Characterization of factors affecting groundwater levels in and near the former Lake Traverse Indian Reservation, South Dakota, water years 1956–2017

First posted May 19, 2021 For additional information, contact: Director, Dakota Water Science CenterU.S. Geological Survey821 East Interstate Avenue, Bismarck, ND 585031608 Mountain View Road, Rapid City, SD 57702 The U.S. Survey (USGS), in cooperation with the Sisseton Wahpeton Oyate, completed a study to characterize water-level fluctuations observation wells relative driving factors that affect water levels and near historical 1867 boundary of Lake Traverse Indian Reservation. investigated concerns regarding potential effects groundwater withdrawals climate conditions on within an area includes reservation surrounding extends 10 miles all directions South Dakota. Characterization was accomplished by statistical trend analysis.Monthly data from Parameter-elevation Regressions Independent Slopes Model (PRISM) were aggregated obtain annual seasonal datasets for total precipitation, minimum air temperature (Tmin), maximum (Tmax) buffer area. Trend tests gridded Tmin, Tmax time series years 1956–2017, which is about 2 before earliest available measurements. A 2-year offset arbitrarily selected because scrutiny precipitation indicated responses many lagged major changes patterns years. Statistically significant upward trends detected Tmin most downward only few 2.5 arc-minute grid cells; however, sparsity spatial coverage reduces confidence these are true trends, contrast completeness Tmin. Spatial distributions statistically generally similar but substantial differences density trends.Potential interactions among streamflow examined through correlation analyses median level each 76 versus mean four streamgages. Potential lake involving 25 lakes. Resulting coefficients used as part approach selecting be plotted conjunction well.Groundwater analyzed three separate parameters (minimum, median, maximum) measured sporadically, biased towards more frequent measurements during periods heaviest irrigation demand. Trends (from PRISM) starting earlier than associated also location individual well. Sen’s slope Mann-Kendall p-values computed series. Graphs showing results well showed sum licensed six specified radii (0.5, 1.0, 2.0, 3.0, 4.0, 5.0 miles) qualitative indicator proximal demand.Trends predominantly upward, 43 having at least one 8 parameter. driven (not same wells) no trends. Significant two eight Groundwater some likely substantially affected surface water, especially lakes increased response wet early 1990s have maintained high ever since. It recognized cases plotting hydrographs not hydraulically connected system or aquifer well; plausible numerous other records available.