RAINFALL AND RUNOFF MODELING AS A JOINT TIME SERIES, CASE STUDY: SOUTHWEST SAUDI ARABIA

The prediction of the future water resources in specific area is very important for water resources development in specific region. There are many methods in which models are used to predict the amount of expected runoff resulted from rainfall. One of these methods is the stochastic methods. One of these stochastic models is called “The Vector Autoregressive Model (VAR)”. In the current study, this model is used for modelling rainfall-runoff relationships as a joint time series. The development of water resources systems in arid and semiarid zones suffer from data availability, especially for storm runoff. Measurements of runoff in arid zones are often not available; therefore, there is a need to estimate runoff that is produced from rainfall events. In the current study, a regional stochastic model is developed to assess the correlation between rainfall and runoff in arid and semiarid zones based on recorded data at five gauged watersheds in the southwestern part of the Kingdom of Saudi Arabia during the period 1984 -1987. We have reached from the Chi^2 test we find the log-normal distribution is the best for both monthly rainfall and runoff data. The p-p plot shows very good agreement between observed data and generated results. Most of the stations show strong correlation between rainfall and runoff data (up to 0.894). The lowest correlation is at station J-416 (0.101). One of the results of the success of the model was the values of the mean of the observed data and the generated data of the rainfall and runoff depth the correlation strength reached 0.9 and the other statistical characteristic is the standard deviation the correlation between the observed values and the generated results also reached 0.9 and it shows a very strong correlation.