Abstract
Each system of prediction has its own sphere of application and associ-ated set of advantages and disadvantages. Since the atmosphere is basically a turbuleat fluid, its future state in not predictable beyond about three weeks although certain mean and persistent states may be predicted with some accu-racy. Thus the statistical treatment of past climatic and other dependentdata is assential to estimate the probabilities of future weather states. Every agricul-tural undertaking has some economic risk depending on the chance of poor or unfavourable weather. Risk assessment is one of the soundest bases for determi-ning the practicability of certain agricultural operatic 11F. Temperature has a direct effect on the growth rates of plants and high temperatures above the optimum for growth, result in reduced growth rates and more serious are the damaging effects at critical stages of development. In a number of crops, high temperatures immediately after flowering may interfere with pollination and this will decrease yields. However, low humidities genera-lly associated with days of high temperature are also known to reduce pollen viability. High temperatures also have direct effects on livestock. Cattle are known to suffer loss of appetite and decreased feed intake and bulls subjected to temperature stress exhibit seminal degradation (Rosenthal and Hammer, 1979). In most of these cases, the farmer must take preventive measures as little can be done when the heatwave occurs. With crops it is necessary to adjust the time of planting and the variety used so that the risk of encountering a heat-wave at the critical stage of development is kept at, an acceptable level. The maximum temperature data for the period 1944 to 1983 were collected at the meteorological observatory of the University of Agriculture, Faisalabad. The frequency distributions for decade mean maximum temperatures were prepared and the probabilities calculated therefrom (W M.O., 1981).
Table 1 gives the decade mean maximum temperatures for a particular
*Department of Agricultural Meteorology, University of Agriculture, Faisalabad.
123
