Evaluating Normal Rainfall for Wetland Hydrology Assessment.
J.P. Sumner, NC State Univ, Soil Science Dept, Box 7619, Raleigh, NC 27695, Michael Vepraskas, Dept of Soil Science, Box 7619, North Carolina State Univ, Raleigh, NC 27695, and Randall Kolka, USDA-Forest Service NC Research Station, 1831 E US Highway 169, Grand Rapids, MN 55744.
It is assumed, but not proven, that wetland hydrology can be evaluated with single-season measurements of water table data, if antecedent rainfall is within a normal or drier than normal range. Two methods of rainfall analysis were compared to long-term records of water table levels to determine which method could be used to identify sites with wetland hydrology. Water table for three sites in NC and MN were obtained that had at least a 40-yr record of data. Plots with wetland hydrology had a water table within 30 cm of the surface, during the growing season, in at least half the years. Years meeting wetland hydrology were identified for each site. Normal rainfall was determined from using the 30th and 70th percentiles obtained from a WETS data set for the nearest available weather stations. Two rainfall-assessment methods were compared. The first computed a 30-day moving rainfall total that was compared with a normal range of rainfall values obtained from the WETS data. The second approach used a 3-month period of antecedent rainfall along with the WETS data to determine whether a given period had a normal range of rainfall. Plots meeting wetland hydrology met it in over 92% of the years at all three sites. For single-year data, the moving total method provided the correct conclusion in less than 62% of the years that met wetland hydrology. The second method that uses rainfall for the prior 3-month period was more accurate, and identified plots with wetland hydrology in over 84% of the years. The results showed that when single-season data are used to identify wetland hydrology, then no more than two years of measurements will be needed to reach a correct conclusion, and in most cases a single season of data may be used.