Skip to main content

Srinivasan, Raghavan

The cascade correlation neural network was used to predict the two-year peak discharge (Q2) for major regional river basins of the continental United States (US). Watersheds ranged in size by four orders of magnitude. Results of the neural network predictions ranged from correlations of 0.73 for 104 test data in the Souris-Red Rainy river basin to 0.95 for 141 test data in California. These results are improvements over previous multilinear regressions involving more variables that showed correlations ranging from 0.26 to 0.94. Results are presented for neural networks trained and tested on drainage area, average annual precipitation, and mean basin elevation. A neural network trained on regional scale data in the...
thumbnail
The potential effects of mid-21st century climate change on the hydrology of the Cook Inlet watershed in south-central Alaska was analyzed in this study. Climate datasets representing a set of potential change scenarios for the period 2041–2070 were developed from the North American Regional Climate Change Assessment Program (NARCCAP) archive of dynamically downscaled climate products. The NARCCAP 50-km scale regional climate output was converted to realistic daily weather time series using a “change factor” method in which observed meteorological time series used for model calibration are perturbed. The perturbations are based on statistical summaries of change for the different climate scenarios, by month, as...
Global climate change will impact the hydrologic cycle by increasing the capacity of the atmosphere to hold moisture. Anticipated impacts are generally increased evaporation at low latitudes and increased precipitation at middle and high latitudes. General Circulation Models (GCMs) used to simulate climate disagree on whether the U.S. as a whole and its constituent regions will receive more or less precipitation as global warming occurs. The impacts on specific regions will depend on changes in weather patterns and are certain to be complex. Here we apply the suite of 12 potential climate change scenarios, previously described in Part 1, to the Hydrologic Unit Model of the United States (HUMUS) to simulate water...
ScienceBase brings together the best information it can find about USGS researchers and offices to show connections to publications, projects, and data. We are still working to improve this process and information is by no means complete. If you don't see everything you know is associated with you, a colleague, or your office, please be patient while we work to connect the dots. Feel free to contact sciencebase@usgs.gov.