Areal Intensity-Duration-Frequency Curves for Short Term Rainfall Events in Lund

»Design Storms«, which are usually derived from point intensity-duration-frequency (i-d-f) relationships, neither reflect the areal variations of the rainfall pattern nor the dynamics of the moving storms. Design storms derived from point i-d-f curves can, no longer be accepted as a reasonable rainfall input for simulation of runoff from large urban catchments. One possible way of improving the rainfall input for runoff simulationes is to develop areal i-d-f relationships and use them for deriving design storms. The raingauge network operating in Lund for the past four years provided the set of good quality data necessary for developing such relationships. This paper describes the procedure of developing point and areal i-d-f relationships and shows the differences between them. Statistically developed factors reducing point values to areal values for areas up to 25 sq.km for different durations and return periods of up to three years are given. Presented relationships will give more realistic design storms in comparison with design storms derived from point i-d-f curves. Observed spatial variations of the maximum rainfall intensities in Lund can probably be explained by the city's influence on the rainfall pattern. Orographic effects, earlier observed in Lund, can not be seen from the data used during this study. Since the used data set includes only short term, high intensity rainstorms, it can be concluded that long and low-intensity rainfalls are mainly responsible for orographic effects. This paper is the first part of a major work aiming at developing a stochastic model simulating a time series of the areal rainfall.