Laserfiche WebLink
private ownership, thus our sample data collection will primarily focus on public lands (Nebraska Game <br />and Parks Commission), as well as properties owned and managed by non-governmental organizations <br />(The Nature Conservancy, Whooping Crane Maintenance Trust). <br />Field Work <br />Technicians will utilize Trimble Geoexplorer XT GPS units installed with ArcPad to check <br />polygon data out for field assessment. Once in the field the technicians will utilize the GPS to navigate to <br />polygons and class the vegetation. Once field data collection is complete data will be checked back into <br />the master project. This master project organizes all of the data and ensures that technicians will not <br />double edit data and that all areas requiring sampling will be visited. <br />Image Classification <br />Utilizing the training data collected by the technicians a testing and training mask (TTA mask) <br />will be created. TTA masks can then be integrated into eCognition to complete a supervised <br />classification. A TTA mask utilizes the field training data to create sample signatures far the defined <br />vegetation classes. This data is integrated into a nearest neighbor classification strategy that classifies the <br />imagery to the appropriate classes. The output can either be a vector or raster output. <br />Accuracy Assessment <br />The accuracy assessment process is completed after a final photo interpretation of the nearest <br />neighbor classification. During the photo interpretation features are separated into finer mapping units <br />that can not be remotely sensed using eCognition. An example of this is separating the water bodies into <br />the following classes: warm water slough, sand pit, stock dam, irrigation reuse pit. Also during the photo <br />interpretation obvious misclassifications are reclassified and input from filed staff is integrated to correct <br />as much error as possible. Once these final improvements are completed a subset of the field data that <br />was with held from the TTA mask creation is used to assess the accuracy of the vegetation classes. From <br />this assessment an accuracy assessment table is created that outlines errors of omission and commission <br />as well as overall accuracy. <br />Final Vegetation Map <br />Once the accuracy assessment process is completed, the dataset is prepared for distribution <br />assuming accuracy requirements are met. The data is prepared in both hardcopy format as well as digital <br />format. This includes of vector and raster datasets. The raster data will be integrated into the HABS <br />landcover and used to assess current habitat provided by the corridar as well as develop estimated <br />carrying capacity far species of concern. <br />Platte River Recovery Implementation Program Benefits <br />The PRRIP would benefit significantly by contributing to this ongoing project. With a partial <br />contribution to help finalize this work, a complete and highly detailed land cover system will be available <br />to use for program implementation activities for the entire program area. The requested funds are only a <br />fraction of what the entire project would cost if initiated from scratch. The Adaptive Management <br />Workplan for 2007 identifies several tasks that would get accomplished by supporting completion of this <br />project and includes: Mapping and identifying wet meadows; developing a priority list of locations for <br />different types of experiments and monitoring; database management and development; identify historic <br />land clearings; and identifying grassland development and restarations. Many other research, <br />management, and monitoring activities identified in the work plan for future years would benefit greatly <br />by having the current land cover system available for planning and development of on the ground <br />projects.