Geostatistical analysis for predicting soil biological maps under different scenarios of land use

• Published in: European journal of soil biology 2013-03, Vol.55, p.20-27
• Main Authors: Shahbazi, F., Aliasgharzad, N., Ebrahimzad, S.A., Najafi, N.
• Format: Article
• Language: English
• Subjects: Agronomy. Soil science and plant productions; apples; Biochemistry and biology; Biological and medical sciences; Biophysical modeling; carbon; Chemical, physicochemical, biochemical and biological properties; crop production; enzyme activity; Fundamental and applied biological sciences. Psychology; geographic information systems; Geostatistics; kriging; land use; microbial biomass; pastures; Physics, chemistry, biochemistry and biology of agricultural and forest soils; prediction; soil biological properties; soil heterogeneity; Soil microbiological indices; Soil quality; soil sampling; Soil science; urease
• ISSN: 1164-5563
• DOI: 10.1016/j.ejsobi.2012.10.009

The ArcGIS Geostatistical Analyst aims to effectively bridge the gap between geostatistics and geographical information system analysis by enabling to model spatial phenomena and accurately predicting values within the study area. This approach was conducted to forecast the distribution patterns of some soil biological indices in Mirabad area, North West of Iran. Three different land uses (apple orchard, crop production, and rich pasture) were selected to conduct the experiments in a randomized completely blocks design with five blocks. Soil samples (0–30 cm) were collected on mid July 2010. Soil biological indices i.e. (i) substrate induced respiration, (ii) microbial biomass carbon, (iii) the activity of urease; (iv) alkaline phosphomonoesterase, and also (v) dehydrogenase were determined. Kriging and inverse distance weighting methods were applied to assess the spatial variability of five stated indices. Ordinary kriging was applied because it is the most general and widely used method. Digital soil biological indices maps will be the last output of integrating geostatistics and geographical information system. The study, while addressing spatial variability of soil biological properties, also discusses the accuracy of modeling as well as spherical model is now distinguished as the best fitted model. Assessing spatial variability of alkaline phosphomonoesterase activity has the lowest accuracy than urease and dehydrogenase activities. The geostatistical results showed that management practices might not have relevant effect on microbial biomass carbon and enzyme activities. But, the statistical analysis revealed significant differences between pasture and two other land uses. ► Geostatistics may be used to predict soil biological maps in different land uses. ► Integrating GIS with geostatistics could help us to track in soil enzyme activities. ► Spherical model as the best fitted model could predict soil biological properties. ► Number of digital soil map units is the key point to study the spatial variability.