A proper soil raster unit resolution for grid sampling design is important to estimate the soil organic carbon (SOC) pool at certain map scales, which is related to the soil sampling density and the accuracy of the estimation. A series of raster soil unit data sets at varying resolutions were derived from different vector soil unit data sets at six map scales of 1:50,000, 1:200,000, 1:500,000, 1:1,000,000, 1:4,000,000, and 1:14,000,000 in the Tai-Lake region of China. Four indices-soil type number (STN) and area (AREA), average SOC density (ASOCD), and total SOC stocks (SOCS) of surface paddy soils- were attributed from all these vector and raster units data sets. Subjected to the four index values (IV) from parent vector unit data set, the relative variability (VIV, %) from raster unit data set was used to assess its accuracy and redundancy, which reflects uncertainty and workload of SOC estimation, respectively. Optimal raster unit resolutions were generated and suggested for each map scale's SOC estimation, in which the soil raster unit data set can hold the same accuracy as its parent vector unit data set without any redundancy when VIV < 1% of all the four indices was assumed as criteria to the assessment. A relationship between map scale (1:x) of soil vector unit and its optimal grid resolution (y, km) was found to be: y = -8.03 × 10-6x2 + 0.0256x- 0.087 (R2 = 0.998, p < 0.05). The results may serve for soil unit conversion from vector to raster and soil grid sampling design at a certain map scale in the investigation of regional SOC pool.
All Science Journal Classification (ASJC) codes
- Soil Science