Third, if actions are undertaken on Harlequin streams, they should be monitored to determine 

 there effects. This allows us to learn the effects of various actions and not repeat mistakes in the 

 future. To determine effects of specific land management or development actions on Harlequin 

 streams, the action should be proceeded by at least two years of baseline marking and surveying 

 for population size and productivity, habitat evaluation, and pre-action levels of human activity 

 and development. The population should be monitored during and following the action to 

 determine the effects of the action. Actions which particularly need attention include road, 

 campsite, and trail construction and upgrading, including any increased accessibility and changes 

 in human use of the area; actions which could resuh in changes to flow regimes or water quality, 

 such as mining, road building, timber harvest, industrial development, and water/hydroelectric 

 development; changes in fishing regulations which could change fishing use of the area; building 

 of structures such as industrial areas, dams, or houses which will increase the access and use of a 

 Harlequin stream. Possibilities for mitigation and habitat restoration can be explored during these 

 projects. 



2) What is the extent and nature of movements in breeding and wintering areas? 



This information is needed to determine the possibilities for naturally recolonizing new and 

 historic Harlequin occurrences; naturally supplementing existing occurrences, particularly small 

 populations, and how strong natal and adult fidelity to particular sites are. All these are needed to 

 successfully model Harlequin populations and their stability, with both breeding and wintering 

 grounds data incorporated. 



Radio-telemetry may give quick results from the standpoint of local daily movements, 

 however, I expect that long distance (>5 km) movements are relatively rare and with limited 

 numbers of ducks radioed may not be best for long distance movement detection. For long 

 distance and moves between years, visibly marking birds is best. For association to natal areas, 

 this will be a long term project, but Montana has the best start with 250 birds banded on the 

 breeding grounds since 1992. With the start we have now, I estimate sufficient information for 

 preliminary modeling will be available following the 1996 field season if funding is received, and 

 good information could be available following the 1998 field season for final modeling. 



For the wintering grounds, much data is available and is currently being collected in 

 Washington, Alaska, and British Columbia. It should be available within 2 years with sufficient 

 information to use in detailed population modeling. For an accurate model, information is 

 necessary from both the breeding and wintering grounds. See Proposals 1, 4, 5. 



3) Are distinct metapopulations (such as a Rocky Mountain breeding population) identifiable 

 within the Pacific range of the Harlequin Duck? 



The degree of genetic differences among and within wintering and breeding subpopulations would 

 allow an assessment of the appropriate management units for various Harlequin conservation 

 strategies. Dan Esler, Alaska National Biological Service is currently examining this question, 

 primarily for wintering areas, but he is also examining blood samples we are providing from 



23 



