Variability of major soil properties of a fallow-acidic-level upland with high and multiple spatial resolutions

Abstract

Variation of the soil attributes of a land in an area is dependent on topography, time, climate, parent material, land use land cover, land management, distance and scale. This variation affects the representation of soil of a land in an area. The study aimed to assess the variations in the representation of major soil properties of a unique fallow-acidic-undisturbed-level upland in different spatial resolutions of soil sampling. A fallow and level upland of 1500 m2 as separately gridded with the spacing of 5mx5m, 10mx10m and 15mx15m and geo-referenced surface (0-20 cm) soil samples were collected from the corner of each grid. The collected soil samples were analyzed for texture (Tx), organic carbon (OC), pH, total N (TN), available P (AP), exchangeable K (exch K), available S (AS), available Fe (AFe), available Zn (AZn) and available Mn (AMn) in soil. Statistical and geospatial analyses of the dataset were done with the relevant softwares. For the nutrients TN, AP, AZn and AFe, coefficients of variation (CV) showed a trend of increment across high-medium-low spatial resolutions, and their variability ranked as AZn (mean CV=104.03%, great variation)>AFe (mean CV=41.67%, moderate variation)>AP (mean CV=20.32%, moderate variation)>TN (mean CV=4.92%, low variation) based on average CV of three spatial resolutions of sampling. In case of other soil attributes, no particular trend of increment or decrement was observed across the resolutions and their variability was moderate except for pH which had low variability. Their variability ordered as exch K (mean CV=35.17%)>AS (mean CV=34.98%)>SOC (mean CV=31.71%)>Tx (mean CV=31.17%)>AMn (mean CV=30.10%)>Soil pH (mean CV=6.96%). Rationale correlations were observed between some soil attributes (pH vs AZ, AFe, OC; Tx vs TN, AP; Exch K vs AZn vs AFe; OC vs Exch K, AZn, AFe) with different degrees of associations (r), and increased trend in r value was found across the resolutions of high-medium-low except for pH and Tx. Different spatially gradient structures of the ordinary krigged interpolated maps were observed for different soil properties and for different spatial resolutions. Quantitatively, calculated (from semivariograms) nugget effects of 0-100% indicated that spatial dependency of studied soil properties could be very strong to very weak. The heterogeneity of soil in the upland as revealed by our results would assist scientists or farm managers to use or compare scale-dependent soil data wisely and precisely.