Soil health across a Kansas precipitation gradient

Abstract

This project used the precipitation gradient across Kansas from less than 400 mm in the west to over 1000 mm in the east with three land uses. The locations, Tribune (472 mm), Hays (579 mm), and Manhattan (850 mm), KS, were selected for soil characterization. The land uses included native prairie (NP), conventional tillage agriculture (AG), and no-till agriculture (EA). Soil biological, chemical, and physical properties were measured based on USDA-NRCS chosen soil health metrics. The objectives of this research were to 1) assess differences in soil health metrics across the precipitation gradient with three land uses, 2) compare soil metric methods for wet aggregate stability and infiltration, and 3) assess soil health metrics to a one m depth. Several soil health metrics distinguished differences between land use and precipitation to a one m soil depth. Soil organic carbon and total nitrogen stocks increased with increasing precipitation. Mean weight diameter was higher in NP, than EA and AG at all locations. Soil microbial biomass was generally higher in NP than cropping systems with increasing precipitation. Surprisingly, EA in Hays had greater soil health metrics than EA Manhattan due to the inclusion of cover crops in no-till at Hays. Positive correlations between biological properties, explained 77.3% of the data variance as determined by principal component analysis. The aggregate methods tested included Mikha and Rice (2004), Kemper and Rosenau (1986), and Soil Survey Staff (2014). All methods were highly correlated to one another with significant p-values. The Bland-Altman plot showed agreement between Mikha and Rice (2004) and Kemper and Rosenau (1986) with relative random error. The choice of aggregate stability method ultimately depends on the level of sensitivity needed, available resources, labor, costs, and data of interest, with the Mikha and Rice (2004) method providing the most robust information. Cornell sprinkler infiltrometer and single infiltration methods were variable in results by location. The single ring method was more sensitive to land disturbance as the infiltration rate increased with decreased soil disturbance. Variability and method approaches makes both infiltration methods difficult to recommend. The single ring method was more portable and required less material, training, maintenance, and water. The Cornell Sprinkler Infiltrometer can simulate a range of rainfall rates, designed to measure runoff and infiltration, and reduces unnatural macropore flow from ponding (Cornell University, 2019). Soil health metrics to assess differences in land use detected surface and subsurface changes. Several soil properties were highly correlated, such as alkaline phosphatase and phosphodiesterase; or soil organic carbon and soil respiration. A single sampling time can not be used to compare fields with differing management. However, a single point in time measurement for the prairie across the precipitation gradient could be used to assesss soil health properties. Thus, the best approach to document soil health is to track changes in a single field rather than compare sites.