LAUSR

Abstract Soil health status should be monitored to allow planning sustainable management, but indicators available do not encourage frequent soil health evaluation because of the complexity, time-consumption, and expensiveness of the methodologies. Aggregate stability (AS) is a good soil physical health indicator associated with soil (SOC) and particulate (POC) organic carbon but is difficult to monitor. Anaerobically mineralizable nitrogen (AN) has been proposed as soil health indicator because is cheap, simple, and safe to measure, is sensitive to soil-use changes, is also related to soil (SOC) and particulate (POC) organic carbon, and is frequently determined by farmers in Mollisols of the Southeastern Argentinean Pampas to support soil fertility diagnosis. We hypothesize that AN is positively related to and can be used as indicator of AS. Soil samples were taken at 0–5 and 5–20 cm depths from 46 sites throughout the southeastern Buenos Aires province, Argentinean Pampas. In each site, we sampled Mollisols under continuous cropping (CC) and others that had not been disturbed for many years (pseudo-pristine, PRIS). We determined texture, SOC, mineral-associated organic C, POC, AS and AN. We also calculated variable values for 0–20 cm. Soil organic carbon, POC, AN and AS were reduced by continuous cropping. Anaerobically mineralizable N was positively related to SOC (R2 = 0.74, 0.46, and 0.62 at 0–5, 5–20, and 0–20 cm) and POC (R2 = 0.73, 0.33, and 0.60, respectively). An important proportion of the total variability in AS was explained by SOC (R2 = 0.77, 0.65, and 0.73 at 0–5, 5–20, and 0–20 cm, respectively), POC (R2 = 0.75, 0.63, and 0.73, respectively), and AN (R2 = 0.78, 0.69, and 0.81, respectively). The AS increased with the increase of SOC, POC, and AN at all three depths, with slopes that did not differ between CC and PRIS, but with intercepts that differed. Neither sand nor clay contents significantly contributed to explain the variations in AS as a function of SOC, POC, and AN. An independent validation of the regression model relating AS and AN at 0–20 cm was done and the output was very good (RPIQ (ratio of performance to interquartile distance) = 2.20). Results support our hypothesis because AN was positively related to AS. Consequently, AN would be a good indicator of AS, SOC, and POC. Based on our results, we consider that a simple and cheap soil analysis as AN can not only be used to diagnose soil fertility, but to monitor soil physical and biochemical health status.