 |
|
|
||||
| Water-Resources Investigations 96-4001 |
Download this report as a PDF file (3.3 mb)
Download a free copy of Acrobat Reader
By Timothy P. Brabets
Prepared in cooperation with the
Alaska Department of Natural Resources and U.S. Forest Service
In 1906, the U.S. Geological Survey (USGS) began operating a network of streamflow-gaging stations in Alaska. The primary purpose of the streamflow-gaging network has been to provide peak flow, average flow, and low-flow characteristics to a variety of users. In 1993, the USGS began a study to evaluate the current network of 78 stations. The objectives of this study were to determine the adequacy of the existing network in predicting selected regional flow characteristics and to determine if providing additional streamflow-gaging stations could improve the network's ability to predict these characteristics.
Alaska was divided into six distinct hydrologic regions: Arctic, Northwest, Southcentral, Southeast, Southwest, and Yukon. For each region, historical and current streamflow data were compiled. In Arctic, Northwest, and Southwest Alaska, insufficient data were available to develop regional regression equations. In these areas, proposed locations of streamflow-gaging stations were selected by using clustering techniques to define similar reas within a region and by spatial visual analysis using the precipitation, physiographic, and hydrologic unit maps of Alaska.
Sufficient data existed in Southcentral and Southeast Alaska to use generalized least squares (GLS) procedures to develop regional regression equations to estimate the 50-year peak flow, annual average flow, and a low-flow statistic. GLS procedures were also used for Yukon Alaska but the results should be used with caution because the data do not have an adequate spatial distribution.
Network analysis procedures were used for the Southcentral, Southeast, and Yukon regions. Network analysis indicates the reduction in the sampling error of the regional regression equation that can be obtained given different scenarios. For Alaska, a 10-year planning period was used. One scenario showed the results of continuing the current network with no additional gaging stations and another scenario showed the results of adding gaging stations to the network. With the exception of the annual average discharge equartion for Southeast Alaska, by adding gaging stations in all three regions, the sampling error was reduced to a greater extent than by not adding gaging stations.
The proposed streamflow-gaging network for Alaska consists of 308 gaging stations, of which 32 are designated as index stations. If the proposed network can not be implemented in its entirety, then a lesser cost alternative would be to establish the index stations and to implement the network for a particular region.
| AccessibilityFOIAPrivacyPolicies and Notices | |
  |
|