Development and evaluation of a record extension technique for estimating discharge at selected stream sites in New Hampshire

Daily mean discharges are needed for rivers in New Hampshire for the management of instream flows. It is impractical, however, to continuously gage all streams in New Hampshire, and at many sites where information is needed, the discharge data required do not exist. For such sites, techniques for estimating discharge are available. The U.S. Geological Survey, in cooperation with the New Hampshire Department of Environmental Services, developed and evaluated the accuracy of estimated discharge records for six discontinued U.S. Geological Survey streamgages in New Hampshire.

The estimated records were developed by using the maintenance of variance extension, type 1 (MOVE.1), record extension technique and were generated for periods with concurrent observed records to allow for evaluation. The six discontinued streamgages were on New Hampshire designated rivers throughout the State and had drainage areas ranging from 35.6 to 395 square miles with little to no regulation.

Estimated records for four of the six streamgages had Nash-Sutcliffe efficiency coefficients greater than 0.85. The other two streamgages had Nash-Sutcliffe efficiency coefficients between 0.45 and 0.60. For the four streamgages with the higher Nash-Sutcliffe efficiency coefficients, more than 35 percent of the estimated record was within 15 percent of the observed record. At the other two streamgages, more than 23 percent of the estimated record was within 15 percent of the observed record.

At lower discharges (exceeded 80 percent of the time), for four of the six streamgages, more than 40 percent of the estimated record was within 15 percent of the observed record. The site with the lowest Nash-Sutcliffe efficiency coefficient had more than 14 percent of the estimated record at low discharges within 15 percent of the observed record.