Thanks to the ACA for this information:
Wolverine Harvest Summary from
Registered Traplines in Alberta,
1985 – 2011
The Alberta Conservation Association is a Delegated Administrative
Organization under Alberta’s Wildlife Act.
25% Post Consumer Fibre
When separated, both the binding and paper in this document are recyclable
Wolverine Harvest Summary from
Registered Traplines in Alberta,
1985 – 2011
Shevenell Webb1, Doug Manzer2, Robert Anderson2, and Michael Jokinen3
Alberta Conservation Association
1 101-9 Chippewa Rd, Sherwood Park, AB, T8A 6J7
2 Box 1139, Provincial Building, Blairmore, AB, T0K 0E0
3 817-4th Ave. South, Lethbridge, AB, T1J 0P6
i
Report Editors
DOUG MANZER GLENDA SAMUELSON
Alberta Conservation Association R.R. #2
Box 1139, Provincial Building Craven SK S0G 0W0
Blairmore AB T0K 0E0
Conservation Report Series Type
Technical
ISBN printed: 978-1-4601-0784-3
ISBN online: 978-1-4601-0785-0
Disclaimer:
This document is an independent report prepared by the Alberta Conservation
Association. The authors are solely responsible for the interpretations of data and
statements made within this report.
Reproduction and Availability:
This report and its contents may be reproduced in whole, or in part, provided that this
title page is included with such reproduction and/or appropriate acknowledgements are
provided to the authors and sponsors of this project.
Suggested Citation:
Webb, S., D. Manzer, R. Anderson, and M. Jokinen. 2013. Wolverine harvest summary
from registered traplines in Alberta, 1985-2011. Technical Report, T-2013-001, produced
by the Alberta Conservation Association, Sherwood Park, Alberta, Canada. 37 pp +
App.
Cover photo credit: David Fairless
Digital copies of conservation reports can be obtained from:
Alberta Conservation Association
#101, 9 Chippewa Rd
Sherwood Park AB T8A 6J7
Toll Free: 1-877-969-9091
Tel: (780) 410-1998
Fax: (780) 464-0990
Email:
info@ab-conservation.com
Website:
www.ab-conservation.com
ii
EXECUTIVE SUMMARY
Wolverines are among the most rare and least understood carnivores in the world. Lack
of data has classified the status of wolverines as Data Deficient in Alberta. Large home
ranges, low densities, and remote habitat use make wolverines challenging to study.
Long-term trapping records are valuable and can be used to assess distribution and
relative abundance. Previous studies have suggested that wolverine harvests and
potential populations may be declining, but more recent data have not been analyzed.
Thus, we completed a comprehensive update on long-term wolverine harvest trends in
Alberta.
We used wolverine harvest data collected from fur affidavits (1985-2011), fur
registrations (1989-2011), and Statistics Canada pelt export records (1971-2010). Our
results show that the total number of registered traplines harvesting a wolverine in a
year, and the average number of harvested wolverines, has increased since the early
1990s. The distribution of harvest data suggest that wolverines primarily occur in the
Rocky Mountains, Foothills, Boreal Forest, and Canadian Shield natural regions of
Alberta. When comparing two time periods (1989-1999; 2000-2011), wolverine harvests
have more than doubled in the Northwest Boreal (105%) and increased in the Northeast
Boreal (47%) and Foothills (33%), while declining in the Canadian Shield (40%) and
Rocky Mountains (32%). However, it is not clear whether these results are related to
changes in trapper effort, wolverine population, or other factors.
Some registered traplines in the Rocky Mountains had the highest wolverine harvest
densities, but wolverines have been harvested the most consistently, over the past 23
years, in the Boreal Forest, primarily north of 58 degrees latitude (WMU 530, 536, 539).
Despite a lack of consistent spring snow cover (thought to be a critical factor for
wolverine occurrence), wolverine harvest success on some traplines in the Boreal Forest
was moderately high, indicating that other factors may be responsible for wolverine
persistence.
We also found a high spatial and temporal overlap between lynx and wolverine
harvests, indicating that the number of lynx is possibly associated with the number of
wolverine harvested each year. However, these results may also be related to trapper
effort or other factors. We will be working with trappers to further investigate these
iii
trends as part of a larger project to determine wolverine occupancy and gene flow across
Alberta.
Key words: wolverine, furbearer, harvest, spring snow cover, traplines, Alberta
iv
ACKNOWLEDGEMENTS
We thank the trappers of Alberta for submitting fur registrations and affidavits over the
years, thus providing information on furbearer populations that is difficult to obtain
otherwise. We also thank Rob Corrigan and Jim Allen of Alberta Environment and
Sustainable Resource Development (ESRD) for access to the furbearer harvest database
and hard copy records. Nate Webb, Lonnie Bilyk, and Jacqui Gerwing (ESRD) also
provided important technical assistance and expertise on the furbearer database,
wolverine ecology, and trapping regulations. Rhonda Kindopp and John McKinnon
(Parks Canada) provided background information about trapping in Wood Buffalo
National Park. Jim Mitchell, Brian Bildson, and Bill Abercrombie (Alberta Trappers
Association), and Len Peleshok (ACA) were an important sounding board for advice
related to general furbearer ecology and harvest details. Jeff Copeland, Kevin
McKelvey, and Audrey Magoun provided fruitful discussions on wolverine distribution
in Alberta and provided the snow cover data used in our analysis. Lastly, we thank Sue
Peters (ACA) for her comments on this manuscript.
v
vi
TABLE OF CONTENTS
EXECUTIVE SUMMARY .................................................. .............................................. ii
ACKNOWLEDGEMENTS .................................................. ........................................... iv
TABLE OF CONTENTS .................................................. ............................................... vi
LIST OF FIGURES .................................................. .................................................. ...... vii
LIST OF TABLES .................................................. .................................................. ....... viii
LIST OF APPENDICES .................................................. ................................................ ix
1.0 INTRODUCTION .................................................. ............................................... 1
2.0 STUDY AREA .................................................. .................................................. .... 3
3.0 MATERIALS AND METHODS .................................................. ........................ 6
3.1 Data sources .................................................. .................................................. .... 6
3.2 Data entry/proofing .................................................. ......................................... 7
3.3. Distribution and harvest success .................................................. ................... 8
4.0 RESULTS .................................................. .................................................. .......... 11
4.1 Data quality .................................................. .................................................. ... 11
4.2 Wolverine distribution .................................................. .................................. 13
4.3 Wolverine harvest success .................................................. ............................ 22
5.0 DISCUSSION .................................................. .................................................. ... 29
6.0 LITERATURE CITED .................................................. ....................................... 34
7.0 APPENDICES .................................................. .................................................. .. 38
vii
LIST OF FIGURES
Figure 1. Distribution of registered traplines (RFMA, n=1667), wolverine harvests
(dot is centered on RFMA polygon), and natural regions in Alberta ........... 5
Figure 2. Wolverine (W) harvests reported on fur affidavits, fur registrations, and
Statistics Canada fur exports, and average wolverine pelt price adjusted
for inflation, in Alberta from 1972-2011 .................................................. ........ 12
Figure 3. Average wolverine harvest reported on fur affidavits, by natural region,
across different time intervals in Alberta .................................................. ..... 14
Figure 4. Wolverine harvest densities in relation to RFMA and WMU, as reported
on fur registrations from 1989-2011, and FWMIS locations from 2002-2011
.................................................. .................................................. .......................... 16
Figure 5. Number of years with a wolverine harvest (dot is centered on RFMA
polygon) in relation to Intact Forest Landscapes, from 1989-2012 in
Alberta. .................................................. .................................................. ............ 20
Figure 6. Distribution of female wolverine harvests reported on fur registrations
from 1989-2011, and total wolverine harvest from fur registrations and
affidavits from 1985-2011 (red dots centered on RFMA), relative to years
with spring snow cover from 2000-2006 in Alberta. ..................................... 21
Figure 7. Total wolverine pelts exported from Alberta from 1921-2010 .................... 23
Figure 8. Five-year average wolverine harvest, and total wolverine harvest
reported on fur affidavits, and wolverine pelt price for Alberta from 1972-
2011 .................................................. .................................................. .................. 23
Figure 9. Total and average number of traplines that reported a wolverine harvest
on fur affidavits from 1985-2011 in Alberta .................................................. . 25
Figure 10. Total and four-year average of wolverine harvests reported on fur
registrations from 1989-2012 in Alberta .................................................. ........ 26
Figure 11. Total and percent of wolverine harvests by gender, reported on registered
traplines by month, from 1989-2012 in Alberta ............................................. 26
Figure 12. Total lynx (n=30,438) and wolverine (n=1,185) harvests reported on fur
affidavits from 1985–2010 in Alberta .................................................. ............ 27
Figure 13. Total lynx (n=25,697) and wolverine (n=671) harvests reported on fur
registrations from 1989-2010 in Alberta .................................................. ........ 28
Figure 14. Total number of lynx (n=168,457) and wolverine (n=2,029) pelts exported
from 1971–2010 in Alberta, as reported by Statistics Canada (2012) .......... 28
viii
LIST OF TABLES
Table 1. Percent of total trapline area, total wolverine harvests, and harvest density,
summarized by natural region, for two time periods in Alberta. .................... 14
Table 2. Wolverine mortalities on registered traplines (RFMA) and non-trapline
locations (roadkill, trapping in Indian Reserve/Wood Buffalo National Park,
accidental harvest by resident trapper), reported on fur registrations from
1995-2012. .................................................. .................................................. ............. 17
ix
LIST OF APPENDICES
Appendix 1. Wildlife Management Units (WMU) and Natural Regions in Alberta. ..... 38
Appendix 2. Wolverine harvest density (# wolverine per km2) by WMU reported on
fur affidavits from 1985-1989 in Alberta. .................................................. ...... 39
Appendix 3. Wolverine harvest density (# wolverine per km2) by WMU reported on
fur affidavits from 1990-1994 in Alberta. .................................................. ...... 40
Appendix 4. Wolverine harvest density (# wolverine per km2) by WMU reported on
fur affidavits from 1995-1999 in Alberta. .................................................. ...... 41
Appendix 5. Wolverine harvest density (# wolverine per km2) by WMU reported on
fur affidavits from 2000-2004 in Alberta. .................................................. ...... 42
Appendix 6. Wolverine harvest density (# wolverine per km2) by WMU reported on
fur affidavits from 2005-2011 in Alberta. .................................................. ...... 43
Appendix 7. Lynx harvest density (corrected by area) in relation to RFMA and WMU
reported on fur affidavits from 1985-2010 in Alberta.. ................................. 44
1
1.0 INTRODUCTION
Wolverines (Gulo gulo) are one of the most rare and least understood carnivores in the
world. They are broadly distributed primarily in circumpolar regions, including
Canada, United States, Finland, Russia, and China. In Canada, the status of the eastern
wolverine population is Endangered, while the western population is classified as Special
Concern (COSEWIC 2003). The status of wolverines in Alberta is Data Deficient, meaning
that there is insufficient data to determine whether the population is at risk or secure
(Alberta Fish and Wildlife 2008). However, there is evidence to suggest that wolverines
may be sensitive to overharvest and need to be carefully managed (Banci 1994).
Wolverines are trapped primarily for their fur, and pelts are sold in a commercial
market to manufacture garments and taxidermy products (rugs, mounts). Since the
inception of the fur trade, wolverine fur has been sought after for its rugged durability,
warmth and frost-resistant properties on parkas and other garments, and has remained
particularly important to northern communities to the present day (Banci 1994). Being
both cryptic and difficult to catch, even basic information on wolverines is a challenge to
obtain. These factors, coupled with a vast home range and naturally occurring low
densities (Banci 1994), make wolverines one of the least understood and charismatic
carnivores in North America. With a paucity of basic information available from
Alberta, we examined wolverine trapping records to gain a better understanding of the
distribution and trend in harvest numbers over time.
Observations and harvest records from trappers have played a key role in the
management of furbearers. Biologists have used harvest records to help monitor
furbearer relative abundance (Smith and Brisbin 1984), adjust harvest quotas (Fryxell et
al. 2001), estimate population densities (Fryxell et al. 1999), examine cyclic fluctuations
(Viljugrein et al. 2001), evaluate status and distribution (Erickson 1982), collect biological
information (Simon et al. 1999), and assess the effects of trapping and other forms of
human disturbance on furbearers (Payer 1999; Webb and Boyce 2009). In Alberta, the
best large-scale data currently available for wolverines is harvest data from registered
traplines. Wolverines can be legally harvested on Registered Fur Management Areas
(RFMAs/registered traplines), which overlap primarily with the forested regions on
public Crown land. Beginning in 1989, a quota of one wolverine per trapline per season
2
(1 November - 31 January, for most of Alberta) was established along with a mandatory
registration program requiring all wolverine pelts be recorded and tagged before they
can be exported.
What little is known about wolverines in Alberta has been primarily obtained from
trapping records and anecdotal information. A trapper opinion survey in 1987 indicated
that wolverines were declining throughout Alberta (Skinner and Todd 1988). The status
of wolverine in the province was assessed, in large part by using trapping records
compiled by Petersen (1997), which showed a downward trend in wolverine harvests
through time (1972-1995). Poole and Mowat (2001) also used trapping records to assess
wolverine harvest trends from 1977-1999, which showed a general decrease, as well as
reduced distribution over time; however, it is likely that the harvest quota initiated in
1989 played an important role in affecting these trends. Thus, previous work suggests
that wolverine harvests, and potential wolverine population and distribution have
declined over time in Alberta, but analysis of more recent data is lacking.
As part of a larger wolverine research program in Alberta, we initiated an analysis of the
trapping data in order to evaluate the spatial and temporal trends in wolverine harvest.
Specifically, our objectives were to use trapping data to determine the distribution of
wolverines and identify some of the variables that may influence the spatio-temporal
pattern of wolverine harvest success. This work provides an updated summary of
wolverine harvest trends on registered traplines in Alberta. Based on discussions with
trappers, biologists and fish and wildlife officers, we predicted that there has been an
increasing trend in wolverine harvest since the last harvest analysis was completed over
a decade ago (Poole and Mowat 2001).
Previous studies have also shown a strong relationship between lynx and wolverine
harvests (Poole and Mowat 2001), so we explored this relationship to better understand
what factors may affect how many wolverines were harvested. We predicted that there
would be no relationship between wolverine pelt price and harvest because wolverines
are primarily sold to taxidermists; however, we expected there to be a relationship
between lynx pelt price and wolverine harvest because of increased effort to catch lynx
when prices were high. We expected to find a positive relationship between the number
of wolverine and lynx harvested, both at the provincial, as well as the trapline scale.
3
Since traps set for lynx can also catch wolverine, we predicted that traplines with a
greater lynx harvest would have a greater opportunity to catch wolverine.
Wolverines are built for living in cold places with deep snow. In fact, the number of
years with persistent spring snow cover has been a key factor in explaining where
wolverines occur and den around the world (Copeland et al. 2010). Researchers believe
that snow serves many purposes for wolverines, including thermal refuge and predator
avoidance for dens, and an ideal environment to “refrigerate” cached prey items for
extended periods of time (Copeland et al. 2010; Inman et al. 2012). Thus, we expected a
high degree of overlap between registered traplines with wolverine harvests and
locations with persistent spring snow cover. Wolverines also tend to occur in remote
places far from humans, so we expected to see increased wolverine harvests in areas
with more intact forest. Because males tend to have much larger home ranges than
females, we expected the wolverine harvests to be biased towards catching more males
than females.
2.0 STUDY AREA
There are three main types of trapping licenses in Alberta: (a) Resident Fur Management
(n=914 trappers), where trapping occurs on private property primarily in the Grasslands
and Parkland; (b) RFMA (n=1543 trappers), where trapping occurs on registered
traplines (n=1667 RFMAs) that overlap Crown land primarily in the Rocky Mountains,
Foothills, Boreal Forest and Canadian Shield; and (c) Indian/Metis Fur Management,
where trapping occurs on Indian Reserves and Metis Settlements across the province
(Government of Alberta 2012; Figure 1; Appendix 1).
We collected harvest data from RFMAs because these areas are mapped and have
remained somewhat consistent over time (Figure 1). Registered traplines give exclusive
rights for individuals (Jr. and Sr. partners) to trap furbearers in a specific area as
managed by Alberta Environment and Sustainable Resource Development (ESRD).
Created in 1938, the registered trapline system was meant to prevent over-harvest by
limiting competition and improve the overall management of furbearers (Pybus 2005).
Trapline boundaries are defined by anthropogenic (roads), political (National Parks) or
natural (rivers, ridges) features. Registered traplines range in size from one-half of a
4
township to 30 townships, with traplines tending to be larger as you move north in the
province. There is no trapping in National Parks, with the exception of Wood Buffalo
National Park in northern Alberta, which allows both commercial and subsistence
trapping (R. Kindopp and J. McKinnon pers. comm.).
5
Figure 1. Distribution of registered traplines (RFMA, n=1667), wolverine harvests (dot
is centered on RFMA polygon), and natural regions in Alberta. (National
Parks and non-trapline areas are excluded.)
6
3.0 MATERIALS AND METHODS
3.1 Data sources
We collected wolverine and lynx harvest data from three main sources: (a) fur affidavits,
(b) fur registrations, and (c) Statistics Canada pelt export records. Additional wolverine
data were gathered from the Fisheries and Wildlife Management Information System
(FWMIS) database and from the literature. The FWMIS data includes credible sightings
and locations of wolverines documented from research from 2002-2011.
Fur affidavit records were exported into an Excel spreadsheet from the Licensing and
Commercial Harvest database (R. Corrigan pers. comm.) for the available time period
(1985-2010); we also entered 2011 hard copy affidavits and appended them to the Excel
database. Trappers are required to submit fur affidavits to record how many of each
furbearer species were caught in the previous trapping season on a given trapline.
Affidavits are typically filled out when a trapper renews their annual trapping license at
a Fish and Wildlife office and are based solely on memory recall. We supplemented the
affidavit data with information from Alberta Fish and Wildlife (1990) that reported
affidavits by species and year from 1971-1989. We estimated the number of wolverine
harvests from a graph in the report but omitted 1971 data because it was suspiciously
low (n=6) as compared to Statistics Canada export records for that same year (n=23).
Fur registration records were exported into an Excel spreadsheet from the Licensing and
Commercial Harvest database (R. Corrigan pers. comm.) for the available time period
(1989-2011). Trappers are required to register all wolverine harvests at a Fish and
Wildlife office or fur dealer within 30 days of the end date of the wolverine trapping
season (January 30th for most of Alberta). Trappers must bring in evidence of the
wolverine (pelt or carcass) so that it can be marked with the metal tag that is required for
the pelt to be exported and sold. Fur registrations record the trapper’s name, trapline
number, harvested species and gender, harvest date, and trap type. Unlike fur
affidavits, fur registration records are based on actual physical evidence that a wolverine
was caught. Since 1989, trappers are only allowed to harvest one wolverine annually
per trapline. Additional wolverine caught over the quota is at the discretion of the Fish
and Wildlife Officer, but typically, trappers are allowed to keep one additional
7
wolverine if accidentally caught; if greater than two wolverines are caught or if caught
out of season, then they are turned over to Fish and Wildlife (J. Gerwing pers. comm.).
In addition to registered traplines, wolverines are also registered as a result of road kill,
trapping in Wood Buffalo National Park, Indian Reserve or Metis Colony, and incidental
capture from resident trapping in the white zone. However, specific locations of nontrapline
mortalities were often not recorded so we removed all non-trapline records
from the database and focused on harvests from registered traplines.
Fur export records and pelt price were extracted from the Statistics Canada CANSIMs
database for the available time period (1971-2010; Stats Canada 2012). Fur export
records are collected consistently across all the provinces and provide the total number
of exported pelts for each species in Canada. Unlike fur affidavits and fur registrations,
export records are reported only by province so they do not include harvest location.
We supplemented the Statistics Canada (2012) records with comparable data reported in
Todd and Geisbrecht (1979) to examine longer term provincial trends.
3.2 Data entry/proofing
We sorted wolverine harvest records by registered trapline number, year of harvest, and
wolverine count in order to identify potential outliers or obvious data entry errors
(duplicate records, wrong species, high counts). We cross-checked trapline numbers
with a trapline description database and a geographic information systems (GIS)
shapefile to determine that the trapline was valid, and updated the database accordingly
when trapline amalgamations occurred. Harvest records without a valid trapline
number were removed from the database (roadkill, Indian Reserve). Multiple affidavits
for a trapline were summed to create a total count of wolverines caught each year on a
given trapline. Suspect records were checked with hard copy records, when available,
in order to verify the data. In addition, we cross-referenced fur affidavits with fur
registrations and vice versa to determine how well the data matched up from 1989 to
2011.
We created a new affidavit record when a registration confirmed a wolverine harvest
but the affidavit appeared to be missing. Affidavits could be missed if the trapper did
not trap in the consecutive year and thus would not complete an affidavit for the
8
previous trapping season. High wolverine counts (>6 wolverine on a trapline in a given
year) that could not be verified with hard copy records were removed from the
database. We also mapped wolverine harvests to determine potential outliers based on
location (edge of white zone) and checked hard copies when possible. We entered
wolverine harvest records from hard copy forms not yet entered into the electronic
database for 2011-2012.
Lynx harvests were a secondary priority and had a large sample size, making it difficult
to adequately verify with hard copy records. We sorted the lynx records by registered
trapline number, year of harvest, and lynx count, and identified potential outliers.
Suspect records, including high counts (>50 lynx on a trapline in a given year) were
verified and updated with hard copy records, when available. Harvest records without
a valid trapline number were removed.
3.3. Distribution and harvest success
We used natural regions to summarize wolverine harvest data into landscapes with
similar climate, soil, topography, and vegetation. Refer to Natural Regions Committee
(2006) for a complete description of natural regions in Alberta. Registered traplines
overlap the Boreal Forest, Rocky Mountains, Foothills, and Canadian Shield natural
regions, so we limited our analysis to these regions. Because the Boreal Forest is large,
we summarized results into Northwest (NW) and Northeast (NE) Boreal based on the
Fur Management Zone (FMZ) map (Zones 1+3 = NE; Zone 2 = NW; Government of
Alberta 2012; Figure 1). To simplify the map and remain consistent with the FMZs, we
grouped the small section of Boreal Forest located south of Grande Prairie into the
Foothills region.
Registered traplines are assigned to the nearest Fish and Wildlife office, so we used the
office location to determine which natural region a fur affidavit would be located in.
Rocky Mountains included offices in Canmore, Blairmore, Pincher Creek, High River,
Sundre, Nordegg, and Grande Cache. Foothills included Rocky Mountain House,
Edson, Hinton, Foxcreek, Slave Lake, and Swan Hills. NW Boreal included Grande
Prairie, Valleyview, Fairview, Fort Vermilion, High Level, High Prairie, Manning, Peace
River, Spirit River, and Red Earth. NE Boreal included Athabasca, St. Paul, Lac La
Biche, Cold Lake, and Fort McMurray. In this case, fur affidavit records from the
9
Canadian Shield were included with the NE Boreal because the majority of the Ft.
McMurray district traplines are located in the Boreal Forest region.
We used ArcMap 10 (ESRI 2011) for making maps and summarizing data. We used a
GIS to determine wolverine harvest distribution from fur affidavit and registration
records. We mapped wolverine harvests by registered trapline and Wildlife
Management Unit (WMU) (Government of Alberta 2012). Refer to Appendix 1 to
determine specific WMUs and natural regions in the province. In order to protect the
identity of individual traplines, we summarized data by WMU when presenting harvest
densities. When traplines overlapped multiple WMUs, we assigned traplines to the
WMU that encompassed the greatest area.
We separated the fur affidavit harvest data into five time periods to compare long-term
changes in harvest distribution and success (1985-1989, 1990-1994, etc.). Similarly, we
compared fur registrations for 2 time periods (1989-1999; 2000-2011) to determine trends
by natural region. For the registration data, we assigned each registered trapline to a
natural region to compare spatio-temporal trends, and in this case, we separated the
Canadian Shield natural region. We determined the area of each natural region that
overlapped the registered trapline map (National Parks were excluded). We examined
the fur registrations to determine the proportion of female versus male wolverine
harvests, as well as the timing of the harvest.
We summarized lynx data using fur affidavits and registrations to determine whether
consistent trends were apparent from the different sources of data. When mapping lynx
distribution, we used the Natural Jenks function in ArcGIS to summarize density by
WMU. We plotted wolverine harvests against lynx harvests and pelt value to examine
whether potential relationships existed. We assumed our data was non-parametric and
used the Spearman rank correlation (rs) to determine the strength of the relationship
between different trapping-related variables (Zar 1999).
Researchers have found a strong relationship between the occurrence of wolverines,
particularly female dens, and years with persistent spring snow cover (snowpack
existing between 24 April - 15 May, 2000-2006; Copeland et al. 2012). We overlaid
wolverine harvest data onto the snowpack layer (Copeland et al. 2012) to determine the
relationship between wolverine occurrence and spring snowpack in Alberta. We
10
selected female harvests and determined the proportion of harvests that overlapped the
snowpack map by natural region.
We compared wolverine harvest registration data to Global Forest Watch Canada’s
(GFWC) map of large, intact forest landscapes (Lee et al. 2010). We measured the
number of years with a wolverine harvest as an indication of consistency in harvest over
time. We used Natural Jenks in ArcGis to group harvest data into three time periods.
An intact forest landscape as defined by GFWC is, “a contiguous mosaic of naturally
occurring ecosystems, including forest, bog, water, tundra, and rock outcrops, that is within a
forest ecozone, and that is essentially undisturbed by significant human influence visible on
Landsat satellite images. Intact forest landscape fragments are the best remaining pieces of our
once-intact forest landscapes and they are therefore critical to the restoration of ecosystem
functioning in areas affected by human development.” (Lee et al. 2010).
We decided that intact forest was an appropriate metric for human footprint (roads,
towns, wells, etc.) and to investigate how it relates to wolverine harvest at a very coarse
level. There were several categories of large, intact forests including: Intact Forest
Landscape (IFL) = >50,000 ha within all forest ecozones; IFL Fragment = 5,000-50,000 ha
within all forest ecozones; Temperate IFL = 1,000-5,000 ha within Temperate ecozone;
and Small Island = <5,000 ha within the Boreal/Taiga or <1,000 ha in Temperate ecozone.
We predicted that wolverine harvest success would be higher in areas with large, intact
forest landscapes based on previous research that shows that wolverines may be
sensitive to human activity and development.
11
4.0 RESULTS
4.1 Data quality
The original harvest database was relatively clean with few errors identified. We
verified approximately 11% of the wolverine registrations, and 20% of the wolverine
affidavits with original hard copy records. For lynx, approximately 1% of the affidavits
were verified with original hard copy records. Organization and availability of hard
copy records varied, so we are more confident in the database in some years than others.
Wolverine registrations in 2011 and 2012 are likely underestimated since we may be
missing records that had not yet been received by Fish and Wildlife district offices at the
time of data entry. The low number of wolverine typically harvested each year made it
relatively easy to identify potential data entry errors based on high counts. We
identified two common errors for fur affidavits, incorrect species (such as marten or
wolf) mistakenly entered by the trapper or data entry staff on the wrong line of the
affidavit, and duplicate records entered (identified based on the same information in
every field). Similarly, mistakes were found in fur registrations when the incorrect
species was checked off on the registration form. By cross-referencing registered
trapline numbers to the GIS shapefile of traplines, we were able to easily identify
harvests from non-registered traplines. However, it is still possible that some of these
harvests may overlap and were included in our analysis.
We observed a discrepancy between the data sources (Figure 2). It was expected that the
Statistics Canada data would not match the other data sources because it only included
pelts that were exported from the province, and also included wolverines caught in all
locations (Wood Buffalo National Park and Indian Reserves). Wolverines used
domestically or for taxidermy purposes, in which the pelt was tanned in Alberta, would
not receive an export permit and thus not be captured by Statistics Canada. In addition,
the Statistics Canada dates may not be comparable to affidavits or registrations because
it is based on what year the wolverine pelt was sold at auction as opposed to the year it
was harvested. However, we expected the fur affidavits and fur registrations to better
align since there is compulsory registration for every wolverine caught after 1989. In
general, the annual number of wolverines recorded by affidavits was higher than on
registrations. Roughly 25% of affidavits had missing registrations (indicating a
wolverine was caught on a given trapline in the same year but was not registered), and
12
24% of registrations were missing affidavits, from 1989–2011. There are some plausible
reasons for the discrepancy, including data entry errors and missing data (records from
district offices or fur dealers were not forwarded to ESRD). However, the timing of
when wolverines are reported on affidavits versus registrations likely explains much of
the variation. Trappers register wolverines during the same winter that the wolverines
were caught, whereas the affidavit may not be reported until the following summer or
fall. Trappers may forget to report it or may not fill out an affidavit altogether if they do
not trap the following season. A certain degree of uncertainty exists with each data
source and we recognize these data limitations.
Figure 2. Wolverine (W) harvests reported on fur affidavits, fur registrations, and
Statistics Canada fur exports, and average wolverine pelt price adjusted for
inflation, in Alberta from 1972-2011. Year represents the trapping season
end date (1990 indicates harvests made from 31 August 1989 – 31 August
1990).
13
4.2 Wolverine distribution
The distribution of wolverines, as determined from harvests reported on fur affidavits
(Figure 1), show wolverines primarily occurring in the Rocky Mountain, Foothills,
Boreal Forest and Canadian Shield natural regions. This distribution has remained fairly
consistent over time, with the possible expansion of wolverine into the Boreal Forest of
northeastern Alberta (Figure 3; Appendices 2-7). When trapping records were grouped
by natural region, the data shows a potential increase and expansion of wolverine
harvests occurring in northeastern Alberta, particularly in the Fort McMurray area
(Figure 3). For example, the average number of wolverines harvested in the NE Boreal
from 1985-1989 was 5 wolverines, as compared to 13 wolverines from 2005-2011. The
majority of the total wolverine harvests (1985-2011) occurred in the NW Boreal (63%),
followed by NE Boreal (includes Canadian Shield) (20%), Rocky Mountains (9%), and
Foothills (8%). The distribution of wolverines is roughly proportional to the percent
area of each natural region, with the exception of the NW Boreal and Foothills.
Approximately 58% of wolverines were harvested in the NW Boreal, which comprises
only 41% of the total trapline area (RFMAs). The Foothills comprise 23% of the trapline
area, but only 10% of the total wolverine harvest (Figure 3; Table 1).
Wolverine harvests reported on fur registrations were also summarized by natural
region to determine changes over space and time. The fur registrations showed a similar
spatio-temporal trend as the affidavits. When comparing two time periods (1989-1999;
2000-2011; Table 1), wolverine harvests have more than doubled in the NW Boreal
(105%), and also increased by 47% and 33% in the NE Boreal and Foothills, respectively.
During the same time periods, wolverine harvests declined by 40% and 32% in the
Canadian Shield and Rocky Mountains, respectively.
14
Figure 3. Average wolverine harvest reported on fur affidavits, by natural region,
across different time intervals in Alberta. Harvests from the Canadian
Shield were included with the NE Boreal natural region.
Table 1. Percent of total trapline area, total wolverine harvests, and harvest density,
summarized by natural region, for two time periods in Alberta.
Natural region
Percent of total
trapline area
Total harvests
1989-1999 (#/km2)
Total harvests
2000-2011 (#/km2)
Canadian Shield 2.6% 13 (0.0014) 8 (0.0009)
Rocky Mountains 6.9% 35 (0.0004) 24 (0.0003)
Foothills 23.4% 28 (0.0011) 37 (0.0015)
NE Boreal 25.8% 49 (0.0005) 72 (0.0008)
NW Boreal 41.4% 121 (0.0008) 249 (0.0016)
Natural region
Average Wolverine Harvest
15
FWMIS locations for wolverine occurrences demonstrate that registered traplines could
have wolverines present without a harvest taking place, so we use caution when
interpreting the distribution of wolverines from harvest data alone (Figure 4). Although
our maps do not indicate wolverine harvests in National Parks, wolverines likely occur
there (Waterton Lakes, Banff, Jasper, and Wood Buffalo National Parks). In addition to
harvests on registered traplines, other mortality events are recorded on fur registrations,
including road kill, harvests on Indian Reserves, Metis colonies or Wood Buffalo
National Park, as well as accidental harvest by resident trappers (Table 2). These
mortality events are challenging to map because they often have no specific location
information. No wolverine mortalities were registered prior to 1995, indicating an
increase in non-trapline mortalities after that point, and particularly in 2012. However,
it’s difficult to determine if the more recent reporting of wolverine mortalities on fur
registrations simply reflects an increased effort (Fish and Wildlife staff, general public)
to report them.
16
Figure 4. Wolverine harvest densities in relation to RFMA and WMU, as reported on
fur registrations from 1989-2011, and FWMIS locations from 2002-2011.
17
Table 2. Wolverine mortalities on registered traplines (RFMA) and non-trapline
locations (roadkill, trapping in Indian Reserve/Wood Buffalo National Park,
accidental harvest by resident trapper), reported on fur registrations from
1995-2012. [2012] data is incomplete since all registrations had not been
received.
Year
Non-trapline
mortalities
Registered trapline
mortalities
1995 1 20
1996 1 24
1997 0 26
1998 0 20
1999 1 21
2000 1 14
2001 0 17
2002 0 31
2003 2 49
2004 3 46
2005 1 36
2006 2 34
2007 4 35
2008 1 23
2009 1 36
2010 1 70
2011 1 22
[2012] 8 39
It is not surprising that approximately 80% of the wolverine harvests occur in the Boreal
Forest in northern Alberta, given that’s where the majority (67%) of the available
trapping area occurs (Table 1). However, when wolverine harvest densities are
corrected by WMU area, the highest densities occur within the Rocky Mountains (Figure
4). According to fur registrations, the highest wolverine harvest densities (>5.5
wolverines/1000 km2) over time (1989-2011) have occurred in the Rocky Mountains:
Castle-Carbondale (WMU 400), Solomon (WMU 439) and Adams Creek (WMU 440).
Medium wolverine harvest densities (2.7-5.5 wolverines/1000 km2) have been reported
in both the Boreal Forest and Rocky Mountains: Siffleur Wilderness (WMU 736), Red
Cap (WMU 437), Narraway (WMU 445), Lakeland Provincial Recreation Area (WMU
841), Panny River (WMU 541), Chinchaga River (WMU 524), Rainbow Lake (WMU 536)
and Bistcho Lake (WMU 539). However, some of these higher densities may be the
18
result of a small number of wolverine harvests reported in a small WMU. Therefore,
harvest density at the scale of a WMU may not be a robust indication of population
numbers.
The total number of years with a wolverine harvest may indicate where a more stable
trapper effort and wolverine population occurs. Since 1989, the registered traplines that
had five or more years with a wolverine harvest are located in the Rocky Mountains
(WMU 400*, 440*, 445*), Foothills (WMU 353), and Boreal Forest (WMU 524*, 527, 530,
532, 534, 535, 536*, 539*, 541*); * indicates WMUs that also had medium to high
wolverine harvest densities (2.7-10.8 wolverines/1000 km2; Figure 5; Appendix 1).
Although, some of the mountain areas had the highest wolverine harvest densities,
places where wolverines have been harvested the highest number of years (12-23 years)
occurred in the Boreal Forest, primarily north of 58 degrees latitude (WMU 530, 536, 539;
Figure 5). Coincidentally, some of these WMUs in the Boreal Forest (WMU 524, 532, 534,
536, 539) have a wolverine trapping season extended by two weeks (1 November – 15
February; Government of Alberta 2012).
In general, the harvest data indicates that the core distribution of wolverine spans along
the Rocky Mountains and across the northern Boreal Forest of Alberta. Some of this
distribution is consistent with our predictions of where wolverines should occur,
assuming spring snowpack is important for recruitment and long term population
persistence (Copeland et al. 2010; Figure 6): Caribou Mountains (WMU 534, NW Boreal),
Chipewyan (WMU 532, Canadian Shield), Bistcho Lake (WMU 539, NW Boreal) (Figure
1; Appendix 1). The spring snow distribution is relatively continuous throughout the
Boreal Forest of Canada, with the exception of Alberta, which tends to have a more cold
and dry climate (The Wolverine Foundation 2012). The snow cover map indicates more
consistent spring snowpack in the Rocky Mountains and a patchy distribution, with
large areas devoid of spring snow cover, throughout most of northern Alberta (north of
56 degrees latitude; Figure 6). And yet, we have evidence of fairly consistent long term
wolverine harvest from many areas that appear to lack large areas of spring snow cover
(WMU 524, 527, 530, 541; Figure 6; Appendix 1). The relationship between wolverine
occurrence and spring snowpack was strongest when considering female dens, where
98% of dens found across the wolverine’s circumpolar range overlapped places where
snow persisted one or more years into spring (2000-2006; Copeland et al. 2010). In
Alberta, we found that approximately 63% of traplines that reported female harvest had
19
at least 1 year (n=86 of 138 traplines) with spring snowpack, and 13% of traplines had 5
or more years (n=18 of 138 traplines) (Figure 6). The majority of traplines that caught a
female wolverine and had spring snowpack for 5 or more years were in the Rocky
Mountains (61%; n=11 traplines), followed by NW Boreal (28%; n=5), and Canadian
Shield (11%; n=2). However, there are other areas in the Rocky Mountains that have
persistent spring snow coverage but lack wolverine harvest (Kananaskis country,
WMUs 410-414; Figure 6; Appendix 1), indicating that other factors must be considered
when determining wolverine distribution (trapping effort, prey, human activity,
development, etc.).
Global Forest Watch Canada indicated that larger patches of continuous, intact forests
occurred in the Rocky Mountains, Boreal Forest and Canadian Shield (Lee et al. 2010)
(Figure 5). The landscapes of the Boreal Forest tend to be more fragmented south of 57
degrees latitude. The areas of the northern Boreal Forest of Alberta that support higher
wolverine harvests appear to be less fragmented; a greater proportion of traplines that
harvested a wolverine for five or more years overlapped with intact forest (WMU 400,
440, 530, 531, 532, 534, 536, 539, 540). However, some areas with large, intact landscapes
in the mountains have low or zero wolverine harvest (Happy Valley (WMU 308), Upper
Saskatchewan (WMU 426), Bighorn (WMU 430), White Goat (WMU 432), Cardinal
(WMU 436), White Goat Wilderness (WMU 738)). Despite a lack of intact forests in the
Foothills, the number of wolverine harvests has increased by 32%, while harvests in the
more intact Rocky Mountain landscape have declined by 31% (Table 1).
20
Figure 5. Number of years with a wolverine harvest (dot is centered on RFMA
polygon) in relation to Intact Forest Landscapes, from 1989-2012 in Alberta.
21
Figure 6. Distribution of female wolverine harvests reported on fur registrations from
1989-2011, and total wolverine harvest from fur registrations and affidavits
from 1985-2011 (red dots centered on RFMA), relative to years with spring
snow cover from 2000-2006 in Alberta.
22
4.3 Wolverine harvest success
4.3.1 Statistics Canada
Wolverine pelt exports have fluctuated over time with the peak export recorded in 1927
(350 wolverines exported) and the lowest number recorded in 1954 (0 wolverines)
(Figure 7), which likely reflects the effects of intense poisoning efforts from 1952-1956 to
control a rabies outbreak (Pybus 2005). The initiation of a quota in 1989 of one
wolverine harvest per year per registered trapline, makes it difficult to compare harvest
data as a reflection of potential population change over pre- and post-quota time
periods. Trapping effort may have changed as a result of the quota, although harvest
numbers are within a similar range during different time periods (1960-1970 vs. 1997–
2010; Figure 7).
Wolverine pelt values have also varied over time (range: $170 in 2005 to $600 in 1976)
and appear to have a positive relationship to wolverine harvest from 1972-2011 (rs=0.495,
d.f.=38, p=0.001; Figure 8). In addition, unadjusted (for inflation) average lynx pelt
values also had a positive relationship to wolverine harvests recorded by Statistics
Canada from 1972-2010 (rs=0.422, d.f.=38, p=0.007), fur affidavits from 1972-2010 (rs=0.544,
d.f.=37, p=0.0003), and fur registrations from 1989-2010 (rs=0.499, d.f.=20, p=0.02). The
strength of the relationship between lynx pelt values and wolverine harvests is even
stronger as they both increase from 1995-2010 (rs=0.72, d.f.=14, p=0.002). Discrepancies
exist between the Statistics Canada, fur affidavit, and fur registration records on an
annual basis, but the general increasing trend across data sources is apparent from 2000-
2010 (Figure 2).
23
Figure 7. Total wolverine pelts exported from Alberta from 1921-2010. Export records
from 1921-1970 were estimated from Todd and Geisbrecht (1979) and records
from 1971-2010 were derived from Statistics Canada (2012). Year represents
the trapping season end date.
Figure 8. Five-year average wolverine harvest, and total wolverine harvest reported
on fur affidavits, and wolverine pelt price for Alberta from 1972-2011.
Wolverine affidavit data from 1972-1984 were estimated from Alberta Fish
and Wildlife (1990). Year represents the trapping season end date. Five-year
time periods are 1972-1976, 1977-1981, etc.
24
4.3.2 Fur affidavits and registrations
Data from fur affidavits show variable wolverine harvests over time, with noticeably
greater numbers prior to 1989 when there was no quota in place (Figure 8). Wolverine
harvests from affidavits (1972-2011) have ranged from a low of 20 wolverines in 1996 to
a high of 104 wolverines in 1983. Wolverine harvests were on a downward trend from
1984-1996, but harvests have shown an increasing trend from 1997-2011. In fact,
wolverine harvests have increased by 1.5 fold when comparing the five-year averages of
1992-1996 to 2007-2011 (Figure 8). Similarly, the average number of traplines reporting a
wolverine harvest has risen, with a 1.6 times increase from the 1990-1994 time period to
the 2005-2011 time period (Figure 9). On average, approximately 3% of all traplines
(n=46, range: 33-61) harvested wolverine in the most recent time period (2005-2011).
Approximately 25% of all traplines (n=418) have reported a wolverine harvest on fur
affidavits at least once in the past 27 years (1985-2011). The number of traplines that
have reported more than one wolverine harvests on an affidavit is ≤1% of the total
traplines (range: 1-12) from 1985-2011; the vast majority of which are located in the
Rocky Mountains and NW Boreal. So despite the increasing trend in the number of
traplines catching wolverine, the proportion of traplines catching more than one
wolverine per year does not appear to be increasing.
Wolverine harvests reported on fur registrations do not match fur affidavits precisely,
but do follow similar trends from 1989-2011 (rs=0.591, d.f.=21, p=0.003). In general, the
number of wolverine harvests registered tends to be lower than affidavit reports, on an
annual basis (Figure 2). The annual number of wolverine harvests from registrations
(1989–2011) has ranged between 14 (1994, 2000) and 70 (2010) (Figure 10). The average
number of wolverines registered has increased 1.5 fold from the 1989-1992 time period
to the 2009-2012 time period (Figure 10). Approximately 18% of traplines (n=305) have
registered a wolverine harvest at least once during the past 23 years (1989-2011).
December (43%) and January (40%) were the primary months when trappers caught
wolverine, as compared to 9% in November and 8% in February. The proportion of
males and females harvested was surprisingly consistent across months; roughly 66% of
the total registered wolverine harvests were male and 32% were female (n=467; range:
male=54-86%, female=14-46%) (Figure 11). Although greater numbers of males were
harvested in all months, the ratio of males to females increased by approximately 20% in
February as compared to the other months (Figure 11). The sex ratio of the Alberta
25
harvest was comparable to wolverines harvested in British Columbia, where females
comprised less than 50% of the total harvest (Lofroth and Ott 2007).
Figure 9. Total and average number of traplines that reported a wolverine harvest on
fur affidavits from 1985-2011 in Alberta. Time periods are 1985-1989, 1990-
1994, 1995-1999, 2000-2004, and 2005-2011.
26
Figure 10. Total and four-year average of wolverine harvests reported on fur
registrations from 1989-2012 in Alberta. 2012 data is incomplete since all
records had not been received. Year represents the trapping season end
date. Four-year time periods are 1989-1992, 1993-1996, etc.
Figure 11. Total and percent of wolverine harvests by gender, reported on registered
traplines by month, from 1989-2012 in Alberta. Wolverines with unknown
gender were excluded.
61%
66%
32%
67%
30%
66%
32%
37%
61%
27
4.3.3 Wolverine and lynx
Wolverine and lynx both had harvest quotas initiated in the late 1980s (1988 for Lynx;
1989 for wolverine). Trapping records reflect the well-known 10-year lynx cyclic pattern
with harvest peaks occurring around 1982, 1992, 2002 and suspected in 2012 (Figure 12).
Although wolverine harvests did not fluctuate at the same magnitude as lynx, we found
that wolverine harvests seemed to track the ups and downs in the lynx harvest data
strikingly well from 1985-2010 (rs=0.372, d.f.=24, p=0.06; Figure 12). Interestingly, this
trend holds up for both fur affidavits (Figure 12) and fur registrations (Figure 13), but
less well for the longer-term Statistics Canada records, where lynx harvests were much
greater in magnitude prior to 1980 (Figure 14). The distribution of lynx, as determined
from fur affidavits from 1985-2010, indicate that lynx are broadly distributed with
harvest densities highest in the NE/East-central Boreal (north of 54 degrees latitude), the
NW Boreal (north of 56 degrees latitude), and to a lesser degree in the Foothills and
Rocky Mountains (Appendix 7). We found a high spatial and temporal overlap of
wolverine and lynx harvests; approximately 67% of traplines that had a wolverine
harvest also reported a lynx harvest in the same year (1985-2011).
Figure 12. Total lynx (n=30,438) and wolverine (n=1,185) harvests reported on fur
affidavits from 1985–2010 in Alberta. Year represents trapping season end
date.
28
Figure 13. Total lynx (n=25,697) and wolverine (n=671) harvests reported on fur
registrations from 1989-2010 in Alberta. Year represents trapping season end
date.
Figure 14. Total number of lynx (n=168,457) and wolverine (n=2,029) pelts exported
from 1971–2010 in Alberta, as reported by Statistics Canada (2012). Year
represents trapping season end date.
29
5.0 DISCUSSION
We provide the most recent and comprehensive summary of wolverine harvest data for
Alberta. In contrast with previous analyses on wolverine data prior to 1999 (Petersen
1997; Poole and Mowat 2001), we show evidence of an expansion in wolverine harvest
distribution (Table 1), which also coincides with an increase in the total number of
wolverine harvests reported on fur affidavits and fur registrations from 1997-2011
(Figures 8, 10). The mean annual number of wolverine harvests reported on affidavits,
as well as the number of traplines reporting a wolverine harvest, has increased since the
mid-1990s. However, there has been no trend in the frequency of traplines reporting >1
wolverine/year over this time period. The locations where more than one wolverine
were caught were very similar to the locations with higher harvest densities and greater
number of years reporting a harvest (NW Boreal and Rocky Mountains). Although
wolverines continue to be harvested in the Rocky Mountains, the traplines where more
than one wolverine was harvested is shifting away from this region (n=7 traplines (1985-
1989), n=5 (1990-1994), n=0 (1995-1999), n=2 (2000-2004), n=2 (2005-2011); total harvest
from the Rocky Mountains has also declined over time (Table 1; Figure 3), and may
indicate a change in trapper effort.
It is not surprising that the vast majority of wolverine harvests (82%) occurred in the
Boreal Forest in northern Alberta given that most of the available trapping area occurs
there (67%). It is not clear whether the increase in wolverine harvests in the NE Boreal is
related to an increase in the wolverine population or simply to trapper effort. Skinner
and Todd (1988) used a trapper questionnaire to estimate that wolverines had a sparse
distribution in NE Alberta, approximately between 53-56 degrees latitude and east of
115 degrees longitude; harvest data also suggested a sparse distribution in this area.
However, the more recent harvest data shows a potential expansion in wolverine
distribution into parts of the Foothills (Slave Lake area) and the NE Boreal (Fort
McMurray area). It is possible that increases in human population and industrial
development, particularly in the Fort McMurray area, have resulted in previously
remote registered traplines now becoming more accessible to trappers.
Although harvest data is useful for plotting wolverine occurrence on a map, we do not
know whether or not wolverines are present on traplines that have not reported a
harvest. While the majority of the FWMIS locations overlapped areas that had
30
wolverine harvests, there were a few FWMIS locations where wolverine harvests had
not occurred (Highwood (WMU 404), Elbow (WMU 406)). This suggests that multiple
sources beyond harvest data can provide a more informed picture of wolverine
distribution. FWMIS locations are useful in determining wolverine range and
distribution even though they are often opportunistic observations recorded by
recreationalists, research scientists, or consultants performing tasks for industry. This is
evidenced by the increase in FWMIS locations in the oilsands area near Fort McMurray,
Willmore Wilderness and some parts of Kananaskis country.
We were surprised to find that large areas of the Boreal Forest lacked persistent spring
snow cover, yet had consistent wolverine harvests through time (1985-2011). Studies
have suggested that wolverines have an obligate need for persistent spring snowpack
which is linked to successful denning recruitment (Copeland et al. 2010). Exploring
other metrics, such as distance to the nearest spring snow cover, would help in
determining whether wolverines tend to occur closer to places with spring snowpack.
New research has shown evidence for a food-based hypothesis to help explain why
wolverines occur in cold, low-productivity environments (Inman et al. 2012). Areas
with deep snow provide wolverines with an ideal place to cache prey and den. Areas
that lack snow during the denning period may still have suitable cover for dens (hollow
trees, woody debris, etc.), prey availability, and microsites to refrigerate cached items for
extended time periods (sphagnum moss). Also, the snow cover map is not 100%
accurate and wolverines may have snow dens that occur outside the snow coverage
map. Investigating wolverine dens in the Boreal Forest of Alberta would give better
information on the characteristics of dens and how they differ from other studies.
Temperature has also been suggested as an important predictor of wolverine occurrence,
so further research could explore how temperature, particularly soil temperature or
mean August temperature differs across Alberta (Copeland et al. 2010). Although
elevation is likely related to both spring snowpack and temperature, it is also an
important variable to consider given that wolverines may exhibit thermal avoidance
(Copeland et al. 2010). Prey availability is another important consideration to wolverine
distribution but it is difficult to quantify; trappers could provide information based on
their experiences, as well as wolverine specimens for further study (stable isotope
analysis). Anecdotal observations from the Boreal Forest suggest that wolverines may
31
benefit from ungulate roadkill near roads, but wolverines can also become victims
themselves; commercial fisheries can also provide an important food source (Bistcho
Lake; J. Gerwing pers. comm.), as well as snowshoe hare populations.
Other research has found that wolverines tend to occur in places further away from
people (Banci 1994). This certainly would seem to be the case when considering the
wolverine harvest distribution in Alberta, with the majority of the harvest occurring in
the less populated and less developed areas of the Boreal Forest and Rocky Mountains.
Large, intact forests are fairly continuous among traplines in the mountain region but
they have a narrow width, while larger, more patchy intact forests occur in the Boreal
Forest (Lee et al. 2010). We anticipated more consistent wolverine harvest among
traplines in the mountain region given the low human footprint.
Further research should include areas where wolverine harvests seem to be
discontinuous, such as the areas northwest of Porcupine Hills, southwest of Chain
Lakes, west of Highway 22, Kananaskis country, the lower foothills west of Drayton
Valley and north to Swan Hills, and the area on the east side of the Athabasca River,
northeast of Wabasca and west of Fort McMurray. Coincidentally, many of these areas
have undergone intense human development and/or have heavy levels of recreational
activity (Herrero 2005; Schneider 2002). As a result, grizzly bear conflicts and/or
mortality are also high in the areas around Swan Hills and Kananaskis country (Alberta
Grizzly Bear Recovery Team 2008). An increased human footprint could be
discouraging trapper effort in some of these areas (Webb et al. 2008), and/or be reflective
of a true lack of wolverine occurrence.
Conversely, the lack of wolverine harvest in some areas may be caused by remote,
inaccessible areas located far away from towns (northeast of Wabasca along the
Athabasca River, some mountain traplines). Without further investigation, we cannot
determine wolverine status in these areas, although discussions with trappers would be
a good place to start. Despite the evidence that wolverines don’t generally occur near
concentrated human activities or development, apparently wolverines can travel
through these areas occasionally. A wolverine was photographed running through a
new subdivision in Airdrie, just north of Calgary in March 2012, and a wolverine was
videotaped eating dog food in a backyard in St. Albert, just north of Edmonton in June
32
2012 (R. Corrigan pers. comm.). A family group of wolverines were also observed living
at the dump in High Level (J. Gerwing pers. comm.).
It is interesting that wolverine harvests would closely track lynx harvests, both at the
provincial and trapline scale. The high degree of spatial and temporal overlap of
wolverine and lynx suggest the two are closely linked. Lynx are considered a “bread
and butter” furbearer species with a wide distribution, relatively high densities when
populations are increasing, and they tend to have valuable pelts. We hypothesize that
the close relationship between lynx and wolverine is likely caused by a combination of
factors including trapper effort, and natural population fluctuations. Trappers likely
adjust their effort in relation to furbearer abundance, and so as lynx numbers increase,
trapper effort increases. As more lynx are caught, the encounter and catch rate of
wolverine (bycatch) also increases, as they can be incidentally caught in lynx traps.
As a scavenger, it is also possible that wolverines are attracted to active traplines to steal
bait and/or trapped lynx, making the trapper more likely to target wolverines. In
addition, we cannot rule out the importance of the snowshoe hare and lynx population
fluctuations. Previous researchers have documented the importance of the snowshoe
hare, particularly when they are very abundant, in the diet of wolverines (Banci 1994).
As the lynx population declines in concert with snowshoe hare numbers, wolverines
may benefit from scavenging lynx that struggle to find an alternate prey source.
Changes to trapping regulations for lynx may affect wolverine harvest, since the former
appears to have some close link to the latter. We will be looking into lynx and wolverine
relationships in future studies.
Although long-term furbearer harvest data is valuable, we use caution in interpreting
these results as a direct reflection of actual wolverine population trends. Trapper effort
is a complex process that is influenced by a wide variety of factors in Alberta including:
furbearer abundance, weather, rising average income, health, work schedule, fur prices,
improved recreational access, and increasing industrial development (Mullen 2006). To
better tease apart the effect of trapper effort from population effects, trapping records
would be more valuable if trappers recorded the number of trapnights per species.
Trapper questionnaires have been used in the past to get a snapshot of furbearer
abundance and continue to be an important source of traditional knowledge. As part of
a larger wolverine project, we will be using a trapper questionnaire, in conjunction with
33
the collection of field data, to measure wolverine occurrence and gene flow in order to
better understand the status of wolverines in Alberta. This updated wolverine harvest
summary gives us good baseline information to work from.
34
6.0 LITERATURE CITED
Alberta Grizzly Bear Recovery Team. 2008. Alberta grizzly bear recovery plan 2008-
2013. Alberta Species at Risk Recovery Plan No. 15, produced by Alberta
Sustainable Resource Development (ASRD), Edmonton, Alberta, Canada. 68 pp.
Alberta Fish and Wildlife. 1990. Fur affidavits in Alberta 1985-1989. Produced by
ASRD, Edmonton, Alberta, Canada.
Alberta Fish and Wildlife. 2008. Report of Alberta’s Endangered Species Conservation
Committee: June 2006. Produced by ASRD, Edmonton, Alberta, Canada. 44p.
Banci, V. 1994. Wolverine. Pages 99-127. In: L. Ruggiero. The scientific basis for
conserving forest carnivores: American marten, fisher, lynx and wolverine in the
western United States. General Technical Report RM-254, produced by United
States Department of Agriculture, Fort Collins, Colorado, U.S.A. 184 pp.
Copeland, J.P., K.S. McKelvey, K.B. Aubry, A. Landa, J. Persson, R.M. Inman, J. Krebs, E.
Lofroth, H. Golden, J.R. Squires, A. Magoun, M.K. Schwartz, J. Wilmot, C.L.
Copeland, R.E. Yates, I. Kojola, and R. May. 2010. The bioclimatic envelope of
the wolverine (Gulo gulo): do climatic constraints limit its geographic
distribution? Canadian Journal of Zoology 88(3): 233-246.
COSEWIC. 2003. COSEWIC assessment and update status report on the wolverine Gulo
gulo in Canada. Produced by the Committee on the Status of Endangered
Wildlife in Canada, Ottawa, Ontario, Canada. vi + 41 pp.
Erickson, D.W. 1982. Estimating and using furbearer harvest information. Pages 53-66.
In: G.C. Sanderson. Midwest furbearer management. Symposium proceedings at
the 43rd Midwest Fish and Wildlife Conference, Wichita, Kansas, U.S.A.
ESRI. 2011. ArcGis Desktop: Release 10. Environmental Systems Research Institute,
Redlands, California, U.S.A.
35
Fryxell, J.M., J.B. Falls, E.A. Falls, R.J. Brooks, L. Dix, and M.A. Strickland. 1999. Density
dependence, prey dependence, and population dynamics of martens in Ontario.
Ecology 80: 1311-1321.
Fryxell, J.M., J.B. Falls, E.A. Falls, R.J. Brooks, L. Dix, and M.A. Strickland. 2001.
Harvest dynamics of mustelid carnivores in Ontario, Canada. Wildlife Biology 7:
151-159.
Government of Alberta. 2012. 2012-2013 Alberta Guide to Trapping Regulations. URL:
http://www.albertaregulations.ca/tra...scriptions.htm (Accessed
28 November 2012).
Herrero, S. 2005. Biology, demography, ecology and management of grizzly bears in
and around Banff National Park and Kananaskis Country: The final report of the
Eastern Slopes Grizzly Bear Project. Produced by the Faculty of Environmental
Design, University of Calgary, Calgary, Alberta, Canada. 248 pp.
Inman, R.M., A.J. Magoun, J. Persson, and J. Mattisson. 2012. The wolverine’s niche:
linking reproductive chronology, caching, competition, and climate. Journal of
Mammalogy 93(3): 634-644.
Lee, P., M. Hanneman, J. Gysbers, and R. Cheng. 2010. Atlas of key ecological areas
within Canada’s intact forest landscapes. Produced by Global Forest Watch
Canada, Edmonton, Alberta, Canada. 54 pp.
Lofroth, E.C., and P.K. Ott. 2007. Assessment of the sustainability of wolverine harvest
in British Columbia, Canada. Journal of Wildlife Management 71(7): 2193-2200.
Mullen, S.M. 2006. Furbearer harvests and landscape change in the Rocky Mountain
foothills of Alberta. Master’s Thesis. University of Alberta, Edmonton, Alberta,
Canada. 120 pp.
Natural Regions Committee. 2006. Natural regions and subregions of Alberta.
Publication No. T/852, produced by the Government of Alberta, Edmonton,
Alberta, Canada. 254 pp.
36
Payer, D.C. 1999. Influence of timber harvesting and trapping on habitat selection and
demographic characteristics of marten. Doctoral dissertation. University of
Maine, Orono, Maine, U.S.A. 298 pp.
Petersen, S. 1997. Status of the wolverine (Gulo gulo) in Alberta. Wildlife Status Report
No. 2, produced by Alberta Environmental Protection, Edmonton, Alberta,
Canada. 17 pp.
Poole, K.G., and G. Mowat. 2001. Alberta furbearer harvest data analysis. Alberta
Species at Risk Report No. 31, produced by ASRD, Edmonton, Alberta, Canada.
51 pp.
Pybus, M. 2005. Furbearers: trapping, fur farms and problem wildlife. Pages 141-174.
In: Fish, Fur and Feathers, fish and wildlife conservation in Alberta: 1905-2005.
Fish and Wildlife Historical Society and Federation of Alberta Naturalists,
Edmonton, Alberta, Canada. 418 pp.
Schneider, R.R. 2002. Alternative Futures: Alberta’s Boreal Forest at the Crossroads.
Produced by The Federation of Alberta Naturalists, Edmonton, Alberta, Canada.
152 pp.
Simon, N.P.P., F.E. Schwab, M.I. LeCoure, and F.R. Phillips. 1999. Fall and winter diet
of martens, Martes americana, in central Labrador related to small mammal
densities. Canadian Field Naturalist 113: 678-680.
Skinner, D.L., and A.W. Todd. 1988. Distribution and status of selected mammals in
Alberta as indicated by trapper questionnaires in 1987. Produced by ASRD,
Edmonton, Alberta, Canada. 64 pp.
Smith, L.M., and I.L. Brisbin. 1984. An evaluation of total trapline captures as estimates
of furbearer abundance. Journal of Wildlife Management 48: 1452-1455.
Statistics Canada. 2012. Census of wildlife pelt production, record number 3428. Table
003-0013, Number and value of pelts produced (wolverine and lynx). URL:
http://www5.statcan.gc.ca/cansim/a01?lang=eng (Accessed 12 January 2012).
37
The Wolverine Foundation (TWF). 2012. TWF homepage. URL: http://
wolverinefoundation.org/ (Accessed 20 August 2012).
Todd, A.W., and L.C. Geisbrecht. 1979. A review of Alberta fur productions and
management, 1920/21 to 1977/78. Produced by Alberta Energy and Natural
Resources, Edmonton, Alberta, Canada 65 pp.
Viljugrein, H., O.C. Lingjaerde, N.C. Stenseth, and M.S. Boyce. 2001. Spatio-temporal
patterns of mink and muskrat in Canada during a quarter century. Journal of
Animal Ecology 70: 671-682.
Webb, S.M., D.J. Davidson, and M.S. Boyce. 2008. Trapper attitudes and industrial
development on registered traplines in west-central Alberta. Human
Dimensions of Wildlife 13: 115-126.
Webb, S.M., and M S. Boyce. 2009. Marten fur harvests and landscape change in westcentral
Alberta. Journal of Wildlife Management 73: 894-903.
Zar, J.H. 1999. Biostatistical analysis (4th edition). Prentice Hall Inc., Upper Saddle
River, New Jersey, U.S.A. 123 pp.
38
7.0 APPENDICES
Appendix 1. Wildlife Management Units (WMU) and Natural Regions in Alberta.
39
Appendix 2. Wolverine harvest density (# wolverine per km2) by WMU reported on
fur affidavits from 1985-1989 in Alberta. National Parks and non-trapline
areas were excluded.
40
Appendix 3. Wolverine harvest density (# wolverine per km2) by WMU reported on
fur affidavits from 1990-1994 in Alberta. National Parks and non-trapline
areas were excluded.
41
Appendix 4. Wolverine harvest density (# wolverine per km2) by WMU reported on
fur affidavits from 1995-1999 in Alberta. National Parks and non-trapline
areas were excluded.
42
Appendix 5. Wolverine harvest density (# wolverine per km2) by WMU reported on
fur affidavits from 2000-2004 in Alberta. National Parks and non-trapline
areas were excluded.
43
Appendix 6. Wolverine harvest density (# wolverine per km2) by WMU reported on
fur affidavits from 2005-2011 in Alberta. National Parks and non-trapline
areas were excluded. FWMIS locations are from 2002-2011.
44
Appendix 7. Lynx harvest density (corrected by area) in relation to RFMA and WMU
reported on fur affidavits from 1985-2010 in Alberta. National Parks and
non-trapline areas were excluded.
Alberta Conservation Association acknowledges the
following partner for their generous support of this project:
wolverine harvest data
Linear Mode
