Evaluating trends in streamflow extremes in the Northeast USA

The goal of this project is to identify statistical trends in observed and simulated maximum, minimum and base (mostly groundwater contribution during low flow months) flows in the Northeast Climate Science Center domain during the 20th and 21st century, assess the temporal (annual and seasonal) and spatial distribution of the trends, and evaluate the impact of warmer climates on the statistical properties of streamflows (mean and variance). A secondary goal is to determine what GCMs best represent the observed climatology of the region using statistical metrics.

Base and minimum flows are vital for fish ecosystem functioning and for riparian vegetation. Climate projections indicate summers will get warmer and drier in the NE CSC domain which will affect aquatic ecosystems. Larger streamflows peaks will affect existing infrastructure, e.g. bridges, dams, cities).  

Keith Kuk-Hyun Ahn completed a non stationary analysis for streamflow in the connecticut river basin, developing regional flood frequency analysis using spatial proximity and basin characteristics in the northeastern United States, and developed drought forecasting methodology using climate indices. He also built the SWAT model for Deerfield basins.

Nirajan Dhakal's findings include the following:

1. The high and low flow seasonality of catchments in the northeastern Unites States is likely to change under climate change because of anticipated alterations of precipitation as well as snow accumulation and melt. Information on this change is crucial for flood protection polices for example for regional flood frequency analysis. The objective of this study is to evaluate the change in seasonality of daily stream flows (both high and low) in northeastern Unites States.                                                                                                       
2. In areas where high flows are generated by more than one distinct hydrologic process (e.g., snowmelt- and rainfall generated peaks), peak discharge data should be considered to be drawn from subpopulations with different statistical characteristics. For an appropriate and effective hydrologic design it is necessary to consider these two characteristics of precipitation to estimate the probability distribution by introducing mixed distribution.