Prediction of Soil Enzyme Activity Along the Transect Using Some Soil Characteristics (Case Study: Miami County in Semnan Province)

Soil quality indicates soil ability to provide biological, ecological and production services. Determination of soil quality indices requires a large number of soil characteristics measurements. Soil enzyme activities have been recognized as suitable indicators in assessing the effect of management and environmental factors on soil quality. This study has been implemented to develop a regression model between soil enzyme activities and soil characteristics along the north-south transection in Miamicounting. For this purpose, 120 soil sampleswereselected from the top layer (30 cm) at three different sites using the standardized classification method. Soil properties and activity of enzymes were determined according to standard methods and also, applying linear regression method of ordinary least square, the relationship between soil properties and three soil enzymes included in b-glucosidase, urease and alkaline phosphatase was modeled. The results showed, overall all soil enzyme activity decreased from north to south and b-glucosidaseenzyme had the highest coefficient of variation (%68). The models provided with accuracy (R2adj) about %50.4 to %70.4 are able to predict the soil enzymes activity. The most important soil properties effective on enzyme activity in the northern region included respectively, total organic carbon, soil texture, cation exchange capacity, Phosphorus and potassium available but in the southern region, total organic carbon, salinity, PH. The results of validation (RMSE, R2adj)ascertained that the predictors were sufficiently accurate. The results of this study used in the improvement of regional planning for sustainable management of soil.