Search

Recently, a single-site stochastic precipitation model called ''the mixture of generalized chain-dependent processes conditioned on a climate variable'' was developed. The model can effectively eliminate overdispersion--that is, underestimation in variance of seasonal precipitation total. In this paper, the single-site model is further developed into a multisite stochastic precipitation model by driving a collection of individual single-site models, but with spatial dependence following a method proposed by D. S. Wilks. Specifically, a computationally effective algorithm for estimating the spatial dependence of precipitation occurrence is developed to replace the construction of the empirical curves in the Wilks method. An effective and straightforward approach for correcting the bias of the spatial correlation of precipitation intensity is also proposed. This model is tested on a small network of sites from a significant hydroelectric power generation region of South Island, New Zealand. [ABSTRACT FROM AUTHOR]

Abstract: Investigation of hydrological impacts of climate change at the regional scale requires the use of a downscaling technique. Significant progress has already been made in the development of new statistical downscaling techniques. Statistical downscaling techniques involve the development of relationships between the large scale climatic parameters and local variables. When the local parameter is precipitation, these relationships are often very complex and may not be handled efficiently using linear regression. For this reason, a number of non-linear regression techniques and the use of Artificial Neural Networks (ANNs) was introduced. But due to the complexity and issues related to finding a global solution using ANN-based techniques, the Genetic Programming (GP) based techniques have surfaced as a potential better alternative. Compared to ANNs, GP based techniques can provide simpler and more efficient solutions but they have been rarely used for precipitation downscaling. This paper presents the results of statistical downscaling of precipitation data from the Clutha Watershed in New Zealand using a non-linear regression model developed by the authors using Gene Expression Programming (GEP), a variant of GP. The results show that GEP-based downscaling models can offer very simple and efficient solutions in the case of precipitation downscaling. [Copyright &y& Elsevier]