关键词: 棉花, 高温热害, 春季霜冻害, Copula, 风险预警

Abstract:

Based on daily meteorological data from 75 national basic meteorological stations in Xinjiang from 1981 to 2024, the events of high temperature damage and spring frost damage were selected and the station ratio, duration and harmful accumulated temperature were calculated to analyze the characteristics of high temperature damage and spring frost damage for cotton. Considering the formation mechanism of natural disaster risk and operational risk early warning requirements, sequences of disaster risk indices and historical daily percentages of disaster affected area with relatively high risks were constructed to evaluate disaster intensity, duration and affected area, as well as to determine risk early warning grade indicators and activation thresholds. Subsequently, the accuracy of risk early warning grade indicator thresholds was verified by comparing cumulative probabilities using the Copula function. The results showed that: (1) from 1981 to 2024, the station ratio, cumulative duration and harmful heat accumulated temperature of high temperature damage increased significantly, with climatic tendency rates of 3.7pp·10y⁻¹, 0.8d·10y⁻¹, 2.5℃·d·10y⁻¹, respectively. The station ratio, cumulative duration and harmful cold accumulated temperature of spring frost damage decreased significantly, with climatic tendency rates of −7.1pp·10y⁻¹, −0.7d·10y⁻¹, −2.2℃·d·10y⁻¹, respectively. The high incidence periods of high temperature damage and spring frost damage for cotton in Xinjiang were July 1^st to August 10^th and April 11^th to May 20^th, respectively. (2) The historical extreme years for high temperature damage and spring frost damage for cotton in Xinjiang were 2015 and 2003, respectively. Their risk indices were 3.2 and 2.7 higher than those of the second highest historical years, respectively. Among them, approximately 50% of the cotton suffered from high temperature damage or spring frost damage for consecutive 4d during July 20−23, 2015 and April 16−19, 2003. (3) There was spatial heterogeneity in the risk early warning grade indicator thresholds for cotton in Xinjiang. The high temperature damage showed a distribution characteristic of “stronger in the east and weaker in the west”, while spring frost damage showed a distribution characteristic of “stronger in the northeast and weaker in the southwest”. The risk early warning grade indicator thresholds were consistent with the thresholds derived from Copula function. The risk early warning activation thresholds for high temperature damage and spring frost damage for cotton in Xinjiang were 3.0% and 5.1% of the cotton planting area affected by disasters at or above the relatively high risk level, respectively. In summary, the risk early warning grade indicators for high temperature damage and spring frost damage for cotton in Xinjiang are reliable. The risk indices can effectively quantify disaster intensity and determine risk early warning indicators, thereby providing a technical reference for carrying out agrometeorological disaster risk early warning services.

Key words: Cotton, High temperature damage, Spring frost damage, Copula, Risk early warning