009) 0.8
–  Aspen, willow, codonwood, and alder (Denny 1952) 0.7
Stem
0.6
•  Preferred sizes: <10 cm diam. allows for mulDple uses Live Stem Relative Density
Basal Area
BF, 15%
Selectivity
0.5
(Pinkowski 1983, Barnes and Mallik 1997, Haarberg & Rosell 2006, Raffel et al. 2009) •  Increase selecDvity with increased distance from pond Photo: Anna M. Harrison Photo: Josh Cousins (Raffel et al 2009) Not for distribu5on (Please contact J. Stella, stella@esf.edu) John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design STM,
5%
BH, 4%
0.4
0.3
AB,
29%
RS,
17%
0.2
RM,
10%
0.1
0.0
r.
St
p
Ma
le
A.
b
ch
ee
ut
ce
ch aple aple
sh
ru
bir
.a
eln
m
m
sp
W
az
Y.
h
S.
R.
R.
B.
fir
B.
SM, 7%
WA, 1%
YB, 8%
Harrison & Stella 2010; Not for distribu5on (Contact J. Stella, stella@esf.edu) Woody tree species
16 9/24/11 ;-%&"!,*+*,,*2"
3*4*)56$&7"
Size selecDvity along a distance gradient :0
!"
0 -,
&1(
$%
5)"
Predicting likelihood of harvest as a
function of size and distance
8-(90,*+*,,*2"
#$%&'()*"+,-."&/*"$.0-1(2.*(&"
Most-­‐preferred: 2-­‐5 cm stems Opportunis<c: <2 and 5-­‐10 cm Non-­‐preferred: >10 cm Harrison & Stella 2010; Not for distribu5on (Contact J. Stella, stella@esf.edu) Large area + concentrated impact on forest tree communiDes Transect  Pond/Reach  Forest Landscape (Harrison &. Stella 2010) Not for distribu5on (Please contact J. Stella, stella@esf.edu) John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design Harrison & Stella 2010; Not for distribu5on (Contact J. Stella, stella@esf.edu) ImplicaDons •  Broad landscape controls on beaver behavior •  Fine-­‐scale foraging impacts of beaver •  ResulDng changes to forest composiDon and abioDc environment •  Extensive areas with intense foraging could alter forest composiDon 3 9 12 4 (Harrison &. Stella 2010) Not for distribu5on (Please contact J. Stella, stella@esf.edu) 17 9/24/11 MulD-­‐stemmed growth habit MulD-­‐stemmed growth habit Photos: Anna M. Harrison Harrison & Stella 2010; Not for distribu5on (Contact J. Stella, stella@esf.edu) Multi-stemmed growth
Harrison & Stella 2010; Not for distribu5on (Contact J. Stella, stella@esf.edu) John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design Woody species regenera<on 82
Photo: Bill Dunker
18 9/24/11 Woody Species Regeneration
Photo: Bill Dunker
Woody Species Regeneration
Harrison & Stella 2010; Not for distribu5on (Contact J. Stella, stella@esf.edu) Not for distribu5on Harrison & Stella 2010; Not for distribu5on 84
(Contact J. Stella, stella@esf.edu) (Contact J. Stella, stella@esf.edu) Summary of Impacts
Acknowledgements
•  Impacts extend to 80
meters from pond edge
•  Greatest impact on nearpond forest
•  Foraging creates
positive feedbacks that
increase preferred sizes
and species
•  Adaptability of beaver
will aid in long-term
presence on landscape
•  Stacy McNulty, Dr. Jacqueline
Frair, Dr. Jake Bendix
•  Drs. Martin Dovciak, Ruth
Yanai, Steve Stehman, and
Carol Johnston
•  Adirondack Ecological Center
•  Field support: Joshua
Cousins, Bill Dunker, Michael
Silberman & James Johnson
Funding:
•  SUNY-ESF McIntire Stennis
Grant Program
Photo: Anna M. Harrison Not for distribu5on Photo: Anna M. Harrison For more informaDon: stella@esf.edu 87
(Please contact J. Stella, stella@esf.edu) John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 19 9/24/11 References Baker, B. W., and E. P. Hill. 2003. Beaver (Castor canadensis). Pages 288-­‐310 in G. A. Feldhamer, B. C. Thompson, and J. A. Chapman, editors. Wild Mammals of North America: Biology, Management, and ConservaDon. Second EdiDon. The Johns Hopkins University Press, BalDmore, Maryland, USA. Barnes, D. M., and A. U. Mallik. 1997. Habitat factors influencing beaver dam establishment in a northern Ontario watershed. Journal of Wildlife Management 61:1371–77. Denney, R. N. 1952. A summary of North American beaver management, 1946– 1948 (Current Report 28). Colorado Game and Fish Department, Denver. Haarberg, O. and F. Rosell. 2006. SelecDve foraging on woody plant species by the Eurasian beaver (Castor fiber) in Telemark, Norway. Journal of Zoology 270: 201-­‐208. Harrison, A.M., J.C. Stella. 2010. Engineering the forest ecosystem: impacts on woody vegetaDon structure and composiDon by beaver, a central place forager. MeeDng of the Ecological Society of America, Pidsburgh, PA. Keeton, W. S., C. E. Kra{, and D. R. Warren. 2007. Mature and old-­‐growth riparian forests: Structure, dynamics, and effects on adirondack stream habitats. Ecological ApplicaDons 17:852-­‐868. Naiman, R. J., H. Decamps, and M. E. McClain. 2005. Riparia: Ecology, ConservaDon, and Management of Streamside CommuniDes. Elsevier, Amsterdam Pinkowski, B. 1983. Foraging behavior of beavers (Castor canadensis) in North Dakota. Journal of Mammalogy 64(2): 312-­‐314. Raffel, T.R., N. Smith, C. Cortright, and A.J. Gatz. 2009. Central place foraging by beavers (Castor canadensis) in a complex lake habitat. American Midland Naturalist 162: 62-­‐73. John C. Stella, PhD, 2011 (stella@esf.edu) Partnering with Beaver in RestoraDon Design 20