HomeMy WebLinkAboutSuWa289sec10-18Alaska Resources Library & Information Services
Susitna‐Watana Hydroelectric Project Document
ARLIS Uniform Cover Page
Title:
Wood frog occupancy and habitat use, Study plan Section 10.18, Study
Completion Report SuWa 289
Author(s) – Personal:
Author(s) – Corporate:
ABR, Inc.-Environmental Research & Services, [Offices in] Forest Grove, Oregon and Fairbanks,
Alaska
AEA‐identified category, if specified:
November 2015; Study Completion and 2014/2015 Implementation Reports
AEA‐identified series, if specified:
Series (ARLIS‐assigned report number):
Susitna-Watana Hydroelectric Project document number 289
Existing numbers on document:
Published by:
[Anchorage : Alaska Energy Authority, 2015]
Date published:
October 2015
Published for:
Alaska Energy Authority
Date or date range of report:
Volume and/or Part numbers:
Study plan Section 10.18
Final or Draft status, as indicated:
Document type:
Pagination:
40, 5, 8 pages
Related works(s):
Pages added/changed by ARLIS:
Notes:
All reports in the Susitna‐Watana Hydroelectric Project Document series include an ARLIS‐
produced cover page and an ARLIS‐assigned number for uniformity and citability. All reports
are posted online at http://www.arlis.org/resources/susitna‐watana/
Susitna–Watana Hydroelectric Project
(FERC No. 14241)
Wood Frog Occupancy and Habitat Use
Study Plan Section 10.18
Study Completion Report
Prepared for
Alaska Energy Authority
Prepared by
ABR Inc.—Environmental Research & Services
Forest Grove, Oregon and Fairbanks, Alaska
October 2015
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page i October 2015
TABLE OF CONTENTS
1. Introduction ....................................................................................................................... 5
2. Study Objectives................................................................................................................ 5
3. Study Area ......................................................................................................................... 6
3.1. Study Area Variance ............................................................................................... 6
4. Methods and Variances .................................................................................................... 6
4.1. Auditory Field Surveys ........................................................................................... 7
4.1.1. Variances ......................................................................................... 8
4.2. Occupancy Modeling and Habitat Associations ..................................................... 9
4.2.1. Variances ....................................................................................... 10
4.3. Acoustic Monitoring ............................................................................................. 10
4.3.1. Variances ....................................................................................... 10
4.4. Chytrid Fungus Bioassay ...................................................................................... 11
4.4.1. Variances ....................................................................................... 11
5. Results .............................................................................................................................. 11
5.1. Auditory Field Surveys ......................................................................................... 12
5.1.1. 2013 Sampling .............................................................................. 12
5.1.2. 2014 Sampling .............................................................................. 12
5.2. Occupancy Modeling and Habitat Associations ................................................... 12
5.2.1. Occupancy Modeling .................................................................... 12
5.2.2. Habitat Associations ..................................................................... 14
5.3. Acoustic Monitoring ............................................................................................. 15
5.3.1. 2013 Sampling .............................................................................. 15
5.3.2. 2014 Sampling .............................................................................. 15
5.4. Chytrid Fungus Bioassay ...................................................................................... 15
6. Discussion......................................................................................................................... 16
6.1. Distribution and Habitat Use ................................................................................ 16
6.2. Occupancy Modeling ............................................................................................ 18
6.3. Acoustic Monitoring ............................................................................................. 19
6.4. Chytrid Fungus...................................................................................................... 20
7. Conclusion ....................................................................................................................... 20
8. Literature Cited .............................................................................................................. 21
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page ii October 2015
9. Tables ............................................................................................................................... 24
10. Figures .............................................................................................................................. 33
LIST OF TABLES
Table 5.1-1. Number of Frog-survey Visits to Water Bodies and Wetlands in the 2013 Study
Area. .............................................................................................................................................. 24
Table 5.1-2. Frog Detection in Shallow- and Deep-water Habitats in 2013. ................................ 24
Table 5.1-3. Number of Frog-survey Visits to Water Bodies and Wetlands in the 2014 Study
Area. .............................................................................................................................................. 25
Table 5.1-4. Frog Detection in Shallow- and Deep-water Habitats in 2014. ................................ 25
Table 5.1-5. Occupancy-model Selection Results for Presence of Wood Frogs in 2013. ............ 26
Table 5.1-6. Top Occupancy-model Selection Results for Presence of Wood Frogs in 2014. .... 27
Table 5.2-1. Model Estimates of Wood Frog Occupancy and Detection Probability in 2013 from
the Best Model. ............................................................................................................................. 27
Table 5.2-2. Model Estimates of Wood Frog Occupancy and Detection Probability in 2014 from
the Best Model. ............................................................................................................................. 28
Table 5.2-3. Model Estimates of Wood Frog Occupancy and Detection Probability in 2014 from
the Third-best Model..................................................................................................................... 28
Table 5.2-4. Habitat Characteristics of Water Bodies and Wetlands where Wood Frogs were
Detected and Not Detected in 2013. ............................................................................................. 29
Table 5.2-5. Habitat Characteristics of Water Bodies and Wetlands where Wood Frogs were
Detected and Not Detected in 2014. ............................................................................................. 31
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page iii October 2015
LIST OF FIGURES
Figure 3-1. Wood Frog Study Area for the Susitna–Watana Hydroelectric Project in 2013. ...... 34
Figure 3-2. Wood Frog Study Area for the Susitna–Watana Hydroelectric Project in 2014. ...... 35
Figure 5.1-1. Locations where Wood Frogs were Detected during Auditory Surveys, plus
Incidental Observations from Other Wildlife Surveys, in 2013. .................................................. 36
Figure 5.1-2. Locations where Wood Frogs were Detected during Auditory Surveys, plus
Incidental Observations from Other Wildlife Surveys, in 2014. .................................................. 37
Figure 5.3-1. Wood Frog Calling Activity by Date and Hour in 2013 (error bars depict 1 S.E.). 38
Figure 5.3-2. Wood Frog Calling Activity by Date and Hour in 2014 (error bars depict 1 S.E.). 39
Figure 5.3-3. Average Hourly Air Temperatures Measured at Acoustic Monitor Stations during
May 20–29, 2014 (error bars depict 1 S.E.). ................................................................................. 40
APPENDICES
Appendix A: Records of Additional Consultation in 2014 with USGS Regarding Sampling
Protocol and Analytical Method for Amphibian Chytrid Fungus.
Appendix B: Photographs from Field Surveys in 2013 and 2014.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page iv October 2015
LIST OF ACRONYMS, ABBREVIATIONS, AND DEFINITIONS
Abbreviation Definition
ADF&G Alaska Department of Fish and Game
AEA Alaska Energy Authority
AICc Akaike’s Information Criteria, corrected for small sample size
AKNHP Alaska Natural Heritage Program
APA Alaska Power Authority
Bd Batrachochytrium dendrobatidis (amphibian chytrid fungus)
C.I. confidence interval
CIRWG Cook Inlet Region Working Group
d degrees of freedom
EC electrical conductivity
FERC Federal Energy Regulatory Commission
ft foot, feet
GIS Geographic Information System
GPS Global Positioning System
ILP Integrated Licensing Process
ISR Initial Study Report
kph kilometers per hour
m meter
min minimum
mph miles per hour
n sample size
n/a not applicable
NHD National Hydrography Dataset
NWI National Wetlands Inventory
p detection probability
pH a measure of the acidity or alkalinity of a solution
PLP Pebble Limited Partnership
Project Susitna-Watana Hydroelectric Project
qPCR Quantitative Polymerase Chain Reaction
RSP Revised Study Plan
S.E. standard error
SPD Study Plan Determination
USFWS United States Fish and Wildlife Services
USGS United States Geological Survey
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 5 October 2015
1. INTRODUCTION
The work described herein for the Wood Frog Occupancy and Habitat Use Study (Wood Frog
Study, for short) was conducted according to Section 10.18 of the Revised Study Plan (RSP)
approved by the Federal Energy Regulatory Commission (FERC or Commission) for the
Susitna–Watana Hydroelectric Project, FERC Project No. 14241. The Wood Frog Study focused
on auditory surveys of calling male wood frogs (Rana [Lithobates] sylvatica), including
deployment of acoustic monitors, during the spring breeding season and on habitat occupancy
modeling using the results of those surveys.
A summary of the development of this study, together with the Alaska Energy Authority’s
(AEA) implementation of the study through the 2013 study season, was presented in the Initial
Study Report (ISR) that was filed with FERC in June 2014 (ABR 2014a, 2014b, 2014c). As
required under FERC’s regulations for the Integrated Licensing Process (ILP), the ISR described
AEA’s “overall progress in implementing the study plan and schedule and the data collected,
including an explanation of any variance from the study plan and schedule” (18 CFR 5.15(c)(1)).
Since filing the ISR in June 2014, AEA has continued to implement the FERC-approved plan for
the Frog Study. For example:
A second year of auditory surveys and acoustic monitoring of calling male wood frogs
was conducted in spring 2014, followed by habitat occupancy modeling.
On October 21, 2014, AEA held an ISR meeting for the Wood Frog Study and the other
studies in the wildlife program.
In furtherance of the next round of ISR meetings and FERC’s Study Plan Determination (SPD)
expected in 2016, this report contains a comprehensive discussion of results of the Frog Study
from the beginning of AEA’s study program in 2013 through the end of calendar year 2014. It
describes the methods and results of the Frog Study and explains how the study objectives set
forth in the FERC-approved Study Plan have been met. Accordingly, with this report, AEA has
now completed all field work, data collection, data analysis, and reporting for this study.
2. STUDY OBJECTIVES
The goal of the Wood Frog Study is to characterize the use of the Project area by breeding wood
frogs to facilitate an assessment of potential impacts on wood frogs from development of the
proposed Project.
The study has four objectives, as outlined in RSP Section 10.18.1:
Review existing data on habitat use and distribution of breeding wood frogs in a broad
region surrounding the study area.
Estimate the current occupancy rate for breeding wood frogs in suitable habitats in the
study area through a combination of field surveys and habitat-occupancy modeling.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 6 October 2015
Collect information on current habitat occupancy and habitat use to enable estimation of
the habitat loss and alteration expected to occur from development of the Project.
Sample frogs opportunistically for the presence of the chytrid fungus that has been linked
to amphibian population declines. (At the request of state and federal management
agencies, AEA agreed to sample for the chytrid fungus to opportunistically take
advantage of planned fieldwork and thereby provide some baseline information on the
potential occurrence of the fungus in the study area before development.)
3. STUDY AREA
As established by RSP Section 10.18.1, the study area included those water bodies and suitable
wetland habitats in the proposed Project area in which habitat loss, habitat alteration, and
disturbance could potentially occur. The study area encompassed the reservoir inundation zone,
associated areas for the dam and camp infrastructure, and the potential access-road corridors
(Gold Creek, Chulitna, and Denali corridors) and material sites (Figures 3-1 and 3-2). Field
sampling in 2013 was focused on the reservoir zone, dam and camp area, Chulitna Corridor, and
a small portion of the Denali corridor (ABR 2014a). In 2014, field sampling was focused on the
Gold Creek corridor and on the Denali West and Denali East corridor options (ABR 2014c).
3.1. Study Area Variance
As described in Section 1.4 of the ISR Overview, when the ISR was filed, AEA explained that it
had decided to pursue the study of an additional alternative north/south-oriented corridor
alignment for transmission and access from the Denali Highway to the proposed dam site.
Referred to the “Denali East Option,” this area was added to the Wood Frog Study area in May
2014 (Figure 3-2). The Denali Corridor surveyed in 2013 and reported on in the ISR and in 2013
figures and tables in this report is essentially equivalent to the Denali West Corridor surveyed in
2014.
In addition, Section 1.4 of the ISR noted that AEA was considering the possibility of eliminating
the Chulitna Corridor from further study, so no surveys were conducted in that corridor in 2014
for the Wood Frog Study. In September 2014, AEA filed with FERC a formal proposal to
implement this change. Thus, this report reflects a change in the study area to no lo nger include
the Chulitna Corridor (Figure 3-2). Removal of the Chulitna Corridor resulted in minor changes
to the 2014 study area buffer around the Gold Creek Corridor. Although the Chulitna Corridor
was dropped from further study in 2014, lakes within the Chulitna Corridor that were surveyed
for the APA Project in the 1980s were surveyed again in 2014 and those data are included herein
for comparative purposes.
4. METHODS AND VARIANCES
The methods for each of the components of the Wood Frog Study are presented in this section.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 7 October 2015
4.1. Auditory Field Surveys
AEA implemented the methods described in the Study Plan (RSP Section 10.18.4.1), with the
exception of the variances explained below (Section 4.1.1) and previously in Section 4.1 of ISR
10.18 Part A (ABR 2014a).
As indicated in the Study Plan, because the study area is large and the calling period of breeding
male frogs is short, this study did not involve a comprehensive survey of all potential frog
breeding habitat present in the study area. Instead, observers surveyed for frogs in suitable
habitats that were stratified into two habitat types (water bodies and wetlands). The study team
used a Geographic Information System (GIS) to compile the full list of possible sampling
locations (148 in 2013 and 221 in 2014) by reviewing available information from existing GIS
data layers (National Hydrography Dataset [NHD] and National Wetlands Inventory [NWI]) and
by conducting additional interpretation of aerial imagery for portions of the study area for which
recent imagery was available. The study team selected suitable individual water body and
wetland habitats for auditory sampling by (1) identifying areas with emergent vegetation; (2)
removing shoreline wetland polygons adjacent to water bodies (and just including the water
bodies); (3) removing locations within 250 m of another suitable location; (4) including sampling
locations on Cook Inlet Region Working Group (CIRWG) lands in 2014, for which access was
not permitted in 2013; and (5) including high-elevation areas in 2014 that were frozen during the
field-sampling period in 2013. Next, the study team selected sampling locations (120 in 2013 and
131 in 2014) by stratifying equally by area (access road or dam/camp area/reservoir zone [2013
only]) and then randomly selecting approximately equal numbers of each habitat type (water
body, wetland) within each area. In 2014, sampling locations were stratified by access corridor
(Gold Creek or Denali), except that all available water bodies and wetlands in the Denali East
and West access corridor options were selected because relatively few were available in those
corridors. The study team included the remaining locations (28 in 2013 and 90 in 2014) as
alternative sampling locations, if needed.
The study team conducted ground-based auditory surveys of the randomly selected water bodies
and wetlands in the study area during the early spring breeding season for wood frogs, the
accepted survey time for this species (Gotthardt 2004; PLP 2011). Before the surveys began,
observers trained by listening to digital audio files of the breeding calls of male wood frogs. Up
to three replicate surveys were made by trained observers at each water body or wetland during
May 30–June 8, 2013, and May 20–29, 2014. In addition to these surveys, incidental detections
of wood frogs were documented during data collection efforts for other studies (mainly ground-
based bird surveys), which provided additional information on the occurrence of frogs in the
study area. The study team reached the survey sites by helicopter and then on foot by navigating
to predetermined sample sites using hand-held Global Positioning System (GPS) receivers. The
field observers listened for calling frogs during 5-min sampling periods along the margins of
each water body or wetland sampled to determine whether or not frogs were calling. At small
water bodies and wetlands, a single observation point was sufficient to detect calling frogs, but
for large water bodies and wetlands, multiple observation points were needed to discern whether
frogs were calling. Up to four observation points were located and sampled for large water
bodies and wetlands, with distances of up to 500 m (1,640 ft) being designated between adjacent
sampling points to achieve adequate survey coverage.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 8 October 2015
Due to variability in the calling frequency of male wood frogs even during the peak of the
breeding season (PLP 2011), at least two, and occasionally three, visits were needed to detect
frogs at some water bodies. The second or third surveys at each site were conducted by a
different observer who generally did not have knowledge of the survey results from the first
survey. Because this study involved the use of a “removal design” to estimate occupancy,
however, additional surveys were not needed if frogs were detected on the first survey (i.e., that
site was removed from further sampling; Mackenzie and Royle 2005). Surveys were conducted
only under favorable weather conditions (e.g., light rain or no rain, air temperature higher than 4°
C [39° F], and wind speed ≤25 kph [15 mph]). Observers spent a minimum of 5 min at each
survey location listening for calling frogs, but terminated the survey sooner if frogs were
detected.
Habitat and environmental characteristics (size and depth of water body or wetland, substrate,
presence and type of emergent aquatic vegetation, water quality characteristics [pH level,
dissolved oxygen, specific electrical conductivity (EC)], ice cover, surrounding terrestrial
vegetation, water and air temperature, precipitation, cloud cover, wind speed, time of day, beaver
activity) were recorded during the field surveys for use in the development of a Project-specific
model to estimate occupancy based on the habitat characteristics of the occupied water bodies or
wetlands.
4.1.1. Variances
The Study Plan (RSP Section 10.18.4.1) proposed that the potential water bodies and wetland
habitats to be sampled would be identified from interpretation of aerial photos or remote-sensing
imagery and from the preliminary mapping of vegetation, wildlife habitats, and wetlands. From
that set of water bodies and wetlands, habitats were to be categorized as having a high or low
probability of supporting breeding frogs (based on likelihood of supporting fish and presence of
emergent vegetation). Lastly, the Study Plan proposed to select 10 sampling regions, two in each
of the three potential access-road corridors and four in the reservoir zone and dam and camp
facilities area. In each sampling region, 12 potential water bodies or wetlands were to be selected
through a stratified random process.
Several factors affected the study team’s ability to implement the sampling approach described
in the Study Plan: (1) current mapping of vegetation and wildlife habitats was not yet available
before the 2013 and 2014 field seasons began; (2) existing wetland information (e.g., NWI
mapping) did not cover the entire study area and was not of sufficient accuracy and resolution for
the study; (3) data were not available regarding the presence of fish in water bodies and wetlands
before field surveys began; and (4) permission for access to CIRWG lands was not granted in
2013, precluding sampling in most of the Gold Creek Corridor and parts of the Chulitna Corridor
and the western portion of the reservoir zone in 2013. Therefore, the study team devised an
alternative approach to selecting sampling locations (120 in 2013 and 131 in 2014) that still
incorporated random selection of suitable sampling sites, as described in Section 4.1 above. This
selection process fulfilled the original intent of the study plan to select sampling locations in a
random manner throughout the study area.
In addition, the Study Plan (RSP Section 10.18.4.1) included the distribution of field survey
times each day, which were originally planned for the period from approximately 12:00 to 22:00
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 9 October 2015
but were conducted from approximately 09:00 to 20:00 (2013) and 09:30 to 19:30 (2014)
instead, due to logistical challenges. The data from acoustic monitors showed that the sampling
times were appropriate for the study, as is described below in Section 5.3. The acoustic monitors
provided excellent results for evaluating the times of day when frogs were calling.
As explained above, the applicable study objectives were achieved with these modified
approaches.
4.2. Occupancy Modeling and Habitat Associations
AEA implemented the methods described in the Study Plan (RSP Section 10.18.4.1) with no
variances, as described previously in Section 4.2 of ISR 10.18 Part A (ABR 2014a).
Because frogs were not always detected during 5-min sampling sessions when they were present,
the study team used occupancy modeling to adjust the observed occupancy rates for non-
detections (Mackenzie et al. 2002). Occupancy modeling uses resurveys of the same locations to
estimate a detection rate (p) and then uses the estimated detection rate to calculate an adjusted
occupancy rate estimate (Ψ). The observed (“naïve”) occupancy rate of frogs in water bodies and
wetlands was adjusted to account for those frogs present but not detected, thereby producing a
corrected occupancy rate for the water bodies and wetlands in this study.
Occupancy modeling also allows the user to compare various models with different
specifications of detectability and occupancy parameters. The study team used a removal design
in which locations were not revisited after frogs were detected, resulting in limited statistical
power to estimate detectability parameters. It was assumed, therefore, that detectability was
constant for all surveys in 2013. In 2014, the study team tested two model types for detectability:
one assumed that detectability was constant for all surveys and the other assumed that
detectability differed between corridors (Denali and Gold Creek); the latter was added to the
analysis because of the large differences observed in seasonal phenology between the Gold
Creek and Denali corridors in 2014. The study team compared four covariates for occupancy:
(1) area (dam/camp area plus reservoir zone vs. access-road corridors) in 2013 or corridor
(Denali vs. Gold Creek) in 2014; (2) water type (wetland or water body); (3) water depth (≤1.5 m
[4.9 ft] or >1.5 m); and (4) percent of hibernation habitat (visual estimate of the percent of
herbaceous cover, low shrubs, and tall shrubs within 50 m of the shoreline). Area was included
in 2013 because the sample was stratified by area, whereas corridor was included in 2014
because the sample was stratified by corridor and the corridors had different elevation ranges.
The other three covariates were chosen because they were expected to be the most biologically
important, based on field observations and the results of other studies. With the available sample
size, the analyses would only support a limited number of covariates.
The study team tested all possible combinations of these four occupancy covariates (without
interactions), including an intercept-only model, and with two detectability models in 2014
(constant and corridor), resulting in a total of 16 different models for the 2013 results and 32
different models for the 2014 results. Model calculations were conducted with a desktop
computer using the single-season analysis format and custom model-building features of the
software program PRESENCE (Hines 2006).
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 10 October 2015
The different models for each year were compared using information–theoretic methods
(Burnham and Anderson 2002). For each model, the study team calculated the Akaike
Information Criterion corrected for small sample sizes (AICc), which compares model fit and
penalizes models for the number of parameters to determine the most parsimonious model (the
best fit with the fewest number of parameters). The number of different locations was used as the
effective sample size. The AICc values were used to calculate the Akaike weight (ωi), which is
the probability that each model is the best model in the candidate set (Burnham and Anderson
2002).
4.2.1. Variances
No variances from the methods described in the Study Plan were implemented in 2013 or 2014.
4.3. Acoustic Monitoring
AEA implemented the methods described in the Study Plan (Section 10.18.4.1) with no
variances.
The study team used Wildlife Acoustics Song Meter SM2BAT+ platforms with SMX-II
microphones to record frog calls onto 32-GB (Class 4 SDHC) data cards. The monitors were
internally powered with rechargeable D-cell batteries (Imedion 9,500 mAh). Five acoustic
monitors were deployed to increase accuracy in calculating the detectability of calling frogs. The
monitors were deployed at a subset of water bodies and wetlands on state, federal, and, in 2014
only, on CIRWG lands known to be occupied by frogs. Although the monitors were programmed
to record full-spectrum audio recordings for the first 30 min of each hour around the clock, the
study team analyzed only the first 10 min of each hour. Analytical results indicated that this
subsampling adequately characterized the calling activity within the hour.
In 2013 the study team used the proportion of 5-min periods with frogs calling as an independent
estimate of the ability to detect frogs at a given location, assuming that frogs were present. Due
to battery failures for several monitors in 2014, however, it was not possible to derive a similar
estimate for that year because of the small sample sizes obtained. The validity of this estimate
relies on several assumptions: (1) individual observers were able to detect frogs calling at least as
well as were the acoustic monitors; (2) the presence of observers did not lower the probability of
frogs vocalizing; and (3) the locations chosen for acoustic monitoring were representative of all
locations at which frogs were present. For each location surveyed, the study team determined the
hour of the day in which the visit occurred and calculated the proportion of 5-min periods in
which frog calls were heard on acoustic monitors during that hour. The study team then
calculated the mean of all these proportions for each visit as a second, independent estimate of
detectability.
4.3.1. Variances
No variances from the methods described in the Study Plan were implemented in 2013 or 2014.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 11 October 2015
4.4. Chytrid Fungus Bioassay
Sampling and laboratory assay methods for the chytrid fungus (Bd) were identified through
consultation with U.S. Fish and Wildlife Service (USFWS) representatives in Alaska, who
recommended that Tara Chestnut, an expert with the U.S. Geological Survey (USGS) in
Portland, Oregon, be contacted for sampling protocols (Appendix A). Biologists wore fresh
nitrile gloves and sprayed boots with a 10 percent bleach solution at each sampling location to
prevent potential contamination among sites.
The study team captured seven frogs in 2013 by hand opportunistically and swabbed the skin of
the abdomen, inner thighs, and undersides of foot webbing for a total of 25 times with a sterile
cotton swab, after which the frog was released unharmed. Swabs were placed in tubes that were
refrigerated until all seven samples were shipped on dry ice to the USGS Microbiology
laboratory in Reston, Virginia. The lab analyzed the samples using a quantitative polymerase
chain reaction (qPCR) technique to test for the presence of Bd fungus.
Sampling for the chytrid fungus (Bd) was not conducted in 2014 (see Section 4.4.1).
4.4.1. Variances
No variances from the methods described in the Study Plan were implemented in 2013.
Because of the small sample size obtained in 2013 (n = 7), opportunistic capture and swabbing of
adult frogs to sample for the presence of amphibian chytrid fungus (RSP Section 10.18.4.2) was
dropped from the field effort in 2014, as was discussed and agreed to in the technical meeting on
March 6, 2014 (see meeting notes here: http://www.susitna-watanahydro.org/wp-
content/uploads/2014/03/2014-03-06TT_Wildlife_MeetingNotes.pdf) and was discussed further
with USGS (see Appendix A in ISR 10.18 Part A). Dropping the opportunistic sampling of frogs
for the presence of Bd in 2014 was a study plan modification described in Section 7.1.2 of ISR
10.18 Part C (ABR 2014c); the study plan objective to sample frogs opportunistically was
fulfilled in 2013.
5. RESULTS
Cumulative data developed in support of the Study Completion Report for 2013–2014 are
available for download at http://gis.suhydro.org/SIR/10-Wildlife/10.18-Wood_Frogs/:
FROG_10_18_2013_2014_ABR.gdb/FROG_2013_2014_AcousticMonitors
FROG_10_18_2013_2014_ABR.gdb/FROG_2013_2014_IncidentalObs
FROG_10_18_2013_2014_ABR.gdb/FROG_2013_2014_SamplingSites
FROG_10_18_2013_2014_ABR.gdb/FROG_2013_StudyArea
FROG_10_18_2013_2014_ABR.gdb/FROG_2014_Actual_Field_StudyArea
FROG_10_18_Acoustic_Monitoring_2013_2014_ABR.xlsx.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 12 October 2015
5.1. Auditory Field Surveys
5.1.1. 2013 Sampling
As described in Section 5.1 of ISR 10.18 Part A(ABR 2014a), the study team surveyed a total of
90 different wetlands and water bodies for the presence of wood frogs in 2013 (Table 5.1-1,
Figure 5.1-1). Additional water bodies and wetlands (n = 17) were visited but were excluded
from the analyses for various reasons (e.g., water still frozen or insufficient water depth). Frogs
were detected at 37 of the 90 locations (41.1 percent) on the first visit (Table 5.1-2) including 35
locations where frogs were heard calling and two locations where frogs were not heard but egg
masses were found. The latter two locations were treated as non-detections in occupancy
modeling, however, because frogs were not detected using the normal survey method. The study
team conducted a second survey visit at 50 of the 53 locations where frogs were not detected on
the first visit, producing detections at 8 more locations (16.0 percent). A third visit was
conducted at five of the 42 sites where frogs were not detected on the first and second visits,
producing detections at two more locations (40.9 percent). Overall, frogs were heard or egg
masses were observed at 47 (52.2 percent) of the 90 locations sampled (Table 5.1-2, Figure 5.1-
1). Therefore, the naïve estimate of frog occupancy (assuming 100 percent detectability) was
52.2 percent.
5.1.2. 2014 Sampling
The study team surveyed a total of 104 different wetlands and water bodies for the presence of
wood frogs in 2014 (Table 5.1-3, Figure 5.1-2). Additional water bodies and wetlands (n = 31)
were visited but were excluded from the analyses for various reasons (e.g., water still frozen or
insufficient water depth). Frogs were detected at 14 of the 104 locations (13.5 percent) on the
first visit (Table 5.1-4). The study team conducted a second survey visit at the 90 locations where
frogs were not detected on the first visit, producing detections at 7 more locations (7.8 percent).
A third visit was conducted at three of the 83 sites where frogs were not detected on the first and
second visits, with no additional detections. Overall, frogs were heard or egg masses were
observed at 21 (20.2 percent) of the 104 locations sampled (Table 5.1-4, Figure 5.1-2).
Therefore, the naïve estimate of frog occupancy (assuming 100 percent detectability) was 20.2
percent across all locations sampled.
5.2. Occupancy Modeling and Habitat Associations
5.2.1. Occupancy Modeling
5.2.1.1. 2013 Sampling
The best model of frog occupancy contained only one variable: water depth. Based on the
Akaike weight, this model had a 31.9 percent chance of being the best model in the candidate set
(Table 5.1-5). The next three competing models contained water depth and one of the other
variables but, in all cases, the 95 percent confidence interval (C.I.) for the other variable
contained zero, suggesting that variables other than water depth added little to the model. After
water depth was included, no statistical evidence was found to indicate that occupancy rates
varied by area, by water-body type, or with increasing hibernation habitat.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 13 October 2015
The estimated detectability from the best model was 60.6 percent (95 percent C.I. = 34.8–81.6
percent; Table 5.2-1). The model results indicated that, if frogs were present in a pond, the study
team would, on average, detect them 60.6 percent of the time with one visit, 84.5 percent of the
time with two visits, and 93.9 percent of the time with three visits.
The estimated occupancy for shallow-water habitats was 36.8 percent (95 percent C.I. = 20.8–
56.5 percent) and the estimated occupancy for deep-water habitats was 81.8 percent (95 percent
C.I. = 44.4–96.2 percent; Table 5.2-1). As would be expected, these estimates were slightly
higher than the naïve estimates of 31.0 percent and 70.8 percent, respectively. The sample
included 42 shallow-water habitats (46.7 percent) and 48 deep-water habitats (53.3 percent).
Assuming that this ratio is representative of the entire area sampled in 2013, the overall
occupancy estimate was 63.4 percent (Table 5.2-1).
5.2.1.2. 2014 Sampling
The best model of frog occupancy contained only one occupancy variable (water depth, similar
to 2013) and one detectability variable (corridor—Gold Creek Corridor or Denali Corridor
options). Based on the Akaike weight, this model had a 16.5 percent chance of being the best
model in the candidate set (Table 5.1-6). The second-best model, which contained the same
variables plus hibernation habitat as an occupancy variable, had a 13.6 percent chance of being
the best model in the candidate set. The confidence interval for hibernation habitat contained
zero, however, suggesting that the variable added little to the model. The third-best model, which
included water depth and corridor as occupancy variables and for which detectability was
constant across all sites, had a 13 percent chance of being the best model in the candidate set.
The first, second, and third-best models all included depth and corridor as variables, but the first
and second models assumed that detectability differed between corridors and occupancy was the
same, whereas the third model assumed that detectability was the same between corridors and
occupancy differed (Tables 5.2-2, 5.2-3). Because the two groups of models gave different
occupancy estimates, one estimate from each group (best model and third-best model) is
presented to acknowledge this difference in interpretation (Tables 5.2-2, 5.2-3).
The estimated detectability from the best model, for which detectability differed between
corridors, was 56.6 percent (95 percent C.I. = 14.5–90.9 percent; Table 5.2-2) in the Gold Creek
Corridor. This model indicated that, if frogs were present at a sampling site in the Gold Creek
Corridor, the study team would, on average, detect them 56.6 percent of the time with one visit,
81.2 percent of the time with two visits, and 91.8 percent of the time with three visits. In
contrast, detectability was estimated to be just 16.0 percent (95 percent C.I. = 5.5–38.7 percent;
Table 5.2-2) in the Denali corridors (West and East options combined). The model indicated that,
if frogs were present at a sampling site in the Denali corridors, the study team would, on average,
detect them 16.0 percent of the time with one visit, 29.4 percent of the time with two visits, and
40.7 percent of the time with three visits.
The estimated detectability from the third-best model, with constant detectability in both
corridors, was 54.6 percent (95 percent C.I. = 22.6–83.2 percent; Table 5.2-3). The model results
indicated that, if frogs were present at a sampling site, the study team would, on average, detect
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 14 October 2015
them 54.6 percent of the time with one visit, 79.4 percent of the time with two visits, and 90.6
percent of the time with three visits.
Based on the results of the best model with different detectability between corridors, the
estimated occupancy for shallow-water habitats was 17.9 percent (95 percent C.I. = 5.9–43.1
percent) and the estimated occupancy for deep-water habitats was 71.9 percent (95 percent C.I. =
19.2–96.5 percent; Table 5.2-2). As would be expected, these estimates were higher than the
naïve estimates of 8.6 percent and 34.7 percent, respectively. The sample included 58 shallow-
water habitats (55.8 percent) and 46 deep-water habitats (44.2 percent). Assuming that this ratio
is representative of the entire area sampled in 2014, the overall occupancy estimate based on the
best model was 39.3 percent (Table 5.2-2).
Based on the results of the third-best model, with constant detectability between corridors, the
estimated occupancy for shallow-water habitats was 21.5 percent (95 percent C.I. = 7.4–49.5
percent) in the Gold Creek Corridor and 4.9 percent (95 percent C.I. = 1.4–16.2 percent) in the
Denali corridors. The estimated occupancy for deep-water habitats was 68.2 percent (95 percent
C.I. = 32.5–90.5 percent; Table 5.2-3) for the Gold Creek Corridor and 28.7 percent (95 percent
C.I. = 12.0–54.3 percent) in the Denali corridors. The sample included 21 shallow-water habitats
(55.3 percent) and 17 deep-water habitats (44.7 percent) in the Gold Creek Corridor and 37
shallow-water habitats (56.1 percent) and 29 deep-water habitats (43.9 percent) in the Denali
corridors. Assuming that these ratios are representative of the entire area of both the Gold Creek
and Denali corridors, the overall occupancy estimates based on the third-best model is 40.8
percent in the Gold Creek Corridor and 11.3 percent in the Denali corridors (Table 5.2-3).
5.2.2. Habitat Associations
5.2.2.1. 2013 Sampling
Occupancy modeling was the primary tool to assess habitat associations with breeding male
wood frogs and water depth was the most important habitat variable. Frogs were detected at a
total of 13 of 42 (31.0 percent) locations with shallow water (≤1.5 m) and 34 of 48 (70.8 percent)
locations with deep water (>1.5 m). The remaining habitat variables were summarized by
locations where wood frogs were detected, not detected, and across all sampling locations (Table
5.2-4). The only other variable that exhibited a significant association was dissolved oxygen
(mg/L), with lower levels being found where frogs were detected (Table 5.2-4).
5.2.2.2. 2014 Sampling
Occupancy modeling was the primary tool to assess habitat associations with breeding male
wood frogs and water depth was the most important habitat variable, as in 2013. Frogs were
detected at a total of 5 of 58 (8.6 percent) locations with shallow water (≤1.5 m) and 16 of 46
(34.8 percent) locations with deep water (>1.5 m). The remaining habitat variables were
summarized by locations where wood frogs were detected, not detected, and across all sampling
locations (Table 5.2-5). The only other associations of significance were water-body type (more
frogs detected at small ponds with emergent vegetation than other water-body types); dissolved
oxygen (higher levels were found where frogs were detected, which was the opposite of the
pattern observed in 2013); and specific EC (lower levels were found where frogs were detected;
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 15 October 2015
Table 5.2-5). Although substrate type and emergent vegetation type did not exhibit statistically
significant differences in wood frog presence (P > 0.05), almost all detections were in areas with
emergent sedges (90.5 percent) and all detections were in areas with organic substrates (Table
5.2-5).
5.3. Acoustic Monitoring
5.3.1. 2013 Sampling
Acoustic recordings from the five monitors provided a sample of 2,015 5-min intervals that were
used to quantify when frogs were heard calling. Calling activity varied by date and time of day
(Figure 5.3-1). The results demonstrated that the surveys were well-timed to capture the peak of
calling activity in the study area; frogs were calling when the acoustic monitors were deployed
on May 31 and calling activity was declining by the end of the survey period on June 8 (Figure
5.3-1 [top]). A very strong diurnal pattern of calling activity was evident. Calling activity peaked
near 01:00, then activity dropping dramatically to a low early in the morning (05:00) and
increased throughout the remainder of the day (Figure 5.3-1 [bottom]).
Based on the time-specific results from the acoustic monitors, the site visits should have had a
detectability of 60.8 percent, which was essentially identical to the estimate of 60.6 percent from
the occupancy modeling. This concurrence provides additional evidence that the occupancy
modeling provided a reasonable estimate of detectability and indicates that occupancy rates were
adjusted appropriately.
5.3.2. 2014 Sampling
Acoustic recordings from the five monitors provided a sample of 442 5-min intervals that were
used to quantify when frogs were heard calling. Unfortunately, four monitors experienced battery
malfunctions during most of the sampling period, substantially reducing the overall sample size
in 2014. As in 2013, calling activity varied by date and time of day (Figure 5.3-2). Although the
data on seasonal changes in calling rates were sparse, the results demonstrated that the surveys
were well-timed to capture the peak of calling activity in the study area; frogs were calling when
the acoustic monitors were deployed on May 20 and calling activity declined by the end of the
survey period on May 29 (Figure 5.3-2 [top]). As in 2013, a strong diurnal pattern of calling
activity was evident, with the lowest rates of calling occurring in the morning (05:00–10:00;
Figure 5.3-2 [bottom]) when air temperatures were low (Figure 5.3-3), followed by increasing
calling rates during the late morning and early afternoon (Figure 5.3-2 [bottom]) as air
temperatures increased (Figure 5.3-3). Because of the small sample size of recordings obtained
in 2014, the acoustic monitoring results were not used to calculate detectability.
5.4. Chytrid Fungus Bioassay
Swab samples collected from seven frogs captured opportunistically in 2013 were sent to the
USGS Reston Molecular and Environmental Microbiology Laboratory in Reston, Virginia, and
tested for the presence of chytridiomycosis (Bd) using standard qPCR protocols (Boyle et al.
2004). All seven samples tested negative for Bd. No samples were collected in 2014 (see Section
4.4.1).
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 16 October 2015
6. DISCUSSION
Amphibian populations appear to have been declining worldwide for several decades (Blaustein
and Wake 1990; McCallum 2007), leading to elevated levels of concern about the conservation
status of a large number of amphibian species. Although populations appear to be healthy in
Alaska (Gotthardt 2004, 2005), concern has been expressed about the conservation status of
wood frogs in Alaska (ADF&G 2006). Because amphibians were not included in the original
Alaska Power Authority Susitna Hydroelectric Project (APA Project) environmental study
program in the 1980s, information on the occurrence of wood frogs in the upper Susitna drainage
was lacking and their status in the study area was unknown at the time this study began.
6.1. Distribution and Habitat Use
A review of the literature shows that wood frogs are widely distributed throughout northern
North America and that, in Alaska, they occur from Southeast Alaska throughout central Alaska
to the crest of the Brooks Range (MacDonald 2010). Closer to the study area, they have been
documented in Denali National Park and Preserve, near Healy, and in the lower Susitna drainage
(Cook and MacDonald 2003; Anderson 2004; Gotthardt 2004, 2005; Hokit and Brown 2006).
Wood frogs were widely distributed throughout the areas sampled in 2013 and throughout the
Gold Creek Corridor in 2014. The distribution of frogs at higher elevation sites in the Denali
East and West corridor options was more dispersed, however. Wood frogs may be reaching their
elevational limits in this region because of the limited time available for breeding. The highest
potential sampling sites (approximately 900–1,100 m [2,953–3,609 ft] above sea level) were still
covered by snow and ice during the 2014 survey period and thus were unavailable for sampling.
Wood frogs occurred in a variety of habitats sampled in 2013 and 2014, ranging from alpine
tundra to forested wetlands (see photographs in Appendix B). Wood frogs are known to inhabit
diverse vegetation communities in Alaska, including tundra, open forests, grassy meadows, and
muskeg (MacDonald 2010). Not surprisingly, the habitat associations of wood frogs are diverse,
so a summary of known habitat associations is presented below and related to the findings of this
study and other similar studies.
Water-body types in the study area ranged from those having adequate water to sample, but
insufficient water depth to allow frog larvae to metamorphose (i.e., the ponds or wetlands would
dry out too early in the season) to deep-water lakes. More frogs were detected at small ponds
with emergent vegetation than at other water-body types in 2014, likely because emergent
vegetation is important for the attachment of egg masses (France 1997).
Water depth was the most important habitat factor analyzed in this study in 2013 and 2014,
which was consistent with the results of a similar study in southwestern Alaska, in which water
depth was an important habitat factor (PLP 2011). In both studies, calling male frogs were
detected more frequently in habitats where water was deeper than 1.5 m.
Water depth may be important because deeper water bodies retain water and often maintain more
consistent water-quality characteristics during the egg and larval growth stages (Knapp et al.
2003). In Denali National Park, Hokit and Brown (2006) found that wood frogs had the highest
breeding activity (defined as eggs or larvae) in sites with 51 to 75 percent of the site < 50 cm (1.6
ft) deep, but with a maximum depth of 1 to 2 m (3.3 to 6.6 ft). Differences in sampling methods,
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 17 October 2015
sampling times, and characterization of water body depths, however, make direct comparison of
their results with this study difficult. Water depth may be one of many factors influencing where
wood frogs choose to breed, judging from the findings of Herreid and Kinney (1966), in which
96 percent of wood frog eggs and larvae died before reaching metamorphosis because of lack of
fertilization, freezing, desiccation of eggs, temperature-related abnormalities, and predation.
Hibernation habitat (herbaceous, low shrub, and tall shrub vegetation within a 50-m radius of the
shoreline) was not associated with frog detectability in this study in 2013 or 2014, in contrast to
the results reported by PLP (2011), in which wood frog occupancy increased as surrounding
hibernation habitat increased. Increased availability of vegetation that provides suitable habitat
for hibernation within 50 m of breeding ponds may be an important factor influencing occupancy
of water bodies in some areas, although seasonal movements of wood frogs up to 300 m or more
away from breeding ponds have been documented in Maine (Vasconcelos and Calhoun 2004,
Baldwin et al. 2006). The PLP (2011) study was conducted in a tundra area with much less tree
cover than in this study area. Differences in habitat occupancy and vegetative cover may help to
explain this difference between studies.
Emergent and aquatic vegetation in water bodies provides a substrate for frog egg masses and
escape cover from aquatic predators, as well as helping to increase dissolved oxygen in the water
(France 1997; Babbitt and Tanner 1998). Although the extent of emergent vegetation did not
differ significantly between ponds with and without frog detections (P = 0.10) in 2014, frogs
were never detected at locations without emergent vegetation and were nearly always detected
(90.5 percent) at locations with emergent sedges. Dissolved-oxygen levels (8.53 mg/L) were
similar between sites occupied and those not occupied by frogs in 2013, whereas in 2014 they
were higher at occupied (8.63 mg/L) vs. unoccupied (7.28 mg/L) sites. The levels in this study
were similar to those observed in a study in Southeast Alaska (approximately 9.0 mg/L;
Carstensen et al. 2003) and were within the range of mean values from new (4.9 mg/L) and old
(10.5 mg/L) beaver ponds in Alberta (Stevens et al. 2006). Increased concentrations of dissolved
oxygen were thought to be important in the latter study because they were correlated with
enhanced larval growth rates of wood frogs in old beaver ponds, although the authors cautioned
that this may have been an artifact of other landscape features (Stevens et al. 2006).
Other aspects of water quality such as pH may be important for breeding-site selection by wood
frogs. A study in Quebec reported that egg mass density and hatching success were negatively
correlated with pH, although hatching success was still fairly high (47 and 80 percent in ponds
with pH of 4.3 and 4.7, respectively; Gascon and Planas 1986). Another study near Juneau,
Alaska, measured pH levels ranging from 4.5 to 5.5 in ponds where larval wood frogs were
present (Carstensen et al. 2003). New and old beaver ponds in Alberta containing wood frogs
had pH levels of 7.6 and 7.8, respectively (Stevens et al. 2006). The pH values in the study were
very consistent throughout the sampling locations (5.73 at occupied sites and 5.72 at unoccupied
sites in 2013; 6.55 at occupied sites and 6.51 at unoccupied sites in 2014), within the range of
other studies where wood frogs bred successfully.
Electrical conductivity, a measure of water quality related to salinity, was found to be
statistically significant in this study in the 2014 results, but these differences evidently were not
biologically meaningful. The difference between sites where frogs were detected and were not
detected was <1 uS/cm, and it is thought that conductivity should not limit tadpole presence at
values <3,000 uS/cm (Smith et al. 2006).
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 18 October 2015
Other habitat variables measured in this study did not show clear relationships with frog
occupancy, including beaver activity, substrate, ice cover, and water temperature. Fish are
known predators of frogs (Hecnar and McCloskey 1997), but data on fish presence and
distribution in the water bodies and wetlands sampled in this study were not available for
analysis.
6.2. Occupancy Modeling
Accurate habitat occupancy estimates are adjusted for the detectability of organisms in the
environment. Detectability in this study was high in 2013, at 60.6 percent from the best model
and 60.8 percent from the acoustic monitors (see Section 6.3 below). Detectability was variable
in 2014, at 56.6 percent in the Gold Creek Corridor and 16.0 percent in the Denali corridors from
the best model, or at 54.6 percent overall from the third-best model. The estimated detectability
in a study in southwestern Alaska (26.6 percent; PLP 2011) was lower than this study and may
have resulted from differences in frog density, habitat characteristics, survey conditions, or the
timing of surveys between studies.
The high detectability in this study during 2013 indicates a robust study design: if frogs were
present at a sampling site, the study team would detect them 60.6 percent of the time with one
visit, 84.5 percent of the time with two visits, and 93.9 percent of the time with three visits
(based on the best model). Similarly, in 2014, if frogs were present at a sampling site the study
team would detect calls in the Gold Creek and Denali corridors, respectively, 56.6 and 16.0
percent of the time with one visit, 81.2 and 29.4 percent of the time with two visits, and 91.8 and
40.7 percent of the time with three visits (based on the best model). Based on the third-best
model, frogs would be detected 54.6 percent of the time with one visit, 79.4 percent of the time
with two visits, and 90.6 percent of the time with three visits.
The best models of frog occupancy in this study in 2013 and 2014 contained only the variable
water depth, with deeper water types having higher occupancy. The estimated occupancy in 2013
for shallow-water habitats (36.8 percent), deep-water habitats (81.8 percent), and all locations
(63.4 percent) suggest a widespread distribution of frogs in the areas surveyed in 2013 (dam and
camp area, reservoir inundation zone, Chulitna Corridor, part of the Denali corridors). In the
Gold Creek and Denali corridors in 2014, respectively, the estimated occupancy was 21.5 and
4.9 percent in shallow-water habitats, 68.2 and 28.7 percent in deep-water habitats, and 40.8
percent and 11.3 percent for all locations. The third-best model suggested a widespread
distribution of frogs in the Gold Creek Corridor and a more limited distribution in the Denali
corridors.
The study team detected fewer frogs in 2014 in the Denali corridors than in the Gold Creek
Corridor. Occupancy modeling provided two different interpretations of the 2014 data, both of
which had statistical support but for which occupancy estimates differed. The best model had
very different detection rates between corridors (perhaps because of differences in the timing of
surveys), resulting in occupancy rates that just varied with water depth. The third-best model had
constant detectability between corridors but different occupancy rates, both by water depth and
corridor (perhaps because of differences in habitats by corridor). Although it is not possible to
reach a definitive conclusion about which interpretation is correct because of the limited number
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 19 October 2015
of detections in the Denali corridors in 2014, it is clear that water depth is very influential on the
occupancy rate of wood frogs in this study.
Few studies have estimated occupancy rates of wood frogs in Alaska. The naïve occupancy rate
in Denali National Park and Preserve was estimated at 45 percent (Hokit and Brown 2006),
which was generally similar to an adjusted occupancy estimate of 49.5 percent in southwest
Alaska (PLP 2011), although adjustment of the Denali Park estimate would likely have resulted
in a higher occupancy rate. In comparison, the adjusted overall occupancy rates in this study
ranged from 39.3–40.8 (2014) to 63.4 percent (2013), bracketing the results from those other
studies in Alaska.
6.3. Acoustic Monitoring
The use of acoustic monitoring devices allowed the study team to collect information to
characterize the calling activity of breeding male wood frogs throughout the survey period and
throughout all hours of the day. Frogs called throughout the survey period (May 30–June 9,
2013; May 20–29, 2014) and incidental observations by other wildlife field crews noted calling
frogs (May 28–June 14, 2013, and May 26–June 14, 2014). These results indicated that the
auditory surveys were well-timed, at least for locations at lower elevations in 2013 (dam and
camp area, reservoir zone, and most of the Chulitna Corridor) and 2014 (Gold Creek Corridor).
Locations at higher elevations in the Denali corridors, however, still were snow-covered and
many water types were either frozen or just beginning to thaw during the survey period in each
year, although much more water was open in 2014 than in the unusually late spring in 2013.
Supplemental data from acoustic monitors deployed in both corridors between May 30 and June
8, 2014, showed a decline in calling activity for detectors in the lower-elevation sites (i.e., Gold
Creek Corridor) and a decreasing or variable rate of calling activity in the Denali corridors. Egg
masses in the early stages of development (n = 5 locations), along with young tadpoles (n = 1)
were discovered in the Gold Creek Corridor, also suggesting that the auditory surveys were well-
timed to detect calling frogs in that corridor.
The information for the higher-elevation sites in 2014 (Denali corridor options) is less clear,
however, for several reasons: (1) the two acoustic detectors deployed there were installed at later
dates (May 24 and 26) than in the Gold Creek Corridor because it was necessary to sample the
lower-elevation sites first and allow the higher-elevation sites to thaw; (2) the detectors collected
limited data during the auditory survey period because of battery failures, and most data were
obtained during May 30–June 8, after the auditory surveys were completed; and (3) the two
detectors in the Denali corridors produced disparate results, in that calling frequency decreased
over time at one detector and was variable over time at the other, making it unclear whether the
Denali corridors were sampled when most frogs were calling.
Diurnal patterns of frog calling activity in 2013 showed a pattern of high calling rates throughout
the late morning and afternoon, with peak calling activity occurring between 01:00 and 02:00.
Frog calling activity in 2014 showed a diurnal pattern of high calling rates throughout the early
afternoon and evening, although all of the data for the period of 00:00–15:00 came from a single
detector located in the Gold Creek Corridor. Additional data obtained after the auditory surveys
finished in 2014 (May 30–June 8) reinforced this pattern, which was similar in overall shape to
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 20 October 2015
that from 2013, although 2014 values were generally at lower levels (lower percentages). The
lowest rates of calling activity in 2014 correlated with some of the lowest temperatures of the
day, which occurred before acoustic surveys began.
The auditory survey sampling times between approximately 09:00 and 20:00 in 2013 mainly fell
within the period of high calling activity, helping to explain the high detectability of the surveys
that year. The survey times between approximately 09:30 and 19:30 in 2014 fell within both low
and high periods of calling activity observed in 2014, with later sampling times having a higher
percentage of calling frequency. The low frequency of frogs calling in the morning and early
afternoon in 2014 may help to explain the lower detectability values obtained in 2014 than in
2013.
An additional use of the 2013 acoustic monitoring data was to calculate the detectability (60.8
percent) of frogs calling when the study team actually sampled and compare that to the estimate
from occupancy modeling (60.6 percent). Concordance between these results provided strong
evidence that the occupancy modeling provided a reasonable estimate of detectability and that
the occupancy rates were adjusted appropriately. This concordance is key to producing
meaningful habitat occupancy results for eventual use in estimating the potential habitat loss and
alteration that may occur from development of the Project.
6.4. Chytrid Fungus
Bd is a chytrid fungus that causes the disease chytridiomycosis in amphibians. Since it was first
discovered in amphibians in 1998, it has devastated amphibian populations around the world,
including in North America (Adams et al. 2007, Olson et al. 2013). Bd is sometimes a nonlethal
parasite and some amphibian species and some populations of susceptible species are known to
survive infection. The fungus is widespread and ranges from lowland forests to cold mountain
tops, and is typically associated with host mortality in high altitude environments and during
winter, with greater pathogenicity at lower temperatures. Bd is believed to spread mainly through
contact between infected frogs or with infected water. USFWS originally requested that
opportunistic sampling for the presence of Bd be added to the Study Plan out of concern for the
potential spread of the fungus through increased road access in the study area.
Wood frogs have been identified as a species susceptible to infection by Bd, and it was first
detected in Alaska in a dead wood frog found in the Kenai National Wildlife Refuge in 2002
(Reeves and Green 2006, Reeves 2008). Another positive detection of Bd occurred near Dyea in
Southeast Alaska in 2006 and was associated with the apparent die-off of western (boreal) toads
in that region (Juneau Empire, May 21, 2006). Bd was documented in boreal toads (Bufo boreas)
and red-legged frogs (Rana aurora) in another study in western Canada and Southeast Alaska
(Adams et al. 2007). Although Bd was not detected in this study, the small sample size of swabs
obtained in this study is considered inadequate to confirm its absence unequivocally.
7. CONCLUSION
During 2013–2014, AEA completed auditory surveys and acoustic monitoring of calling male
wood frogs to document the distribution, habitat use, and occupancy of wood frogs in water
bodies throughout the study area during the spring breeding season. The status of wood frogs in
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 21 October 2015
the Project area was unknown prior to this study and few studies have established occupancy
rates of wood frogs in Alaska. A total of 90 randomly selected wetlands and water bodies were
surveyed for the presence of wood frogs in 2013 and 104 wetlands and waterbodies in 2014.
Frogs were found to be widely distributed in the areas surveyed over a variety of habitat types
from tundra to forested wetlands. The field work, data collection, data analysis, and reporting for
the Wood Frog Study successfully met the first, second, and fourth study objectives in the
FERC-approved Study Plan; the third objective will be addressed in the draft license application
for the Project, using information obtained in this study. The results of the Wood Frog Study are
reported herein and earlier by AEA (ABR 2014a, 2014b, 2014c). With this report, AEA has now
completed Study 10.18, Wood Frog Occupancy and Habitat Use.
8. LITERATURE CITED
ABR. 2014a. Wood frog occupancy and habitat use, Study Plan Section 10.18; Initial Study
Report, Part A: Sections 1–6, 8–10; Susitna–Watana Hydroelectric Project (FERC No.
14241). Prepared for Alaska Energy Authority, Anchorage, by ABR, Inc.—
Environmental Research & Services, Forest Grove, Oregon, and Fairbanks, Alaska.
21 pp. + appendices.
ABR. 2014b. Wood frog occupancy and habitat use, Study Plan Section 10.18; Initial Study
Report, Part B: Supplemental information (and errata) to Part A (February 3, 2014 draft
Initial Study Report); Susitna–Watana Hydroelectric Project (FERC No. 14241).
Prepared for Alaska Energy Authority, Anchorage, by ABR, Inc.—Environmental
Research & Services, Forest Grove, Oregon, and Fairbanks, Alaska. 1 p.
ABR. 2014c. Wood frog occupancy and habitat use, Study Plan Section 10.18; Initial Study
Report, Part C: Executive summary and Section 7; Susitna–Watana Hydroelectric Project
(FERC No. 14241). Prepared for Alaska Energy Authority, Anchorage, by ABR, Inc.—
Environmental Research & Services, Forest Grove, Oregon, and Fairbanks, Alaska. 3 pp.
+ appendix.
Adams, M. J., S. Galvan, D. Reinitz, R. A. Cole, S. Pyare, M. Hahr, and P. Govindarajulu. 2007.
Incidence of the fungus Batrachochytrium dendrobatidis in amphibian populations along
the Northwest Coast of North America. Herpetological Review 38: 430–431.
ADF&G (Alaska Department of Fish and Game). 2006. Our wealth maintained: A strategy for
conserving Alaska’s diverse wildlife and fish resources. Alaska Department of Fish and
Game Juneau. 824 pp.
AEA (Alaska Energy Authority). 2012. Revised Study Plan: Susitna-Watana Hydroelectric
Project FERC Project No. 14241. December 2012. Prepared for the Federal Energy
Regulatory Commission by the Alaska Energy Authority, Anchorage,
Alaska.http://www.susitna-watanahydro.org/study-plan.
AKNHP (Alaska Natural Heritage Program). 2008. Alaska Wood Frog Monitoring Project
results, 2002–2008. Available online: http://aknhp.uaa.alaska.edu/zoology/citizen-
science/alaska-wood-frog-monitoring/results-2002-2008/ (accessed October 2012).
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 22 October 2015
Anderson, B. C. 2004. An opportunistic amphibian inventory in Alaska’s national parks, 2001–
2003. Final report, National Park Service, Alaska Region Survey and Inventory Program,
Anchorage. 44 pp.
Babbitt, K. J., and G. W. Tanner. 1998. Effects of cover and predator size on survival and
development of Rana utricularia tadpoles. Oecologia 114: 258–262.
Baldwin, R. F., A. J. K Calhoun, and P. G. deMaynadier. 2006. Conservation planning for
amphibian species with complex habitat requirements: a case study using movements and
habitat selection of the wood frog Rana sylvatica. Journal of Herpetology 40: 442–453.
Blaustein, A. R., and D. B. Wake. 1990. Declining amphibian populations: a global
phenomenon? Trends in Ecology and Evolution 5: 203–204.
Boyle, D. G., D. B. Boyle, V. Olsen, J. A. T. Morgan, and A. D. Hyatt. 2004. Rapid quantitative
detection of chytridiomycosis (Batrachochytrium dendrobatidis) in amphibian samples
using real-time Taqman PCR assay. Diseases of Aquatic Organisms. 60: 141–148.
Burnham, K. P., and D. R. Anderson. 2002. Model Selection and Multimodel Inference: A
Practical Theoretic Approach. Second edition. New York: Springer-Verlag.
Carstensen, R., M. Willson, and R. Armstrong. 2003. Habitat use of amphibians in northern
southeast Alaska. Unpublished report to Alaska Department of Fish and Game, Juneau.
Cook, J. A., and S. O. MacDonald. 2003. Mammal inventory of Alaska’s national parks and
preserves: Denali National Park and Preserve. 2002 annual report for National Park
Service, Alaska Region Survey and Inventory Program, Anchorage, by Idaho State
University, Pocatello. 24 pp.
France, R. L. 1997. The importance of beaver lodges in structuring littoral communities in boreal
headwater lakes. Canadian Journal of Zoology 75: 1009–1013.
Gascon, C., and D. Planas. 1986. Spring pond water chemistry and the reproduction of the wood
frog, Rana sylvatica. Canadian Journal of Zoology 64: 543–550.
Gotthardt, T. 2004. Monitoring the distribution of amphibians in the Cook Inlet watershed: 2003
final report. Alaska Natural Heritage Program, University of Alaska, Anchorage.
Gotthardt, T. 2005. Wood frog conservation status report. Alaska Natural Heritage Program,
University of Alaska, Anchorage.
Hecnar, S. J., and R. T. McCloskey. 1997. The effects of predatory fish on amphibian species
richness and distribution. Biological Conservation 79: 123–131.
Herreid, C., and S. Kinney. 1966. Survival of Alaskan wood frog (Rana sylvatica) larvae.
Ecology 47: 1039–1041.
Hines, J. E. 2006. PRESENCE — Software to estimate patch occupancy and related parameters.
U.S. Geological Survey, Patuxent Wildlife Research Center. Available online:
http://www.mbr-pwrc.usgs.gov/software/presence.html (accessed August 2013).
Hokit, D. G., and A. Brown. 2006. Distribution patterns of wood frogs (Rana sylvatica) in Denali
National Park. Northwestern Naturalist 87: 128–137.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 23 October 2015
Knapp, R. A., K. R. Matthews, H. K. Preisler, and R. Jellison. 2003. Developing probabilistic
models to predict amphibian site occupancy in a patchy landscape. Ecological
Applications 13: 1069–1082.
MacDonald, S. O. 2010. The amphibians and reptiles of Alaska: a field handbook. Version 2.0.
University of Alaska Museum, Fairbanks, and Museum of Southwestern Biology,
Albuquerque, NM. Available online: http://aknhp.uaa.alaska.edu/wp-content/uploads/
2011/02/Herps-of-Alaska-Handbook-Final-Version-2-reduced.pdf (accessed March
2012).
MacKenzie D. I., J. D. Nichols, G. B. Lachman, S. Droegem J. A. Royle, and C. A. Langtimm.
2002. Estimating site occupancy rates when detection probabilities are less than one.
Ecology 3: 2248–2255.
MacKenzie, D. I., and J. A. Royle. 2005. Designing occupancy studies: general advice and
allocating survey effort. Journal of Applied Ecology 42: 1105–1114.
McCallum, M. L. 2007. Amphibian decline or extinction? Current declines dwarf background
extinction rate. Journal of Herpetology 41: 483–491.
Olson, D. H., D. M. Aanensen, K. L. Ronnenberg, C. I. Powell, S. F. Walker, J. Bielby, T. W. J.
Garner, G. Weaver, The Bd Mapping Group, and M. C. Fisher. 2013. Mapping the global
emergence of Batrachochytrium dendrobatidis, the amphibian chytrid fungus. PLoS ONE
8(2): e56802. Available online: http://www.plosone.org/article/info:doi/10.1371/
journal.pone.0056802 (accessed September 2013).
PLP (Pebble Limited Partnership). 2011. Pebble Project Environmental Baseline Document,
2004 through 2008. Chapter 16.12: Wood Frog—Mine Study Area. Pebble Limited
Partnership, Anchorage. Available online: http://www.pebbleresearch.com/ (accessed
August 2013).
Reeves, M. K., and D. E. Green. 2006. Rana sylvatica wood frog chytridiomycosis.
Herpetological Review 37: 450.
Reeves, M. K. 2008. Batrachochytrium dendrobatidis in wood frogs (Rana sylvatica) from three
national wildlife refuges in Alaska, USA. Herpetological Review 39: 68–70.
Smith, M. J., E. Sabine, G. Schreiber, M. P. Scroggie, M. Hohout, K. Ough, J. Potts, R. Lennie,
D. Turnbull, C. Jin, and T. Clancy. 2006. Associations between anuran tadpoles and
salinity in a landscape mosaic of wetlands impacted by secondary salinisation.
Freshwater Biology 52: 75–84.
Stevens, C. E., C. A. Paszkowski, and G. J. Scrimgeour. 2006. Older is better: beaver ponds on
boreal streams as breeding habitat for the wood frog. Journal of Wildlife Management 70:
1360–1371.
Vasconcelos, D., and A. J. K. Calhoun. 2004. Movement patterns of adult and juvenile Rana
sylvatica (LeConte) and Ambystoma maculatum (Shaw) in three restored seasonal pools
in Maine. Journal of Herpetology 38: 551–561.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 24 October 2015
9. TABLES
Table 5.1-1. Number of Frog-survey Visits to Water Bodies and Wetlands in the 2013 Study Area.
Location First Visit Second Visit Third Visit
Dam/Camp Area and
Reservoir Zone
Water body 28 9 0
Wetland 21 9 0
Total 49 18 0
Corridors
Water body 28 21 4
Wetland 13 11 1
Total 41 32 5
Grand Total 90 50 5
Table 5.1-2. Frog Detection in Shallow- and Deep-water Habitats in 2013.
First Visit Second Visit Third Visit Overall
Location Detected
Not
Detected Detected
Not
Detected Detected
Not
Detected Detected
Not
Detected
Shallow water
(<1.5 m)a 8 34 3 29 2 3 13 29
Deep water
(>1.5 m) 29b 19 5 13 – – 34 14
Total 37 53 8 42 2 3 47 43
Notes:
a. 1.5 m = 4.9 ft.
b. Two locations were included where egg masses were observed but no frog calls were detected.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 25 October 2015
Table 5.1-3. Number of Frog-survey Visits to Water Bodies and Wetlands in the 2014 Study Area.
Location First Visit Second Visit Third Visit
Gold Creek Corridor
Water body 18 14 0
Wetland 20 15 0
Total 38 29 0
Denali Corridor Options
Water body 54 49 0
Wetland 12 12 3
Total 66 61 3
Grand Total 104 90 3
Table 5.1-4. Frog Detection in Shallow- and Deep-water Habitats in 2014.
First Visit Second Visit Third Visit Overall
Corridor
Location Detected
Not
Detected Detected
Not
Detected Detected
Not
Detected
Detected
Not
Detected
Gold Creek Shallow water
(<1.5 m)a 1 20 2 18 0 0 3 18
Deep water
(>1.5 m) 8 9 2 7 0 0 10 7
Total 9 29 4 25 0 0 13 25
Denali Shallow water
(<1.5 m)a 1 36 1 35 0 3 2 35
Deep water
(>1.5 m) 4 25 2 23 0 0 6 23
Total 5 61 3 58 0 3 8 58
Total Total 14 90 7 83 0 3 21 83
Notes:
a. 1.5 m = 4.9 ft.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 26 October 2015
Table 5.1-5. Occupancy-model Selection Results for Presence of Wood Frogs in 2013.
Modela –2*LLb Kc AICcd ΔAICce ωif
Ψ (Water Depth), p (.)g 162.35 3 168.63 0.00 0.348
Ψ (Water Depth, Habitat), p (.) 161.71 4 170.18 1.55 0.160
Ψ (Water Depth, Water Type), p (.) 161.97 4 170.44 1.81 0.141
Ψ (Water Depth, Area), p (.) 162.00 4 170.47 1.84 0.139
Ψ (Water Depth, Water Type, Habitat), p (.) 161.51 5 172.22 3.59 0.058
Ψ (Water Depth, Area, Habitat), p (.) 161.61 5 172.32 3.69 0.055
Ψ (Water Depth, Area, Water Type), p (.) 161.84 5 172.55 3.92 0.049
Ψ (Global), p (.) 161.48 6 174.49 5.86 0.019
Ψ (Habitat), p (.) 168.78 3 175.06 6.43 0.014
Ψ (Water Type, Habitat), p (.) 168.53 4 177.00 8.37 0.005
Ψ (Area, Habitat), p (.) 168.75 4 177.22 8.59 0.005
Ψ (Area, Water Type, Habitat), p (.) 168.48 5 178.87 10.24 0.002
Ψ (Area), p (.) 172.59 3 179.19 10.56 0.002
Ψ (.), p (.) 175.18 2 179.32 10.69 0.002
Ψ (Area, Water Type), p (.) 171.93 4 180.40 11.77 0.001
Ψ (Water Type), p (.) 174.81 3 181.09 12.46 0.001
Notes:
a. Ψ = occupancy variable; p = detection probability; Water Depth = 1 if depth > 1.5 m (4.9 ft); Habitat = proportion of shoreline
containing hibernation habitat; Water Type = water body or wetland; and Area = dam, camp, and reservoir area or road corridors.
b. Negative 2 times the log-likelihood value.
c. Number of estimable parameters in the approximating model.
d. Akaike’s Information Criterion, corrected for small sample size.
e. Difference in value between the AICc of the current model and that of the best approximating model.
f. Akaike Weight = Probability that the current model (i) is the best approximating model in the candidate set.
g. p (.) indicates that detection probability was held constant across all locations in the model.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 27 October 2015
Table 5.1-6. Top Occupancy-model Selection Results for Presence of Wood Frogs in 2014.
Modela –2*LLb Kc AICcd ΔAICce ωif
Ψ (Water Depth), p (Corridor) 112.25 4 120.65 0 0.1645
Ψ (Water Depth, Habitat), p Corridor) 110.42 5 121.03 0.38 0.1360
Ψ (Corridor, Water Depth), p (.)g 112.72 4 121.12 0.47 0.1300
Ψ (Water Type, Water Depth, Habitat), p(Corridor) 108.55 6 121.42 0.77 0.1119
Ψ (Water Type, Water Depth), p (Corridor) 111.19 5 121.80 1.15 0.0925
Ψ (Corridor, Water Depth, Habitat), p (.) 111.70 5 122.31 1.66 0.0717
Ψ (Corridor, Water Type, Water Depth), p (.) 111.97 5 122.58 1.93 0.0627
Ψ (Corridor, Water Depth), p (Corridor) 112.25 5 122.86 2.21 0.0545
Ψ (Corridor, Water Depth, Habitat), p (Corridor) 110.36 6 123.23 2.58 0.0453
Ψ (Corridor, Water Type, Water Depth, Habitat), p (.) 110.53 6 123.40 2.75 0.0416
Ψ (Corridor, Water Type, Water Depth, Habitat), p (Corridor) 108.53 7 123.70 3.05 0.0358
Ψ (Corridor, Water Type, Water Depth), p (Corridor) 111.19 6 124.06 3.41 0.0299
Ψ (Water Depth), p (.) 120.24 3 126.48 5.83 0.0089
Ψ (Water Type, Water Depth), p (.) 119.97 4 128.37 7.72 0.0035
Ψ (Water Depth, Habitat), p (.) 120.17 4 128.57 7.92 0.0031
Ψ (Corridor, Water Type), p (.) 121.98 4 130.38 9.73 0.0013
Notes:
a. Ψ = occupancy variable; p = detection probability; Water Depth = 1 if depth > 1.5 m (4.9 ft); Habitat = proportion of shoreline
containing hibernation habitat; Water Type = water body or wetland; and Corridor = Gold Creek or Denali Corridors.
b. Negative 2 times the log-likelihood value.
c. Number of estimable parameters in the approximating model.
d. Akaike’s Information Criterion, corrected for small sample size.
e. Difference in value between the AICc of the current model and that of the best approximating model.
f. Akaike Weight = Probability that the current model (i) is the best approximating model in the candidate set.
g. p (.) indicates that detection probability was held constant across all locations in the model.
Table 5.2-1. Model Estimates of Wood Frog Occupancy and Detection Probability in 2013 from the Best Model.
Variable Estimate S.E. 95% C.I.
Occupancy
Shallow water (<1.5 m deep)a 0.368 0.095 0.208–0.565
Deep water (>1.5 m deep) 0.818 0.131 0.444–0.962
Overall b 0.634 n/a n/a
Detection Probability
Overall 0.606 0.129 0.348–0.816
Notes:
a. 1.5 m = 4.9 ft.
b. Occupancy based on weighted average of parameter estimates.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 28 October 2015
Table 5.2-2. Model Estimates of Wood Frog Occupancy and Detection Probability in 2014 from the Best Model.
Variable Estimate S.E. 95% C.I.
Occupancy
Shallow water (<1.5 m deep) a 0.179 0.093 0.059–0.431
Deep water (>1.5 m deep) 0.719 0.245 0.192–0.965
Overall b 0.393 n/a n/a
Detection Probability
Gold Creek Corridor 0.566 0.256 0.145–0.909
Denali Corridor 0.160 0.082 0.055–0.387
Notes:
a. 1.5 m = 4.9 ft.
b. Occupancy based on weighted average of parameter estimates.
Table 5.2-3. Model Estimates of Wood Frog Occupancy and Detection Probability in 2014 from the Third-best Model.
Corridor Variable Estimate S.E. 95% C.I.
Occupancy
Gold Creek Shallow water (<1.5 m deep) a 0.215 0.106 0.074-0.495
Deep water (>1.5 m deep) 0.682 0.165 0.325-0.905
Overall b 0.408 na na
Denali Shallow water (<1.5 m deep) a 0.049 0.031 0.014-0.162
Deep water (>1.5 m deep) 0.287 0.113 0.120-0.543
Overall b 0.113 n/a n/a
Detection Probability
Overall Overall 0.546 0.179 0.226-0.832
Notes:
a. 1.5 m = 4.9 ft.
b. Occupancy based on weighted average of parameter estimates.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 29 October 2015
Table 5.2-4. Habitat Characteristics of Water Bodies and Wetlands where Wood Frogs were Detected and Not Detected in 2013.
Habitat Type / Variable Description Wood Frog Detection a
P-value Detected Not Detected Overall
Water-body Structure
Water-body type (%) Big lakes (> 20 acres) 2.1 2.3 2.2 0.158 b
Small ponds w/o emergents 27.7 11.6 20.0
Small ponds w/ emergents 44.7 41.9 43.3
Seasonally flooded ponds 25.5 44.2 34.4
Beaver activity (%) No 91.3 76.7 84.3 0.157 b
Yes 8.7 23.3 15.7
Aquatic Habitat Characteristics
Emergent and submergent
vegetation (%) 22.6 (4.2) 32.7 (5.2) 27.5 (3.3) 0.132
Emergent vegetation (%) Grass 6.4 14.0 10.0 0.158 b
Sedge 80.9 62.8 72.2
None 12.8 23.3 17.8
Substrate (%) Boulder 4.3 2.3 3.3 0.179b
Gravel 0.0 7.0 3.3
Mud/silt 14.9 23.3 18.9
Organic 80.9 67.4 74.4
Aquatic Features
Ice cover (%) c 36.7 (5.6) 26.1 (5.0) 31.7 (3.8) 0.165
Water temperature (%) c 7.0 (0.6) 5.7 (0.8) 6.4 (0.5) 0.175
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 30 October 2015
Habitat Type / Variable Description Wood Frog Detection a
P-value Detected Not Detected Overall
Water depth (%) Shallow (≤ 1.5 m) 27.7 67.4 46.7 <0.001b
Deep (> 1.5 m) 72.3 32.6 53.3
Water Quality
Dissolved oxygen (%) c 64.77 (2.77) 70.63 (3.50) 67.57 (2.22) 0.193
Dissolved oxygen (mg/L) 7.96 (0.38) 9.16 (0.46) 8.53 (0.30) 0.047
Specific EC c 0.039 (0.006) 0.040 (0.008) 0.039 (0.005) 0.950
pH c 5.73 (0.10) 5.72 (0.12) 5.73 (0.07) 0.932
Terrestrial Habitat within 50-m Radius
Herbaceous (%) 18.0 (1.9) 26.4 (3.2) 22.0 (1.9) 0.029
Dwarf shrub (%) 12.7 (2.2) 11.4 (2.6) 12.1 (1.7) 0.709
Low shrub (%) 21.2 (2.1) 22.4 (2.4) 21.8 (0.6) 0.709
Tall shrub (%) 28.5 (2.3) 27.7 (3.6) 28.1 (2.1) 0.847
Trees (%) 19.0 (2.6) 12.8 (3.1) 16.4 (2.0) 0.130
Notes:
a. Parenthetical values in table cells indicate 1 S.E.
b. P-value from chi-square test (other P-values are from t-tests for two independent samples).
c. Measured on first visit.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 31 October 2015
Table 5.2-5. Habitat Characteristics of Water Bodies and Wetlands where Wood Frogs were Detected and Not Detected in 2014.
Habitat Type / Variable Description
Wood Frog Detection a
P-value
Detected Not Detected Overall
Water-body Structure
Water-body type (%) Big lakes (> 20 acres) 4.8 1.2 1.9 0.016 b
Small ponds w/o emergents 0 18.1 14.4
Small ponds w/ emergents 95.2 65.1 71.2
Seasonally flooded ponds 0 15.7 12.5
Beaver activity (%) No 76.2 68.7 70.2 0.501 b
Yes 23.8 31.3 29.8
Aquatic Habitat Characteristics
Emergent and submergent
vegetation (%) 16.0 (5.6) 21.2 (3.0) 20.1 (2.6) 0.097
Emergent vegetation (%) Grass 4.8 9.6 8.7
Sedge 90.5 67.5 72.1
Shrub 4.8 2.4 2.9
None 0 20.5 16.3
Substrate (%) Boulder 0 4.8 3.8 0.080 b
Gravel 0 1.2 1.0
Mud/silt 0 21.7 17.3
Organic 100.0 69.9 76.0
Sand 0 2.4 1.9
Aquatic Features
Ice cover (%)c 6.2 (4.4) 8.6 (2.3) 8.1 (2.0) 0.625
Water temperature (%)c 8.6 (0.8) 7.7 (0.4) 7.9 (0.3) 0.273
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 32 October 2015
Habitat Type / Variable Description
Wood Frog Detection a
P-value
Detected Not Detected Overall
Water depth (%) Shallow (≤ 1.5 m) 23.8 63.9 55.8 0.001 b
Deep (> 1.5 m) 76.2 36.1 44.2
Water Quality
Dissolved oxygen (%) c 74.3 (3.71) 61.58 (2.36) 64.1 (2.08) 0.006
Dissolved oxygen (mg/L) 8.63 (0.42) 7.28 (0.29) 7.55 (0.25) 0.011
Specific EC (uS/cm) c 0.042 (0.005) 0.058 (0.005) 0.055 (0.004) 0.023
pH c 6.55 (0.08) 6.51 (0.08) 6.53 (0.06) 0.761
Terrestrial Habitat within 50-m Radius
Herbaceous (%) 26.9 (3.8) 28.1 (1.9) 27.9 (1.7) 0.775
Dwarf shrub (%) 10.5 (2.3) 10.5 (1.5) 10.5 (1.2) 0.983
Low shrub (%) 23.6 (5.0) 23.9 (22) 23.8 (2.0) 0.956
Tall shrub (%) 33.0 (5.0) 30.7 (2.5) 31.1 (2.2) 0.679
Trees (%) 6.0 (3.4) 6.6 (1.5) 6.5 (1.4) 0.877
Notes:
a. Parenthetical values in table cells indicate 1 S.E.
b. P-value from chi-square test (other P-values are from t-tests for two independent samples).
c. Measured on first visit.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 33 October 2015
10. FIGURES
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 34 October 2015
Figure 3-1. Wood Frog Study Area for the Susitna–Watana Hydroelectric Project in 2013.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 35 October 2015
Figure 3-2. Wood Frog Study Area for the Susitna–Watana Hydroelectric Project in 2014.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 36 October 2015
Figure 5.1-1. Locations where Wood Frogs were Detected during Auditory Surveys, plus Incidental Observations from Other Wildlife Surveys, in 2013.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 37 October 2015
Figure 5.1-2. Locations where Wood Frogs were Detected during Auditory Surveys, plus Incidental Observations from Other Wildlife Surveys, in 2014.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 38 October 2015
Figure 5.3-1. Wood Frog Calling Activity by Date and Hour in 2013 (error bars depict 1 S.E.).
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 39 October 2015
Figure 5.3-2. Wood Frog Calling Activity by Date and Hour in 2014 (error bars depict 1 S.E.).
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Page 40 October 2015
Figure 5.3-3. Average Hourly Air Temperatures Measured at Acoustic Monitor Stations during May 20–29,
2014 (error bars depict 1 S.E.).
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix A - Page 1 October 2015
APPENDIX A: RECORDS OF ADDITIONAL CONSULTATION IN 2014
WITH USGS REGARDING SAMPLING PROTOCOL AND ANALYTICAL
METHOD FOR AMPHIBIAN CHYTRID FUNGUS.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix A - Page 2 October 2015
EMAIL RECORD
From: Todd Mabee tmabee@abrinc.com
Date: 4/4/14
To: Chestnut, Tara chestnut@usgs.gov
Hi Tara,
eDNA has been mentioned as an idea for sampling for Bd in Wood Frogs. Would this technique be an
appropriate method to sample for Bd for Wood frogs? If so, how many samples would we need to obtain
useful information? Do you have any idea on labs that process these samples and costs?
This may be easier to discuss on the phone, and if so, please feel free to suggest a time that would be
convenient for you next week.
Hope you are well, and thank you!
Todd J. Mabee
Senior Scientist/Research Coordinator
ABR Inc., Environmental Research & Services
P.O. Box 249 Forest Grove, Oregon 97116
Work: (503) 359-7525 ext 110
Mobile: (503) 537-7749
Fax: (503) 359-8875
www.abrinc.com
From: Chestnut, Tara chestnut@usgs.gov
Date: 4/8/14
To: Todd Mabee tmabee@abrinc.com
Hi Todd,
Hmm... I would need to know more about the study goals and objectives. I generally recommend against
sampling for Bd for the sake of sampling for it. We know it's widespread so we don't gain much
information by collecting samples without specific questions. Can you tell me more about the details?
Thanks,
Tara
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix A - Page 3 October 2015
From: Todd Mabee tmabee@abrinc.com
Date: 4/8/14
To: Chestnut, Tara chestnut@usgs.gov
Hi Tara,
The idea was suggested specifically for the SuWa study by Dave Tessler of ADFG to detect whether Bd
is present in this remote study area before any roads/transmission corridors are developed.
That's all the detail I have on this topic. Let me know if you want to discuss on the phone, might be
easiest to talk through ideas and potential study options?
Thank you!
Todd
From: Brian Lawhead <lawhead@abrinc.com>
Date: Apr 8, 2014
To: Todd Mabee, ABR
Thanks, Todd. Please call her and tell her that it was an idea suggested specifically for the SuWa study
by Dave Tessler of ADFG (whom she knows). And yes, it would be primarily intended to detect whether
Bd is present in this remote study area before any roads/transmission corridors are punched in.
An idea of sampling intensity and sample analysis costs would be useful.
Please document these interactions using the appropriate contact log forms.
Thanks!
Brian
From: Todd Mabee
Sent: Tuesday, April 08, 2014 2:15 PM
To: Brian Lawhead
Cc: Todd J. Mabee
Subject: Fwd: eDNA sampling for Bd?
Todd Mabee <tmabee@abrinc.com>
Date: Apr 8, 2014
to Tara Chestnut, USGS
Hi Tara,
The idea was suggested specifically for the SuWa study by Dave Tessler of ADFG to detect whether Bd
is present in this remote study area before any roads/transmission corridors are developed.
That's all the detail I have on this topic. Let me know if you want to discuss on the phone, might be
easiest to talk through ideas and potential study options?
Thank you!
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix A - Page 4 October 2015
From: Todd Mabee <tmabee@abrinc.com>
Date: Apr 11, 2014
To: Tara Chestnut, USGS
Hi Tara,
Can we set up a time to talk on the phone next week? I'd like to get your thoughts on this latest sampling
idea. I'm around most of the week, so just let me know if there would be a good time to talk.
Thank you!
From: Todd Mabee <tmabee@abrinc.com>
Date: Apr 14, 2014
To: Brian Lawhead, ABR
Hi Brian,
Had a good conversation with Tara, here's the summary. She thinks there is a high probability that Bd is
in the SuWa area (and throughout AK) and that trying to answer the question of "is Bd there" is not a very
useful one. Rather she thinks that it would be useful to understand how different strains of Bd affect the
native populations of frogs. To answer this question you need swabs of frogs and their actual skin
(therefore would need permit to collect frogs).
She thinks we could use the same approach as last year, except keep the frogs and send them to lab for
sampling (@~$125/frog). We only got 7 samples last year, so this wouldn't be terribly expensive.
FYI, eDNA does not allow you to detect strains of Bd, but only would answer the question of "presence".
It is also expensive and requires a lot of samples to have a high probability of detecting Bd.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix A - Page 5 October 2015
RECORD OF TELEPHONE
CONVERSATION
AEA Team Member Other Party
Name: Todd Mabee Name: Tara Chestnut
Organization: ABR Inc. Organization: USGS, Portland, OR
Study Area: Phone Number: 503 251 3283
Date: 14 April 2014 Time: 1500
Call Placed by: X AEA Team Other Party
Others on Call: none
Subject: Environmental DNA (eDNA) sampling for chytrid fungus
Discussion:
ADFG is interested in whether Bd is present in SuWa study area.
Tara Chestnut (USGS expert on wood frogs and Bd in AK) thinks one should assume Bd
is everywhere in Alaska, given its known occurrence in Denali SP, ANWR, & other
locations. Doesn’t think focusing on this question is very useful to advance the scientific
knowledge of Bd in AK. Recommended that wood frogs be swabbed and captured, so
that IF frogs tested positive for Bd, then the frogs could provide a strain of Bd that could
be used in future experiments. She felt this would help advance the understanding of Bd
in Alaska and elsewhere. Agreed that the small sample size obtained last year (7
samples) is inadequate to provide any level of certainty about the presence of Bd in the
study area.
eDNA technique can be used to determine if Bd is present but it can’t isolate strains. It
also is expensive: would need approximately 10 samples/wetland at a cost of
approximately $70/sample; therefore, $700/wetland. This expense would be cost
prohibitive at the scale of the SuWa project.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix B – Page 1 October 2015
APPENDIX B: PHOTOGRAPHS FROM FIELD SURVEYS IN 2013 AND
2014.
Example of Water Body at which Wood Frogs were Detected in the Chulitna Access Corridor in 2013.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix B – Page 2 October 2015
Example of Water Body in the Reservoir Inundation Zone at which Wood Frogs were Detected in 2013.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix B – Page 3 October 2015
Example of Water Body with Emergent Vegetation at which Wood Frogs were Detected in 2013.
Wood Frog Egg Mass, 2013.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix B – Page 4 October 2015
Example of Acoustic Monitoring Device used to Supplement Auditory Surveys in 2013.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix B – Page 5 October 2015
Example of Wetland at which Wood Frogs were Detected in the Gold Creek Access Corridor in 2014.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix B – Page 6 October 2015
Example of Water Body in the Gold Creek Access Corridor at which Wood Frogs were Detected in 2014.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix B – Page 7 October 2015
Example of Water Body with Emergent Vegetation in the Denali West Access Corridor at which Wood Frogs were
Detected in 2014.
Wood Frog Egg Mass, 2014.
STUDY COMPLETION REPORT WOOD FROG OCCUPANCY AND HABITAT USE (STUDY 10.18)
Susitna–Watana Hydroelectric Project Alaska Energy Authority
FERC Project No. 14241 Appendix B – Page 8 October 2015
Example of Acoustic Monitoring Device used to Supplement Auditory Surveys in 2014.
Wood Frog Adult, 2014.