Correlating Denitrification Rates to Soil Texture using Hierarchical Cluster Analysis

Abstract

Widespread pollution of ground and surface waters from nitrate (NO3-) contamination is of global concern to both public health and the environment. Quantifying the attenuation potential of nitrate in the sub-surface is thus very important. Denitrification rates have been found to vary significantly both spatially and temporally depending on different soil characteristics. Current knowledge of sub-surface denitrification is site specific; thus limiting its use for making NO3- removal predictions on a watershed scale. In the present work, attention has been focused on assessing the influence of soil textural classes on denitrification. Data were compiled and analyzed from the literature of relevant work, and the variability in denitrification rates was assessed. A multivariate statistical approach was used to determine if denitrification rates could be correlated to soil textural classes. In addition, cumulative frequency diagrams (CFD) proved useful for quantifying the variations in denitrification rates. The hierarchical cluster analysis (HCA) provided valuable information regarding the similarities between the different soil experiments by grouping them together into three distinct groups of experiments that were closely related to the grain size distribution. Although the HCA did not provide any information how the three different soil groups are related to denitrification rates, the CFD showed that the denitrification rates vary significantly with soil group. Clayey soils were found to have the highest denitrification rates, while sandy and silty soils have the lowest. This work will allow for more realistic assessments of denitrification on a watershed scale.