Kriging Interpolation Method for Estimation of Continuous Spatial Distribution of Precipitation in Cyprus

Abstract

Aims: Development of a precipitation prediction model for Cyprus. Study Design: Precipitation data collected at 78 stations were used: data from 66 stations for model development and data from 12 stations for additional tests. Four topographic factors – altitude, slope, longitude, and latitude – were taken into account for model development. Place and Duration of Study: All variables were obtained from the observation archives of the Water Development Department of the Ministry of Agriculture, Natural Resources and Environment of Cyprus, between 1961 and 1990. Methodology: Multiple regression analysis, combined with residuals correction, was carried out to develop a precipitation prediction model. Results: The multiple regression model can explain 61.3% of the spatial variability of precipitation over the whole year, 57.5% of variability in the wet season (October–April), and 99.6% of variability in the dry season (May–September). Interpolation-based residuals correction improved the accuracy of our model (Adj_ R2=65.1%, 62.6% and 99.7%, respectively). Conclusion: This approach, as presented in this paper, could potentially be applied to Cyprus’ climate research.