## The isohyetal method

This is considered one of the most accurate methods, but it is subjective and dependent on skilled, experienced analysts having a good knowledge of the rainfall characteristics of the region containing the catchment area. The method is demonstrated in Fig. 9.8. At the nine rain-gauge stations, measurements of a rainfall event ranging from 26 to 57 mm are given. Four isohyets (lines of equal rainfall), are drawn at 10-mm intervals across the catchment interpolated between the gauge measurements. Areas between the isohyets and the watershed (catchment boundary) are measured. The areal rainfall is calculated from the product of the inter-isohyetal areas (ai) and the corresponding mean rainfall between the isohyets (R) divided by the total catchment area (A).

Fig. 9.8 Isohyetal method of estimating areal rainfall.

In the illustration (Fig. 9.8), there are five subareas and the areal rainfall is given by: 1 5

In drawing the isohyets for monthly or annual rainfall over a catchment, topographical effects on the rainfall distribution can be incorporated. The isohyets are drawn between the gauges over a contour base map taking into account exposure and orientation of both gauges and the catchment surface. It is in this subjective drawing of the isohyets that experience and knowledge of the area are essential for good results. The isohyetal method is generally used for analysing storm rainfalls, since these are usually localized over small areas with a large range of rainfall amounts being recorded over short distances.

The UK Met Office has used the isohyetal method for many years and has also developed a modified version in which isopercental lines are drawn instead of isohyets. For a required areal rainfall for a given event, the measurements at each station are plotted as a percentage of the station standard long-term average annual rainfall. Then the lines of equal percentages (isopercentals) divide the catchment into areas that are measured, applied to the mean percentages between the isopercentals, and an overall areal percentage is obtained. This is applied to the standard long-term average annual areal rainfall for the catchment, previously derived, to give the required areal rainfall for the event. This improved technique is more reliable and objective, and can be carried out by computer. However, although it is readily applicable in countries with many long homogeneous rainfall records from which reliable long-term average annual areal rainfall values for all catchments can be evaluated, it is not so useful in developing countries where records are limited.

0 0