Nested Augmentation of Rainfall Monitoring Network: Proposing a Hybrid Implementation of Block Kriging and Entropy Theory

One of the most influential environmental variables is rainfall which has significant effect on water resources management, agricultural development, hydrology, and climate change studies. Due to high spatiotemporal variability of rainfall, its monitoring network design can be considered as a useful tool to improve the efficiency of recorded rain gauge stations within the study area. In this study, a new methodology of augmentation of rain gauge network is developed using coupled Block Kriging (BK) and entropy theory methods. In the proposed method, a nested approach of a two-stage positioning of rain gauge stations has been demonstrated. In the first stage, large-scale or fast positioning was done in which the optimal number of candidate blocks was identified. Then, local scale or fine-tuned positioning was done in the second stage. In this stage, to develop the network, accurate locations of rain gauge stations in each block are determined. Besides the main point of this paper, the effect of two kriging estimators, BK and Ordinary Kriging (OK), on the developed network has been investigated and compared. The study area is the Namak Lake watershed with various climates and altitudes. To assess the performance of the optimal rainfall network, three diagnostics were utilized; spatial distribution of annual precipitation, Estimation Error Variance (EEV) maps and histograms. Based on the results, 30 (more than 30% percent of the current stations) rain gauge stations have been proposed scattered over the watershed. Evaluation of the results has shown that the augmented rain gauge network proposed by the BK method outperformed dramatically that of the OK method. EEV maps and also statistical analysis of EEV values confirms the EEV value reduction of almost 25% in augmented network, as well.