PakCat catalogue record

Prediction Models for Reference Crop Evapotranspiration in Punjab of Pakistan 'Source Title' Bulletin of Irrigation, Drainage and Flood Control Research Council Volume Vol.15, No.2

By Shahid Ahmad

Bibliographic details

Summary

Reference crop evapotranspiration (Et) for six locations of Punjab was computed for weekly periods using historical climatic data and Penman equation. The average of historical Et values were used to develop prediction models. Two step piecewise smooth exponential function was used to fit the average weekly Et curve for Julian calender year. The regression analysis of exponential prediction model inidcated that R2 values were in the rnage of 0.967-0.987. This shows that prediction models can be used with very high confidence for prediction of average weekly Et values. Sensitivity analysis indicates that these models over predict 0.42 - 3.777% on an annual basis, which is within acceptable range takeing into consideration benefits of the prediction models for planning, design and operation of irrigation projects in Punjab. The normal rainfall and evapotranspiration deficit values for six locations also indicated the need to use separate prediction models for different locations because of significant variation in Etr values.