Topographic and physicochemical controls on soil denitrification in prior converted croplands located on the Delmarva Peninsula, USA

Abstract Topographic and soil physiochemical characteristics exert substantial controls on denitrification, and the effect of these controls is especially evident in fertilized agricultural lands. To depict these controls at a landscape scale for decision support applications, metrics (i.e., proxies) must be developed based on commonly available geospatial data. In this study, we carried out an observational study on denitrification potential (DP) and capacity (DC) in three actively farmed crop fields that were converted from forested wetlands (i.e., prior converted croplands). The combined effects of ten topographic and physiochemical factors, including three topographic attributes (relief, topographic wetness index, and positive openness), two soil texture indices (sand and clay), and five soil properties (soil moisture, pH, electrical conductivity, soil organic carbon and total nitrogen), on DP and DC were analyzed. The three topographic attributes were developed using a digital elevation model (DEM) derived from light detection and ranging (LiDAR) data. Nitrate and carbon addition led to a doubling in DP compared to DC without soil amendment. Topography explained the greatest amount of variation in DP across the three sites. The relationship between topography and DP may partly be explained through the relatively robust relationships between topography and soil moisture, texture, and carbon content. Soil electrical conductivity (EC) exhibited the highest correlation with DC ( r 2  = 35%). DP and DC were higher under drought conditions with low soil moisture, relative to average conditions with relatively higher soil moisture, which may be related to the substantial increase in soil EC under drought conditions. However, DP and DC were less responsive to soil EC at sandy sites that tended to have low soil moisture. Results of this study suggest that the spatial-temporal variations in denitrification at these croplands were primarily caused by complex interactions between soil properties and landscape position. Topographic metrics derived from LiDAR data have the potential to improve understanding of denitrification variability at the landscape scale.