Laserfiche WebLink
showed a preference for agricultural lands, presumably for feeding, but this preference was <br />limited to those lands in proximity to preferred channel widths and depths. <br />The predictive capabilities of the habitat models are limited and we caution that specific <br />estimated values should be viewed in light of the width of the corresponding confidence <br />intervals. Other models, with different predictor variables, would lead to slightly different <br />conclusions. We do support the use of this model to indicate general direction of increasing or <br />decreasing probability. <br />RECOMMENDATIONS <br />1) In regards to the detectability trails, Pollock et al (2002) assert that detection probabilities <br />are not constant for wildlife monitoring and detection probabilities should be estimated as <br />part of a scientifically rigorous monitoring design. We agree with this approach and <br />recommend continuing the implementation of the detectability trials each survey season. <br />2) Habitat selection analyses should be reanalyzed when updated data become available. <br />Examples of current data that would be useful include GIS coverages of: <br />a. vegetation community types (our analysis used land use/land cover from the 1998 <br />photographs), <br />b. areas that have received Program management for whooping crane habitat, and <br />c. areas that have received other land management practices for whooping crane <br />habitat development/restoration (i.e., management of conservation lands not a part <br />of the Program). <br />3) HECRAS appears to be a suitable approach to estimate channel characteristics at in- <br />channel use sites and available locations at the time of crane use. We recommend this <br />approach for future analyses, although with the inclusion of the entire study area. We <br />agree with the data collection recommendations made by Fotherby and Russell (2008), <br />future measurements of the channel should be made with survey grade GPS and tied to <br />vertical control. <br />4) The index of use should be used to evaluate the occurrence of whooping cranes in the <br />study area in response to Program management. The interpretation of these results would <br />be improved if an estimate of crane use could be determined for one or more additional <br />locations along the migration route. <br />18