Develop Evaporation Model Using Multiple Linear Regression in the Western Desert of Iraq –Horan Valley

Abstract

Evaporation is influenced by several meteorological parameters, evaporation data are usually difficult to obtain compared to rainfall data, especially in arid regions. Developing a monthly evaporation prediction model in arid regions in terms of available meteorological data is a significant step. The data used in this study for modeling are monthly measurements to cover substantial continuity over a period of 18 years between January 2000 and December 2017. Stepwise and backward multiple linear regression techniques were used with a new procedure of variable selection to select the best model. Temperature, wind speed, relative humidity and sunshine hours were used as a independent variables in the multiple linear regression (MLR) technique to establish the best prediction of the evaporation model. To examine the MLR evaporation developed model in the current study, MLR results were compared with the most common evaporation models commonly used in arid regions such as Kharufa and Khosla methods. The results of performance indicators shows that the R2 values are approximately 0.937, 0.90 and 0.85 for MLR evaporation developed model, Kharufa and Khosla methods, respectively. Moreover, the values of the error measures, namely RMSE and NAE for MLR evaporation developed model were 36.3 and 0.123, Kharufa model 71.22 and 0.241 and Khosla model was and 173.7 and 0.581 respectively. Based on the foregoing, the results of the MLR developed evaporation model in the current study outperforms in all performance indicators and proves to be better than the Kharufa and Khosla models.

Prediction of Monthly Evaporation Model Using Artificial Intelligent Techniques in the Western Desert of Iraq-Al-Ghadaf Valley

“…NAE and RMSE should have zero for a successful pattern. In contrast, NSE and R 2 should be closer to one [35].…”

“…The choice of the appropriate predictors can be regarded as one of the significant steps in designing a robust predictive model [2]. In the present study, the selected groups of evaporation and the best influencing and co-predicting variables were adopted in this study based on [35] in order to apply artificial intelligence techniques to these groups. Table 1 clarifies that Group one that includes E as the dependent variable and T, WS, RH and SS as independent variables and Group two consists of E as the dependent variable and T, WS and SS as independent ones.…”

Daily Evapotranspiration Prediction at Arid and Semiarid Regions by Using Multiple Linear Regression Technique at Ramadi City in Iraq Region

“…Multiple linear regression is a popular approach for linear modelling that is used for investigating the relationship between a dependent (response) variable and many independent (predictor) variables [18][19][20][21]. The following equation can be used to calculate MLR [21].…”

Evaluation of Fluoride Concentration in Drinking Water of Some Cities within Anbar Province, Iraq

“…Human health can be adversely affected if the origin of supplied water is polluted. Good drinking water quality must contain appropriate chemical and minerals content, Fluoride is one of the essential minerals [9][10][11].…”