Copulas-based probabilistic characterization of the combination of dry and wet conditions in the Guanzhong Plain, China

It is of great significance to investigate the joint probabilities of the combination of dry and wet conditions between adjacent seasons under the background of global warming. The Guanzhong Plain, a very important agricultural production area, was selected as the study area. The joint probabilities of the combination of dry and wet conditions between adjacent seasons based on the Standardized Precipitation Index (SPI) in 1961-2010 were computed by copulas functions. The Mann-Kendall trend test method and wavelet analysis were used to investigate the trend and period of dry and wet conditions, respectively, and the Bayesian copula selection method was used to choose the best copula. Furthermore, the cross wavelet analysis was utilized to reveal the correlation between dry and wet conditions and sunspot activities. Results indicate the following: (1) the Guanzhong Plain is characterized by an relatively obvious trend towards dry in non-flood season and a relatively striking tendency towards wet in flood season; (2) the dry and wet conditions in the Guanzhong Plain have a statistically significant in-phase relationship with sunspot activities in 1983-1997 with a period of 8-12 years, whilst they have a non-statistically significant anti-phase relationship with sunspot activities in 1961-1973 with a period of 9-12 years, hence, sunspot activities exert a strong influence on dry and wet conditions in the Guanzhong Plain; (3) the plain is characterized by a high-frequency continuous dry condition in the entire year and a high risk of continuous wet condition in the season of Summer-Autumn;(4) the joint return period of the combination of dry and wet conditions between adjacent seasons gradually increases from east to west in the west and middle of the Guanzhong Plain, which is possibly closely linked to different geomorphic terrains, intensity of human activities and ENSO events. (C) 2014 Elsevier B.V. All rights reserved.