Statistical analysis of daily rainfall data

Thumbnail Image
Haankuku, Nchimunya Urban
Journal Title
Journal ISSN
Volume Title
Zambia, where annual rainfall is moderate to meagre over large parts of the country, and year to year variations are large, unpredictable weather can prove to be a serious obstacle to stable and systematic agricultural production. For effective agricultural planning on a rational and scientific basis, appropriate analysis of rainfall proce-dures need to be set up. Fairly accurate predictions within a given season or future seasons are of great importance to agriculturalists and designers (such as builders of reser-voirs, dams flood control works, airports). Agricultura-lists and designers need to know with confidence when the rainy season is expected to begin/end and how much rain is likely to fall in a given season. They may also want to have reasonable expectations of wet and dry spells. This project tries to provide answers to the following questions: (a) When is the start of the rain season expected? (b) When is the rainy season expected to end? (c) What is the expected total rainfall for part or the whole of the rainy season? (d) How much rain may be expected during a season given data on the early part of a rainy season Daily rainfall data recorded at Chingombe weather station(Kabwe) over a period of 25 years was used as the source of data to be analysed. Rainfall data from 1st September, 1960 to 30th June, 1985 were available for the project. It is evident from the data that there is little or no rainfall in the period from 30th June to 31st August in any one year. Consequently, rain at this station is restricted to the period from early September of one year to the middle of May of the next year at latest. This period, therefore, more than adequately covers the rainy season at this station. This project has three distinct parts: the first part of the analysis is found in Chapter II. In this Chapter the ANALYSIS OF RAW DATA is outlined, especially the distribu-tions of rainfall over each rainy season for the period of 25 years. The second part of the analysis is found in Chapter III. In this chapter the predictions of future seasons were considered assuming that each variable had a truncated exponential distribution. The third part of the analysis is found in Chapter IV where with-in-season predictions were considered using regression analysis. In this Chapter, the rainfall data for the early part of the 1985/86 season was available, and predictions as to the other aspects of the 1985/86 rainy season were attempted using tested statistical procedures. The analysis of variance tables for all the variables, the plot of standardized residuals versus predicted standardized depe¬ndent variables for measures of model adequacy were obtained using the Statistical PACKAGE for the SOCIAL SCIENCES (SPSS) Computer package.
Statistical weather forcasting(Rainfall)--Zambia , Statistics