Abstract
The present study was done to study the effect of climate change on weather parameters like highest possible temperature, lowest possible temperature, average temperature and precipitation. Multiple linear Regression (MLR), Artificial Neural Network (ANN) and Statistical Downscaling Model (SDSM) models were tested in the Dal lake catchment area of Jammu and Kashmir State. Twenty seven year weather data (1985-2012) obtained from SKUAST-Kashmir weather station was used for the study. The modeling results showed a first-rate agreement between the observed data and predicted values for temperature series with high coefficient of determination R2 values varying from (0.87-0.97) for different models. In case of precipitation R2 values varied from (0.112-219) for different models. The low values of coefficient of determination in precipitation time series are due to lot of uncertainty in occurrence of precipitation which could not be defined by the selected models. The SDSM showed the best results of the three models tested for prediction of weather parameters. Thus SDSM was used for climate scenario generation. By comparing daily precipitation and temperature series for 1985-2012 with 2015-2030, an overall increasing pattern of 0.46%, 1.96%, 0.95% and 2.66% was observed for monthly, highest possible temperature, lowest possible temperature, average temperature and precipitation.
Highlights
For the past century, increase in temperature and CO2 concentration due various factors including change in the pattern of land use[4] and greenhouse gas (GHG) emissions from industrial and agricultural sectors[7], have caused changes in the earth’s climate
A comparison was first made between the observed data and predictor data obtained from CCISC to check the best fit model amongst Multiple Linear Regression (MLR), Artificial Neural Network (ANN) & Statistical Downscaling Model (SDSM) using statistical indices like Coefficient of Determination (R2), Mean Square Error (MSE), Root Mean Square Error (RMSE) and Absolute Average Deviation (AAD)
The observed values of daily maximum temperature, minimum temperature, average temperature and precipitation were fitted in the MLR, ANN & SDSM models respectively, which showed a good fit with high values of R2 for each model
Summary
Increase in temperature and CO2 concentration due various factors including change in the pattern of land use[4] and greenhouse gas (GHG) emissions from industrial and agricultural sectors[7], have caused changes in the earth’s climate. This increased GHG’s concentration was likely to influence earth’s temperature as well as precipitation along with the pattern of storms as well as changes to sea levels[3,1,6]. Comparison between Multiple Linear Regression (MLR) model of MS-Excel, Artificial Neural Network (ANN) model of MATLAB software and Statistical Downscaling Model (SDSM) was done to determine the best fit model for downscaling the climate chànge in Dàl Làke càtchment of Srinagar City, Jammu & Kashmir, India
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
