Conversion of rain data from surface stations to forecast models data

الملخص EN

According to Statistical analysis of daily rainfall values (2007_2016) for the study areas(Baghdad, Karbala, Al-Hay, Mosul, Kirkuk, Khanaqin, Basra, Nasiriya, Diwaniya, and Rutba) could been concluded that the 40 % of rainfall very variation and 30% were moderately variable the remaining 30 % had little variation, The most variation in (Baghdad, Diwaniyah, Nasiriyah and Karbala) with CV (50) within the study period.

The spatial and temporal variation of rain is one of the reasons that the measured rainfall data from surface stations cannot be directly used in forecasting weather models, they do not give good values for prediction.

Therefore, Curve fitting is a good way to convert this data to data calculated from a statistical model without changing the actual value and can use as inputs data in the models of prediction.

In this study, more than one function testing and found the best is) waveform-sin 4parameter) the set data of this function used was used in the prediction model SDSM and gave very good results.