Search published articles

Showing 1 results for Bayesian Regression

Ali Hashemi, Hojjatollah Yazdanpanah, Mehdi Momeni,
Volume 0, Issue 0 (3-1921)

Studying the effect of climatic variables on vegetation indices (Case study: Orange orchards in Hassan Abad, Darab County)
Climatic variables are the most significant factors affecting vegetation changes. Nowadays, the satellite imagery is widely used to investigate the effect of fluctuations in climatic variables on vegetation changes. This research aims to investigate the effect of climatic variables of precipitation, temperature, and humidity on changes in vegetation indices of orange orchards in Hassan Abad, Darab County using satellite data. Hence, observational data, including orange tree phenology data and meteorological data on the agricultural weather station have been collected for over 10 years (2006 to 2016). MODIS images from 2006 to 2016 were referenced based on the territorial data and 1:25000 maps of the Iran National Cartographic Center. These images were used to calculate the remote sensing vegetation indices including normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI). Results demonstrated that the variables of maximum humidity, minimum temperature, and precipitation have a significant positive effect on the NDVI variable. Additionally, the variables of maximum temperature and minimum humidity have a significant negative effect on the NDVI and EVI dependent variables. To determine the significance of each of the independent variables in predicting the dependent variables, the artificial neural network method was used. Findings showed that the climatic elements of precipitation, minimum temperature, maximum temperature, minimum humidity and maximum humidity with values (0.39, 0.3, 0.13, 0.1 and 0.06) had the greatest effect on EVI, respectively. Moreover, the effect of these variables on the NDVI index is equal to their coefficients (0.2, 0.28, 0.22, 0.11 and 0.17), respectively. Finally, ARMAX regression method was used to increase the explanatory power of the model. Results showed that this method could increase the explanatory power of the model and reduce the forecasting error.

Page 1 from 1     

© 2024 CC BY-NC 4.0 | Journal of Applied researches in Geographical Sciences

Designed & Developed by : Yektaweb