Consulting Solutions that Work for People and the Environment

Upper Connecticut River Site-Specific Probable Maximum Flood Development


Gomez and Sullivan is currently undertaking a project to develop site-specific Probable Maximum Floods (PMFs) for three hydroelectric projects. This project will serve to update the PMF for these three sites which are located on the Connecticut River; the longest river in New England. The projects are located in the upper portion of the Connecticut Basin, with a total project drainage area of approximately 3,400 square miles situated between the White Mountains and the Green Mountains. Site-specific PMP and PMF studies can be important in assessing dam-safety concerns.

Due to the mountainous location of the study basin, a site-specific approach to determining the Probable Maximum Precipitation (PMP) is ideal for these projects. Standard methods for determining PMP such as Hydrometeorological Report 51 (HMR-51) do not fully address the effects of mountains in blocking moisture inflow. A site-specific study also utilizes actual extreme storms which could occur in the study basin in determining the PMP. For other site-specific studies that Gomez and Sullivan has worked on, the site-specific determination of the PMP has been significantly different from that determined by standard methods.

This project utilized extreme storms identified by prior studies. These storms were maximized in place and transposed to the study basin to estimate the site-specific PMP. The storms are maximized using meteorological data to determine a worst case scenario for producing the storm, whereby the storm is able to attain the most moisture deemed reasonably feasible. To transpose the storm to the study basin, GIS is used to identify topographic barriers which diminish the moisture carried by the storm. The GIS was also used as an aid in determining factors such as maximum dewpoint at the storm source which affect the development of the storm.

In order to estimate the PMF for the study basin, a hydrologic model was developed. The GIS was used in developing the hydrologic model for delineating sub-basins as well as making initial estimates of basin unit hydrograph parameters. The hydrologic model was calibrated to significant storms in the basin and is used to convert the site-specific PMP to run-off, which is routed through the basin for determining the warm-season PMF.

In northern basins such as this, it is possible that the warm season PMP does not lead to the PMF. It is possible that a lesser rainfall combined with snowmelt will produce a higher runoff. For the Upper Connecticut River, the flood of record was a result of a rain on snow event which occurred during March of 1936. This event is the flood of record for many basins in Northern New England.

Since cold-season events have produced many of the high flow events in the basin, Gomez and Sullivan evaluated their efficacy in producing the PMF for the site. A screening analysis suggested that a rain on snow event could potentially cause the PMF, therefore an in-depth analysis is being performed. For the rain on snow event, the PMP is reduced by a seasonal adjustment factor and combined with a computed coincident snowmelt. Snowmelt is computed via an energy budget method. This computation assesses heat transfer to the snowpack via solar radiation, convective transfer caused by wind and heat added through rainfall. In order to compute the snowmelt it is necessary to determine an appropriate temperature and wind sequence to input with the rainfall to the equation. To get an appropriate estimate of the runoff from the rain on snow event, the hydrologic model is re-calibrated to high flow events that were caused by combined rain and snowmelt. To determine the PMF for the site, the higher of the rain on snow and warm-season PMF is utilized.