中国农业气象 ›› 2025, Vol. 46 ›› Issue (12): 1759-1769.doi: 10.3969/j.issn.1000-6362.2025.12.007

• 农业生物气象栏目 • 上一篇    下一篇

基于MaxEnt模型的广东沙田柚种植适生区预测

陈金星,罗碧瑜,李敏勋,林立进,余东柏,黄浩明,刘雯丽,林思华   

  1. 1.广东省梅州市气象局,梅州 514000;2.广东省大埔县气象局,大埔 514200;3.广东省五华县气象局,五华 514400
  • 收稿日期:2025-01-14 出版日期:2025-12-20 发布日期:2025-12-16
  • 作者简介:陈金星,工程师,研究方向为农业气象服务与应用,E-mail:2576473699@qq.com
  • 基金资助:
    广东省气象局科技项目(GRMC2024M45);梅州市气象局科技项目(2024M01);梅州市社会发展科技计划项目(2023C0301003)

Prediction of Suitable Areas for Cultivation of Shatian Pomelo in Guangdong Province Based on MaxEnt Model

CHEN Jin-xing, LUO Bi-yu, LI Min-xun, LIN Li-jin, YU Dong-bai, HUANG Hao-ming, LIU Wen-li, LIN Si-hua   

  1. 1. Meteorological Bureau of Meizhou City, Meizhou 514000, China; 2. Meteorological Bureau of Dabu County, Dabu 514200; 3. Meteorological Bureau of Wuhua County, Wuhua 514400
  • Received:2025-01-14 Online:2025-12-20 Published:2025-12-16

摘要:

利用广东省25个沙田柚样地地理分布数据和27个环境因子,运用ENMeval包优化的最大熵模型(MaxEnt)和ArcGIS软件,预测历史时期和CMIP6气候模式中未来温室气体低排放、高排放情景下广东沙田柚种植适生区分布,分析影响沙田柚分布的主导环境因子。结果表明:(1MaxEnt模型预测广东沙田柚种植适生区分布效果较好,曲线下面积AUC值达0.912。(2)基于历史时期(19702000年)环境因子分析,广东沙田柚低适生区分布范围最广,约10.56km2,占广东省土地总面积的58.7%,主要分布在粤西地区、珠三角地区和粤东沿海地区;广东沙田柚中、高适生区面积约5.59km2,占广东省土地总面积的31.1%,主要分布在梅州市、河源市、韶关市、清远市和肇庆市。其中梅州市沙田柚高适生区总面积最大,韶关市次之。(3)影响广东沙田柚适生种植的主导环境因子为最潮湿月降水量(Bio13)、平均昼夜温差(Bio2)、最暖季降水量(Bio18)、坡向(ASPECT)。(4)在SSP12.6SSP58.5温室气体排放(分别代表低、高排放)情景下,20212040年广东沙田柚的中、高适生区总面积相比历史时期分别扩大2.5倍、2.3倍。其中,低温室气体排放情景下,广东沙田柚高适生区面积最大,约5.72km2,占广东省土地总面积的31.8%。总体而言,未来气候变化将有利于广东省沙田柚适生区范围扩大,建议在粤西地区及珠三角北部市县开展沙田柚栽种实验,验证沙田柚在当地环境中的综合适宜性,探索大规模推广引种的可行性。

关键词: 广东, MaxEnt模型, 沙田柚, 适生区预测

Abstract:

To rationally plan the planting distribution of Shatian pomelo in Guangdong, this study employed the ENMeval-optimized Maximum Entropy (MaxEnt) model and ArcGIS software, utilizing 25 geographical distribution records of Shatian pomelo and 27 environmental factors. The aim was to simulate the potential suitable planting areas under historical climatic conditions and future scenarios based on the CMIP6 Shared Socioeconomic Pathways (SSPs), specifically SSP1-2.6 and SSP5-8.5. Additionally, the dominant environmental factors influencing the distribution of Shatian pomelo were analyzed. The simulation was performed using the Maximum Entropy Model (MaxEnt), optimized by the ENMeval package, and analyzed using ArcGIS software. The goal was to identify and analyze the dominant environmental factors influencing the distribution of Shatian pomelo. The results showed that: (1) the MaxEnt model effectively simulated the distribution of suitable cultivation areas for Guangdong Shatian pomelo, with the area under the curve (AUC) reaching 0.912, indicating high model accuracy. (2) During the historical period (19702000), low−suitability areas for Guangdong Shatian pomelo covered the largest extent, approximately 105600 km², accounting for 58.7% of the total land area of Guangdong province. These areas were mainly located in the western region of Guangdong, the Pearl river delta and the eastern coastal areas. In contrast, medium− and high−suitability areas, covering about 55900km² (31.1% of the province’s total area), were primarily concentrated in the cities of Meizhou, Heyuan, Shaoguan, Qingyuan and Zhaoqing cities. Among these, Meizhou had the largest highly suitable area, followed by Shaoguan. (3) The dominant environmental factors influencing the suitable cultivation of Shatian pomelo were identified as the precipitation of the wettest month (Bio13), the mean diurnal temperature range (Bio2), the precipitation of the warmest quarter (Bio18) and slope aspect (ASPECT). (4) Under the SSP1−2.6 and SSP5−8.5 scenarios, the total area of medium− and high−suitability zones for Shatian pomelo in Guangdong was projected to expand by 2.5 and 2.3 times, respectively, during the period 2021 to 2040 compared with the historical period. The SSP1−2.6 scenario was expected to produce the largest area of highly suitable zones, covering approximately 57200km², or 31.8% of Guangdong's total land area. Overall, climate change was expected to contribute to the expansion of suitable areas for Shatian pomelo cultivation in Guangdong province. Based on these findings, it is recommended that experimental trials for Shatian pomelo planting be conducted in the western part of Guangdong and the northern cities and counties of the Pearl river delta to assess the comprehensive environmental suitability and explore the feasibility of largescale cultivation and introduction of Shatian pomelo.

Key words: Guangdong, MaxEnt model, Shatian pomelo, Fertile zone prediction