توصيفگر ها :
تابع مفصل , سناريو انتشار , شدت , مدت , يخبندان , GCM
چكيده انگليسي :
Occurrence of extreme climatic phenomena such as frost will cause significant risks and costs to many sectors, especially agriculture, horticulture, and forests. Frost will cause the worst damage when it occurs at the critical stages of crops, especially in spring. The frost phenomena are one of the important climatic and environmental hazards that cause a lot of damage to the agricultural sector of Iran every year. In this sstudy, copula functionswere used to calculate the joint bivariate return period of frost events in two statistical periods, past (1984-2014) and future (2023-2053), in Isfahan province. For the future period, a combination of 10 general circulation models (GCMs) from the sixth assessment report of climate change under three SSP scenarios (SSP1-2.6, SSP2-4.5, and SSP5-8.5) was employed using a weighting method. This study examines the risk and effects of frost events in orchards by modeling the joint distribution of event duration and severity factors and analyzing the dependency structure of influential parameters using the capabilities of joint functions. Additionally, the spatial and temporal variations of frost parameters (the first early fall frost, the last late spring frost, the growing season length, the frost period length, and the number of frost days) based on models from the sixth climate change report under three emission scenarios are analyzed for the future statistical period compared to the past period in Isfahan province.The results showed that the average date of the first early autumn frost in the stations of Ardestan, Khur and Biabanak, and Natanz in the future period under three emission scenarios occurs one to six days later compared to the past period. Additionally, the average date of the last late spring frost in Ardestan station in the future period under the three emission scenarios occurs five, four, and three days later, respectively, compared to the past period. Considering the dates of the first early autumn frost and the last late spring frost, the greatest change in the length of the growing season occurs at the Chadegan station, which under the SSP5-8.5 scenario, increases by 21 days compared to the past period. The most significant reduction in the length of the frost period occurs at the Khur and Biabanak station under the SSP5-8.5 scenario, and the greatest decrease in the number of frost days occurs at the Chadegan station under the SSP5-8.5 scenario, which is 26 days. After fitting the marginal distribution functions to the severity (S) and duration (D) of frost events, it was determined that the Generalized Pareto (Gen. Pareto) was the best marginal distribution function for both severity and duration at most stations. The Frank copula (for the stations of Chadegan, Fereydunshahr, Khansar, and Daran) and the Gumbel copula (for the other stations) were selected as the best copula functions. At all stations, the joint bivariate return period in both "AND" and "OR" cases increased in the future period under the emission scenarios compared to the past period, although the increase was less pronounced in the "OR" case. The joint bivariate return period of extreme frost events showed that in both periods, severe frost events occurred in the western parts of the province (Chadegan, Khansar, Daran, and Fereydunshahr), although under the influence of climate changes, the frequency, severity, and duration of frost and freeze events decreased in the future period under the three emission scenarios. The copula function developed in this study has been shown to be a robust tool for predicting the joint return period of frost events.
