Optimized number of suction cups required to predict annual nitrate leaching under varying conditions in Denmark

Abstract

Nitrate concentrations in soil water leaving the root zone measured by suction cups combined with water transport modeling is a commonly used practice in Denmark for calculating nitrate leaching. Two suction cups installed in one plot giving one water sample and replicated four times, (eight total suction cups) to reduce variability between samples. For practical reasons, it would be beneficial to minimize the number of suction cups used yet maintain reliable predictions. To assess the variability in reducing suction replicates, this study analyzed data from five research sites across Denmark representing annual field nitrate leaching predictions for different combinations of soil, weather conditions, crops, N-fertilizer rates, and winter soil cover, covering a total of 173 annual nitrate leaching experiments. The analysis was conducted having different nitrate leaching predictions using different numbers of replicates of suction cup measurements. Linear regression was used to identify the different influences of leaching year (hydrological year), N rate applied, soil characteristics, and crop sequence on nitrate leaching. The analyses were set up on three 2-yr and two 3-yr field experiments in five different sites. Crop effects showed that cereals and winter cover sown in autumn 2017 had significantly more nitrate leaching than in 2015 and 2016 leaching years due to high precipitation rates in the autumn. Furthermore, decreasing the number of suction cup replicates from four (eight total) to three replicates (six total) did not have a significant effect on nitrate leaching prediction. In contrast, decreasing replicates from four to two (four total) and one (two total) replicates did show a significant difference. Therefore, using three replicates is a viable solution for future sampling strategies and a good trade-off between costs and accuracy.