| The factors affecting non-point source nitrate pollution potential in part of the Equus Beds Aquifer in Central Kansas were studied. Factors were selected for study based on a review of available literature, field research, and four basic criteria. (1) The factor data must be quantitative (ordinal data or better). (2) The factor data must be available in geographic information systems (GIS) format or they must be easily created from existing datasets. (3) The factor data must exhibit spatial variability. (4) There must be a sound, physically based reason for the factors to affect groundwater nitrate pollution.; A total of seventeen factors from five broad categories were selected for statistical comparison with groundwater nitrate concentration data from wells. GIS coverages were obtained or generated for each of the seventeen factors. A raster grid for each factor was created from these GIS coverages. Circular buffers with five different diameters from 100 to 1600 meters were generated around each well point where recent nitrate concentration data was available. Average values for each of the seventeen factors were calculated for each buffer diameter around each sampled well. Average factor values were statistically compared with the nitrate data using Spearman Rank correlation coefficients.; Nine of the seventeen factors studied had consistent and statistically significant correlations (alpha = 0.02) with nitrate. The strongest correlations were observed between nitrate and aquifer chemistry, lithology, landuse, and soil factors. Two statistically significant relationships were notable because they did not exhibit the expected trend. Nitrate concentrations tended to increase with decreases in estimated fertilizer application rates. Also, nitrate concentrations tended to increase as the clay content at shallow depths (<50) increased. The reasons for these apparently spurious correlations are not known at this time.; A simple model using three factors (maximum soil clay content, percent clay from 25--50 feet below ground, and dissolved iron concentrations) was developed and spatially compared with historical nitrate concentration data. Apparent anomalies between the model and the nitrate concentration data probably result from inaccuracies in the factor data, proximity to point sources of nitrate, and other site factors such as well depth and construction. |
