Radar cells from the GATE experiment, from Spain and from the Montreal area were followed and measurements of their total rain flux and area were extracted. The values of the flux and area, for each one of the cells, for initial time increments T(,0) were used to evaluate various extrapolation schemes for forecasting purposes. The extrapolation procedure that gave the smallest error in forecasting the changes in flux and area, was found to be the linear one and the optimum increment T(,0) was about 30 min. However, even though these techniques have the advantage of establishing a trend in the behavior of the flux and area with time, a comparison of the forecast errors from the linear extrapolation scheme with those from the assumption of no change in cell area and rain flux shows insignificant improvements. A technique including both cell motion and internal changes in flux and area of the rain cells was developed to evaluate the accuracy of rain accumulation forecasts. Again, it was found that the errors were similar with those generated by the assumption of no change in rain flux and area from the moving cell.