Irrigation scheduling based on reference evapotranspiration (ETo) data depends on continual availability of ETo data. For high frequency irrigation, daily ETo data are inevitable. When the current ETo value cannot be estimated or it is of questionable quality, it must be estimated. Often historical data are used as estimates. This study presents a technique, based on correlation and regression analysis, for estimating missing data. Multiple correlation was performed on long term average daily ETo for each month of the year from ten California Irrigation Management Information System (CIMIS) stations. Of the 540 possible average ETo pairs representing 55 station pairs, only 22 pairs had at least one month of the year with a correlation coefficient less than 0.975. Some of these poorly correlated stations were located relatively close to each other. The mean and standard deviation of normalized difference in daily ETo for station pairs with multiple correlation coefficients greater than 0.975 were calculated. Regression equations, the end product of the estimation technique, were developed for station pairs with standard deviation of normalized difference equal to or less than 0.1. The ability of the regression equations to predict missing data was tested by comparing the 2001 daily ETo values (current ETo) for selected stations with ETo predicted by regression. Also, the accuracy of historical data as estimates for current ETo values was evaluated. The regression equations were a better predictor of current ETo values than historical data.
