Simulating soil organic carbon changes across toposequences under dryland agriculture using CQESTR

•CQESTR, a soil C model, successfully simulated SOC changes across toposequences along a precipitation gradient.•CQESTR successfully simulated SOC at summit, shoulder, backslope, footslope positions, but under-predicted SOC at toeslope.•Toeslope had the highest SOC - a position with high annual deposition of C-rich soil eroded from the upper landscape position.•No significant changes in SOC were found between summit, shoulder, backslope, and footslope positions after 20 years.•CQESTR could be used to predict SOC changes across toposequence and at the landscape scale level with reasonable accuracy.

Soil organic carbon (SOC) and its management under dryland cropping systems are very critical for both crop productivity and environment health. The objective of this study was to evaluate the performance of CQESTR, a process-based C model, in simulating SOC changes across toposequences of selected fields and agriculture management practices along a precipitation gradient in a dryland region of Oregon, USA. Geo-referenced soil samples were collected from summit (SU), shoulder (SH), backslope (BS), footslope (FS), and toeslope (TS) positions during early 1980s and early 2000s. Simulation scenarios were developed based on field management practices, crop rotations, soil properties, and climatic data. CQESTR simulated results were compared with the measured SOC from each landscape position. Significant (P < 0.0001) correlations (r = 0.93) were found between the measured and the simulated SOC at SU, SH (r = 0.91), BS (r = 0.83), FS (r = 0.89), and TS (r = 0.89). The smallest correlation value at BS could be from soil deposition due to erosion. No significant changes in SOC were found between SU, SH, BS, and FS landscape positions; however, TS had the highest SOC (10.8 ± .8 g C kg−1). CQESTR successfully simulated SOC at most of the studied sites and landscape positions, except at TS for a location with high annual deposition of C-rich soil eroded from the upper landscape position. CQESTR could be used to predict SOC changes across toposequence and at the landscape scale level with reasonable accuracy. The results were supported by a linear relation with an r2 of 0.89 and a low mean square deviation (MSD = 0.24) between the measured and the simulated SOC.