The goal of this study was to test for the effects of land use and soil type on soil organic carbon (SOC) and total soil nitrogen (N) stocks in an arid region with a long history of cultivation and animal husbandry. Arid to semi-arid landscapes cover a large proportion of the Earth's surface and are sensitive to agricultural intensification and projected changes in climate, yet data on landscape controls on SOC and N remain scarce.
We used a state factor approach to study the effects of soil type and land use on surface SOC and total soil N (0–25 cm) across 35 sites in a 200 km2 area in the Konya Basin, an important agricultural region of Turkey. Using Landsat-based land cover maps, we identified three representative land uses – annual-crop agriculture, orchards and grazing lands – stratified across Aridisols formed on three different parent materials: alluvial clay-loam, lacustrine clay and terrace sandy loam soils.
SOC and N stocks depended strongly on soil type with strong interactions between soil type and land use. On alluvial soils, grazing land SOC and N stocks were 37% and 23% greater, respectively, than those of agricultural sites and 63% and 50% greater than at orchard sites. In contrast, agricultural sites on lacustrine soils contained 41% more SOC and 42% more soil N. There were no land use effects on terrace soils. The vertical distribution in SOC and N within the top 25 cm, representing the plow layer in the agricultural sites, differed by soil type and land use.
Soil type best explained landscape-scale variability of SOC and N stocks. Interactions between soil type and land use indicate that the long-term effects of land use on SOC and N were mediated by soil type. Differences in SOC and N stocks across soil types even within the same soil order highlight the importance of identifying pedogenic differences in soil properties, such as texture and mineralogy, which can influence the response of SOC and N to land use.
Our study contributes baseline data on SOC and N for a semi-arid region, which can be used to aid development of landscape-scale models of SOC and N dynamics and inform land management. Our data reveal that assessments of regional land use effects on SOC and N in arid to semi-arid environments should account for landscape to regional variability in soils developing on different geomorphic surfaces and parent materials.