Soil and carbon losses from five different land management areas under simulated rainfall

Abstract

The loss of soil and carbon (C) from erosional processes is of greater importance to agricultural productivity and sustainable resource management. Land management practices influence structural properties of soil aggregates, its size and stability, and the stabilization effect of soil organic carbon (SOC), and other cementing agents. Water erosion affects all these properties and consequently decreases productivity of land. Thus, relative significance of loss of soil and C from dry and wet aggregates was studied from soil sampled from the North Appalachian Experimental Watersheds (NAWEP). Aggregates of 2–8 mm from a depth of 0–10 cm were obtained from soils under five different land management areas (LMAs) including forest, pasture, no-till with manure (NT + M), no-till without manure (NT − M), and conventional tillage (CT). Soil and C losses and associated surface runoff, and percolation water were measured using a rainfall simulator at 5 min interval for 1 h using slope gradient (8.5%) under dry and wet runs. Soil and C losses under simulated rainfall conditions varied over time and among different LMAs in both dry and wet runs. In dry runs, highest soil loss of 1.55 kg/m2 was observed in NT − M, followed by 1.34 kg/m2 in CT, 0.96 kg/m2 in NT + M, 0.55 kg/m2 in pasture and 0.51 kg/m2 in forest. The soil loss from NT − M was significantly different from forest and pasture but not of NT + M and CT. General trends in soil loss were similar for dry and wet runs and were in the order NT − M > CT > NT + M > Pasture > Forest. While trends were similar, wet runs merely increased the amount of soil loss. However, the magnitude of soil loss differed significantly among time intervals (P < 0.01). Highest soil loss (0.2 kg/m2) was observed under CT and occurred much earlier in 15 min of rainfall than under LMAs in dry runs. In comparison, the highest soil loss (0.17 kg/m2) was observed in 20 min for NT − M, 40 min for forest, 45 min for NT + M, and 55 min for pasture. In the wet runs however, the highest soil loss commenced after 10 min in all LMAs. In dry runs, highest C loss was in NT + M (0.045 kg/m2), followed by NT − M (0.036 kg/m2), CT (0.016 kg/m2), pasture (0.014 kg/m2), and least C loss was observed in forest (0.0085 kg/m2). Total C loss among time interval in dry runs shows different trends compared to soil loss and highest C loss was observed in 40 min in all LMAs. But in wet runs, C loss was the highest within 5 min of rainfall for all LMAs. Surface runoff in dry runs, differed significantly (P < 0.01) among CT and other LMAs. There was no significant difference in the amount of surface runoff among LMAs from wet aggregates (P < 0.05). Percolation loss also varied among LMAs. The CT treatment had the lowest percolation followed by that from forest, pasture, NT − M and NT + M. The results show that soil wetness is one of the important factors in erosional process, which accelerate higher soil and C loss. Forest, pasture, NT + M management practice areas indicated better possibility to mitigate soil and C loss because soil aggregates from these LMAs were characterized by comparatively higher percolation rate, lesser surface runoff, and longer time to incipient soil loss in comparison with CT and NT − M treatments.