نوع مقاله : مقاله پژوهشی
نویسندگان
بخش تحقیقات خاک و آب، مرکز تحقیقات و آموزش کشاورزی و منابع طبیعی استان اصفهان، سازمان تحقیقات، آموزش و ترویج کشاورزی، اصفهان، ایران
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Knowing about the spatial dependence of different soil characteristics in farms is important to achieve more production and better management. The aim of this study was to evaluate the spatial variability and frequency distribution of some physical and chemical properties, including pH, EC, organic carbon, phosphorus and potassium that can be used by plants, texture and cation exchangble capacity, within the various landforms of Isfahan province. The study was conducted on 118 soil samples. The spatial correlation of each variable with a specific semi-variable and the best fitting model for each variable were selected using GS+ version 9 software. Interpolation was done using normal Kriging, Cokriging, and Inverse Distance Weighting with powers of 1 to 3. The accuracy of the distribution maps of these variables were evaluated by the mean deviation of error (MBE) and the root mean square error (RMSE). The results of the geostatistical analysis showed that potassium, sand percentage and pH had strong spatial dependent and the other characteristics had moderate spatial dependent. The exponential model was the most accurate to predict phosphorus, EC, CEC, clay and silt variables while potassium, pH, sand and organic carbon percentage were best fitted with an sphericalmodel. Also, EC had the smallest effective range (14.86 km) and pH had the largest effective range (around 71 km). For potassium, pH and EC variables, the Inverse Distance Weighting with the power of 1 (IDW-1) and for other variables the normal kriging method were recognized as the best interpolation methods.
کلیدواژهها [English]
EXTENDED ABSTRACT
Knowing about the spatial dependence of different soil characteristics in farms is important to achieve more production and better management. Soil pH, Electrical Conductivity (EC), organic matter, available phosphorus, potassium, Cation Exchange Capacity (CEC), and soil structure are some of the most important indicators of soil fertility. These soil parameters are highly variable in space and time, especially in agricultural areas, with implications for crop production.The aim of this study was to evaluate the spatial variability and frequency distribution of some physical and chemical properties, including pH, EC, organic carbon, soil available phosphorus and potassium, texture and cation exchange capacity.
The study was conducted on 118 soil samples, in agricultural lands of different regions of Isfahan province in 2016. The spatial correlation of each variable with a specific semi-variable and the best fitting model for each variable were selected using GS+ version 9 software. Interpolation was done using normal Kriging, Cokriging, and Inverse Distance Weighting with powers of 1 to 3 methods. The accuracy of the distribution maps of these variables were evaluated by the Mean bias error (MBE) and the root mean square error (RMSE).
The results of the geostatistical analysis showed that potassium, sand percentage and pH had strong spatial dependent and the other characteristics had moderate spatial dependent. The strong spatial dependence due to the effect of intrinsic factors such as parent material, relief and soil types. The moderate spatial dependence could result from variation in environmental factors such as flood water, irrigation, fertilizeir addition, high water table level or agriculture practices and different agricultural managements. The exponential model was the most accurate to predict phosphorus, EC, CEC, clay and silt variables while potassium, pH, sand and organic carbon percentage were best fitted with an spherical model. Also, EC had the smallest effective range (14.86 km) and pH had the largest effective range (around 71 km). For potassium, pH and EC variables, the Inverse Distance Weighting with the power of 1 (IDW-1), RMSE equal to 0.171, 0.152, and 0.171 respectively, and for organic carbon, Phosohorus, texture and CEC, RMSE equal to 0.11, 0.199, 0.155, and 0.156, respectively, the normal kriging method were recognized as the best interpolation methods. Spatial distribution maps according to the most accurate techniques showed that the greatest soil salinity and the sand percentage were measured in north and especially the northeast parts (in the cities of Ardestan and Aran and Bidgol in Isfahan province), also the amount of nutrient elements and soil organic carbon reduced in these areas. The soil fertility was better in the southern and western parts of the province. In addition to management factors such as fertilization and irrigation, this is due to the climate and more rainfall in these areas, as well as the higher quality of irrigation water in these areas. In general, according to the obtained results, it was found that the geostatistical modeling can show continuous changes of soil parameters with acceptable accuracy.
)