Determination of relationships between dominant plant species with environmental factors and satellite data using logistic regression (case study: Rineh Rangeland, Mazandaran province)

This research was aimed to investigate the relationships between dominant rangeland species and environmental factors in the Rineh rangelands located on the southern slope of Damavand Mountain. Stratified random sampling method was applied and the study area was classified to 37 sample units based upon elevation, slop and aspect. In the study area, 750 plots 1m2 and 75 plots 25m2 were established. One hundred and fifty soil samples were dug at the depth of 0-30 cm and 16 soil properties including pH, CaCo3, bulk density, particle density, total phosphorus, total nitrogen, absorbed potassium, organic matter, saturation moisture, soil texture, field capacity, permanent wilting point, available water capacity and water holding capacity were measured in laboratory. After collecting the climate data, 16 climate factors including average relative humidity, average rainfall, average temperature, average minimum temperature, average maximum temperature in spring, summer and annual as well as annual frost days were selected to reconstruct the missing data. IRS images including red and near-infrared bands and PAN band as well as combined bands derived from RVI, NDVI, and PCA were selected. Regression equation was developed for each species using logistic regression. Results indicated that most of the factors including slop, aspect, average minimum temperature in spring, number of frost days, average daily temperature in summer, and most of the soil factors were entered in regression models. Also, satellite data were used as effective tools in showing the presence of plant species.