Laserfiche WebLink
Groundwater depths were calculated by subtracting the groundwater elevation from the surface <br />elevation. River stage was presented as an elevation above mean sea level. Boxes were <br />constructed to represent the median (connected by a line through each month), the middle 50 % <br />of the observations (the box), the lower 10 to 25 % and the upper 75 to 90 % of the <br />observations (the whiskers), and the lower 0 to 10% and upper 90 to 100% of the observations <br />(individual data points). For box -and- whisker plots based on continuous observations (i.e., <br />wells and gages with recorders), each box shows the percent of time during the month that a <br />water level was maintained or above (similar to the duration curves). Periodic observations <br />were used for wells and gages without recorders (Appendix A), but the data must be <br />interpreted as the percentage of observations instead of the percentage of time. <br />Depth to Groundwater and River Stage Duration Curves <br />Depth to groundwater and river stage duration curves were generated from the <br />continuous daily mean groundwater depths and river stage. These curves are cumulative <br />frequency distribution curves that show the percent of time that a particular depth or stage was <br />equaled or above that level for the period specified. Values are typically expressed as a <br />duration followed by a depth or stage. For example, D90 = -3.25 ft means that 90% of the <br />time the water level was at or above 3.25 ' ft below the surface. The duration curves were <br />calculated using procedures similar to those used to calculate a typical flow duration curve for <br />streamflow analysis (Searcy 1959). Duration curves are useful for predicting the availability <br />and variability of sustained levels, but they do not represent the actual sequence of observed <br />events (Viessman et al. 1977). <br />PROC FREQ (SAS Institute Inc. 1985a) was used to generate the duration data and <br />PROC GPLOT (SAS Institute Inc. 1990b) was used to graph the data. Graphical data were <br />based upon 0.01 ft size classes for more accurate representations. <br />River Stage and River Flow Relationships <br />The river flow (Q) for a given river stage at each site was estimated by using a simple <br />power function: <br />18 <br />