Abstract

STOCHASTIC SIMULATION OF DAILY PRECIPITATION OCCURRENCE AND AMOUNT IN MAKURDI, NIGERIA.

aAgada, P. Onuche b Agada, O. I. and c Ogar, Sylvester

---

Abstract Future precipitation occurrence and amount if accurately forecasted can be used for crop planting planning and for making crop and water management decisions, but this is not without the challenge of weather uncertainty. Weather simulation models are well known to help surmount this challenge. This work is therefore in the right direction as it has successfully constructed and validated a stochastic simulation model, for simulating precipitation occurrence and amount in Makurdi, Nigeria. This was achieved by determining the occurrence of precipitation (wet) days using Markov chain transition probabilities of order one and generating precipitation amounts as random variates from the fitted probability distribution of precipitation amount. The probability distributions were fitted to thirty four (34) years data of daily precipitation amount in Makurdi metropolis for each wet month of the year (March-October). The methodology further include the determination of the monthly mean and standard deviation of precipitation amount, the determination of the monthly number and standard deviation of precipitation days and the validation of the model results using the real data set. The simulation model was constructed using a Pascal Simulation shell (PSIM) developed by Davis and O?Keefe (1989). The result shows that the exponential distribution fit the precipitation amount for the month of March, while the gamma distribution is the best fit for the months of April, May, June, July, August, September and October. Ten (10) years forecast of monthly number of precipitation days and the mean precipitation amounts were made. It was recommended that the model be used for simulating precipitation occurrences and amounts in Makurdi metropolis. Keywords: Stochastic, Simulation, Model, Precipitation

PDF