Investigating the Spatial Distribution of Soil Organic Carbon in Fandooqlo Region of Ardabil Province, Iran

Estimating the amount of soil organic carbon on a regional scale determines the potential of a landscape soil as a carbon storage site. This study was carried out in the Fandoqloo Region, Ardabil Province, Iran, and aimed to introduce: (i) the most suitable soil and environment characteristics as ancillary data for estimating soil organic carbon density (SOCD) and (ii) the most appropriate method for mapping SOCD among ordinary kriging (OK), co-kriging (CoK) and regression-kriging (RK) models. To fulfil the objectives, geographic information systems' database of the study area was developed by introducing soil, topographic and satellite data in the first step. Next, using Latin Hypercube (LHC) techniques and soil, land use, and geology maps and 140 sites were determined in the study area for collecting surficial (0-15 cm) compound soil samples. Results indicated that the land use type significantly affected SOCD (P ≤ 0.01) and SOCD of rangelands was higher than of croplands. Soil total nitrogen and mean weight diameter (MWD) were significantly (P ≤ 0.001) correlated with SOCD and could be applied as ancillary data for estimation of SOCD. Statistical indices revealed that application of co-kriging along with total soil nitrogen as ancillary data improved SOCD mapping compared with ordinary kriging. In general, this research indicated that regression-kriging was the most efficient method for mapping SOCD and total soil nitrogen, MWD, normalized difference vegetation index (NDVI) and plane curvature were significant soilscape characteristics that affect SOCD distribution.