2002 Wildlife Society Conference
Poster Presentation

title
by:
S. McNay, K. Zimmerman, D. Heard, R. Ellis, L. Giguere, R. Sulyma, G. Watts, D. Seip, K. Schmidt, P. Hengeveld, S. Walshe, G. Keddie, L. Wilson, R. McCann, T. Barrett, A. Walton, A. Doucette, and J. Joy

1. OBJECTIVES

  1. To estimate and test preferences for pine-lichen winter ranges used by northern caribou

  2. To evaluate the efficacy of current conservation policies in providing a sustainable supply of caribou seasonal ranges
 Caribou Group


Capture Crew
2. METHODS

  • Implement a caribou monitoring program for 3 northern caribou herds
  • Develop a model to predict preference for caribou ranges
  • Using empirical data, test for associations between range preference and classes of range availability
  • Determine the amounts of available current and potential ranges within herd boundaries that are covered by policy, and forecast range supply over time

3. STUDY AREA

  • North-central British Columbia
  • Three caribou herds monitored, models and tests were run on the Wolverine Herd area (7500 km2, or 2900 miles2)
 Akie Mountains

NATIONAL STATUS OF CARIBOU HERDS IN BC:




Captured Caribou Calf
4. Monitoring Program
  • Monitored VHF and GPS radio-collared caribou, moose, and wolves in 3 caribou herd areas
  • > 1500 hrs heli-time, > 453 days fixed-wing flights
  • > 40,000 animal locations (200 collared caribou, 78 moose, and 33 wolves)
  • > 500 sites investigated with ~ 1500 individual visits
  • > 176 mortalities investigated
  • ~ $900,000 per year over 4 years

5. CARIBOU HABITAT ASSESSMENT & SUPPLY ESTIMATOR (CHASE):

a) This model forecasts through time the distributions of seasonal caribou ranges and timber supply by linking:
  • Caribou and seasonal range information
  • Predation risk factors
  • Forest development planning and silviculture activities
b) The CHASE model predicts:
  • Moose Density and Wolf Predation
  • Movement Corridors
  • Pine-Lichen Winter Range (PLWR)
  • Calving and Summer Range
  • High-Elevation Winter Range
c) Uses Bayesian Belief Networks:
  • The simplified belief network shown to the right displays the primary factors that influence the probability that a caribou will show preference for a 1 ha (2.5 ac) site as PLWR
Pine-Lichen Winter Range

Pine-Lichen Winter Range



Simplified PLWR Network

Simplified PLWR Network

6. PINE-LICHEN WINTER RANGE AVAILABILITY TEST RESULTS

  • OBJECTIVE A - To estimate and test caribou preference for PLWRs, which is where forestry has the greatest interaction with caribou
  • Caribou management policies guided the harvest simulator in forecasting current & future development over 200 yrs in the Wolverine area. This data was then modeled in Netica to predict the distribution of Range Availability states (before predation effects).
  • Winter locations (n = 2202) of 36 collared caribou (Fig. 1) were used to test the hypothesis that caribou would show preference for model-predicted Available ranges, avoidance of Unavailable ranges, and neutral (equal) use of Reduced Availability ranges
  • Location data were grouped by individual animal and year into 82 observations of range preference with a minimum of 5 locations per Range Availability state
  • We tested for preference (i.e. departure from equal use vs. availability) using Chesson’s alpha value, which is the ratio of range use and range availability corrected for unequal availabilities of the range states (Equation 1)
  • To determine the 3 Range Availability states, the model calculated: E(x) = summation(i-h) x p(x), (Eq’n 2) where x is the expected mean state value (+1, 0, -1 for the 3 states), and p(x) is the probability of that state occurring. Results were divided into 3 equal classes (Table).
  • We found that 65% (i.e. 35% error) of the observations corresponded with the areas that the model predicted would be available, reduced availability or unavailable

Figure 1

7. POLICY ASSESSMENT

  • OBJECTIVE B - To evaluate the efficacy of the following combination of current conservation policies in providing a sustainable supply of caribou seasonal ranges:
    1. Parks - disturbance is minimal; probable fire suppression and no timber harvest;
    2. Large Patch - timber harvest is guided toward large patches (i.e, large areas of clear-cuts >1000 hectares and comparably sized temporary or 70 year-leave patches);
    3. Restricted - only light (i.e., less than 30% volume removal) harvest is allowed.
  • In addition to the timber-harvest scenario (Fig. 1) we also modeled potential PLWR availability (Fig. 2), i.e. predation effects not included, and all stands were set to 70-140 years (optimal age for terrestrial lichen production)
  • The amount of current Available range within policy areas is 63%; however, only 47% of potential Available range occurs within the policy areas
  • Within policy areas, a large component of the land base is managed as a park, where stand age will increase and terrestrial lichens will be replaced by mosses, such that habitat value to caribou is significantly reduced
  • The forecasted supply of pine-lichen winter range is likely to decrease sharply over the next 80 to 100 years (Fig. 2, graph) as we see the current atypically high supply of available range grow into older age classes or get harvested

Figure 2

8. CONCLUSIONS

Despite the amount of continuous work still required, modeling caribou habitat supply has provided a basis for interpreting the implications of our policy options:

Caribou
  1. Preservation policy of parks will not help us maintain a large amount of Pine-Lichen Winter Range;
  2. Management of caribou range could be improved by including more or all areas currently outside policy;
  3. Although we’re addressing a high proportion of current PLWR within policy areas, we can expect a dramatic decrease in the amount of range over the next 100 years;
  4. While this poster does not describe predation effects, it has a strong influence on how the model predicts range value