Whooping Crane Migrational Habitat Use Draft ‎(2)‎

• File Number: 8461.100
• Description: Adaptive Management Workgroup
• State: CO
• Basin: South Platte
• Water Division: 1
• Date: 8/6/2008
• Author: Shay Howlin, Clayton Derby, Dale Strickland, Western Ecosystems Technology, Inc.
• Title: Whooping Crane Migrational Habitat Use Draft
• Water Supply Pro - Doc Type: Report/Study

Means were estimated for each habitat parameters measured on the ground with the systematic
observations (Table 11). The percentage of fine sand was 61% (95% CI: 51, 71), the percentage
of coarse sand was 30% (95% CI: 21, 39), the percentage of small gravel was 8% (95% CI: 4,
12), and the percentage of large gravel was <1% (95% CI: 0, 0.90). The average distance to
visual obstruction was 285 meters (95% CI: 219, 351). The average number of acres clear of
visual obstruction was 54 (95% CI: 40, 67). The average obstruction to obstruction width was
407 meters (95% CI: 296, 519). The means estimated from the combined systematic and
opportunistic samples are in Appendix B.
Means were estimated for each flow dependent habitat parameter with the systematic
observations (Table 12). The average wetted width was 246 meters (95% CI: 206, 286). The
average depth was 0.35 meters (95% CI: 0.29, 0.42), which is equivalent to 1.10 feet (95% CI:
0.95, 1.38). The average flow was 913 cfs (95% CI: 748, 1078). The average sandbar elevation
was 0.31 meters (95% CI: 0.22, 0.40), which is equivalent to 1.02 feet (95% CI: 0.72, 1.31). The
means estimated from the combined systematic and opportunistic samples are in Appendix B.
Means were estimated for each land cover habitat parameter, habitat parameter measured on the
ground, and flow dependent habitat parameter for the available samples used in the estimation of
resource selection models (Appendix C). The models estimated from the combined systematic
and opportunistic samples are in Appendix D. Steps for the model selection procedure for the
systematic sample of observations are in Appendix E.
Study area selection of land cover
We predicted the relative probability of selection for the study area based on the 1998 CIR
photographs (Figure 18). The final resource selection model, w(x), for all observations (wetted
channel and upland) contained the linear effects of percent of wetted channel within 0.25
miles(P_WC), percent of open water within 0.25 miles(P_OW), and percent of agriculture within
0.25 miles (P_AG). The form of the final model was:
W(x) = exp[(0.078*P_WC) + (0.300*P_OW) + (0.017*P_AG)]
The standard errors for the coefficients were 0.01, 0.05, and <0.01 for percent of wetted channel,
percent of open water, and percent of agriculture respectively. This model indicated significant
selection for areas with large proportions of wetted channel, open water, and agriculture,
regardless of the observation being in the wetted channel or out of the channel.
Study area in-channel selection of land cover
We predicted the relative probability of selection for the in-channel area based on the 1998 CIR
photographs (Figure 19). The final resource selection model, w(x), for in-channel observations
contained the linear effects of percent of open water within 0.25 miles (P_OW), the percent of
wetted channel within 0.25 miles (P_WC), the percent of agriculture within 0.25 miles (P_AG),
distance to transportation (trdist), and the linear and quadratic effects of distance to shrub and
forest (sfdist). The form of the final model was:
12