Estimating reference evapo-transpiration in Mosul (Iraq)‎ using cascade neural networks

Abstract EN

Recently artificial neural network (ANN) has been applied for estimating reference evapo-transpiration (ET°).

In this study a mathematical model was built by Recently artificial neural network (ANN) has been applied for estimating reference evapo-transpiration (ET°).

In this study a mathematical model was built by application the cascade forward network technique (CCANN) to estimate the daily reference evapo-transpiration in the city of Mosul, north of Iraq.

The input parameters for the CCANN were the : temperature, solar radiation, wind speed at 2m height, and relative humidity.

A check for the accuracy of the performance of the network was made using values of reference evapo-transpiration obtained from pan evaporation method.

The results revealed linear correlation between the network output and the data of the measured pan evapo-transpiration with correlation coefficient of (0.9679).

This indicates the possibility of use of CCANN to determine the daily reference evapo-transpiration.

The results also show that the CCANN model performs better more accurate compared to other models.

application the cascade forward network technique (CCANN) to estimate the daily reference evapo-transpiration in the city of Mosul, north of Iraq.

The input parameters for the CCANN were the : temperature, solar radiation, wind speed at 2m height, and relative humidity.

A check for the accuracy of the performance of the network was made using values of reference evapo-transpiration obtained from pan evaporation method.

The results revealed linear correlation between the network output and the data of the measured pan evapo-transpiration with correlation coefficient of (0.9679).

This indicates the possibility of use of CCANN to determine the daily reference evapo- transpiration.

The results also show that the CCANN model performs better more accurate compared to other models.