Platte River Wetland Hydrology Study

Home Browse Search

involved. This groundwater recession appeared to fit an exponential decay curve described by the following equation (Viessman et al. 1977): APIs = K(APIt_d + Pt where APIs = antecedent precipitation index for the current time period APIt_1 = antecedent precipitation index from the previous time period (assumed = 0 at the start) K = recession constant, empirically derived Pt = total precipitation for the current time period () The recession constant K is normally reported in the range of 0.85 to 0.98 for modeling soil moisture, with higher values approaching an additive effect for precipitation. For modeling the water table recession at Crane Meadows, however, a value of 0.60 provided the best fit to the data. Increasing K from 0.5 to 0.9 increased the correlation with the groundwater level, but it also increased the correlation with the river stage. One notable exception to this observation was for the month of April, when increasing K seemed to dramatically increase the correlation with the groundwater level while also decreasing the correlation with the river stage. Different values of K were tried for each month, but a fixed value of 0.60 consistently improved the correlation with the groundwater level while minimizing the increased correlation with the river stage. Since 0.60 fit the data best and provided the best balance between correlations, a fixed value of 0.60 was used for all analysis. River stage and groundwater levels were rescaled by subtracting a constant to reduce their range compared to their mean, (e.g., instead of using the elevation above sea level, the elevation was adjusted by subtracting a constant such as 1,800 ft). PROC CORR (SAS Institute Inc. 1985a) was used to calculate the correlations. Missing Data Estimation Missing data for precipitation, mean daily groundwater elevations, and river stage were estimated to provide a complete record of continuous hydrologic data and to calculate the 25