Document Type : Research Paper
Authors
1
M.Sc. Student of Soil Sciences, Faculty of Agriculture, Tarbiat Modares University, Tehran, Iran
2
Professor, of Soil Sciences, Faculty of Agriculture, Tarbiat Modares University, Tehran, Iran
3
Science Staff of Soil and Water Research Institute (SWRI), karaj, Iran
4
Science Staff of Soil and Water Research Institute (SWRI), karaj, Iran;
Abstract
Determining soil parameters variability is a great necessity for precision agriculture. To achieve higher yield with better management, knowledge of physico-chemical properties of soil and nutrients critical levels in the fields is essential. In this regard, geostatistics is one of the methods recently used for investigating the spatial variability of soil properties. In this study, 188 soil samples were taken from the eastern farms of Mazandaran Province and were analyzed for their pH, Organic Carbon (OC), CaCO3, P, K, Fe, Mn, Zn, Cu, and soil texture, in 2009 . The spatial correlation of variable Zn and the best fitted model (Gaussian model) were determined by variogram. The effective range of this value was 40 km. In order to determine the distribution map, interpolation methods, including Kriging, Inverse Distance Weighted (IDW) and Splines were used. Moreover, the precision of these interpolation methods was calculated through mean base error (MBE), mean absolute error (MAE) and Root Mean Square Error (RMSE). A pot experiment with soil samples collected from 29 fields out of the 188 studied fields and containing different Zn concentration was conducted to determine the critical level of soil in response prediction of soybean (Glycine max L.) to Zn fertilization. The experiment was done in a factorial arrangement in randomized complete block design with three replications (174 pots), in the Agricultural Research Center of Mazandaran Province. Soil samples from 29 fields were considered as factor 1, and two Zn fertilizer treatments (0 and 10 mg Zn kg−1 as zinc sulfate) as factor 2. After harvesting and relative yield determination, the critical soil Zn concentration was found to be 1.40 mg Zn kg−1 . This finding was based on pot study for obtaining 95% relative yield, using Cate-Nelson graphical method. Furthermore, using the probability map with Indicator Kriging method revealed that about 80 percent of soils in the studied region had Zn deficiency.
Keywords
)
