我国金柑属植物气候适宜区的预测

Prediction of suitable climatic areas for Fortunella species in China

  • 摘要: 近年来, 由于金柑商业种植品种单一导致种植效益降低, 种植面积不断萎缩, 而且随着黄龙病等病害的爆发导致野生金柑属资源分布面积不断缩小。在我国湖南、福建、广西等省份已开展过对金柑属野生的资源收集工作, 但尚未考虑气候对其分布的影响。因此, 亟需了解金柑气候适宜分布区, 为其种质资源的收集和保护提供基础。本研究运用最大熵模型(MaxEnt)和地理信息系统(ArcGIS 10.3), 结合金柑属(Fortunella)植物的实际分布数据和18个影响分布的气候因子, 对金柑属6个种的潜在气候适宜区进行预测, 并利用刀切法(Jackknife test)筛选主导气候因子。结果表明, 金柑属潜在气候适宜区范围主要分布于大别山—大巴山以南地区, 山金柑(F. hindsii)、罗浮金柑(F. margarita)、罗纹金柑(F. japonica)、长叶金柑(F. polyandra)、金弹(F. crassifolia)和长寿金柑(F. obovata)的最适宜区面积为354 000 km2、276 100 km2、495 800 km2、613 600 km2、474 400 km2和663 403 km2; 除长叶金柑外, 其他5个种的气候最适宜区主要分布于湖南、江西、广西、福建、浙江、广东6个省份; 长寿金柑的最适宜区除上述各省份外还包括重庆; 长叶金柑的气候最适宜区包括重庆、广西、广东、海南、云南南部地区。金柑属实际分布点均在预测图中的最适宜区域内, 说明预测结果与实际分布情况高度一致。金柑属6个种适宜区预测模型的曲线下面积(AUC)值均大于0.9, 表明本研究预测分布范围精确度高, 可作为金柑属野生资源的区域调查及保护、优良品种推广的科学依据。

     

    Abstract: Fortunella is one group of the citrus fruit trees in southern China, a kind of edible hesperidium with rich flavonoids, carotenoids, limonoids, coumarins, and the fruit of these evergreen trees also have ornamental value. In recent years, areas of growth of commercial species of Fortunella have been decreasing due to a low net interest of planting single species. Meanwhile, natural resources have also withered due to the Huanglong disease caused by gram-negative bacteria. National collection of germplasm resources of Fortunella has been completed in parts of China, including Hunan, Fujian, Guangxi, etc.; however, this work did not take the effects of climatic factors on their distribution into consideration, which could provide a scientific basis for protection and collection of natural resources of Fortunella species. In this study, we predicted suitable climatic areas for six species (F. hindsii Swingle, F. margarita Swingle, F. japonica Swingle, F. polyandra Tanaka, F. crassifolia Swingle, and F. obovata Tanaka) of Fortunella using the MaxEnt (the maximum entropy, MaxEnt 3.3.3) model and the ArcGIS (the geographic information, ArcGIS 10.3) system to analyze actual geographical distribution data and 18 climate factors affecting their distribution. The dominant climate factors were screened through Jackknife test. Results showed that the suitable climatic region for the six species analyzed were distributed mainly in the southern areas of Dabie—Daba Mountain. The suitable region for the six species covered 354 000 km2 (F. hindsii Swingle), 276 100 km2 (F. margarita Swingle), 495 800 km2 (F. japonica Swingle), 613 600 km2 (F. polyandra Tanaka), 474 400 km2 (F. crassifolia Swingle), and 663 403 km2 (F. obovata Tanaka). The optimum climate region for F. polyandra Tanaka mainly extended to Chongqing, Guangxi, Guangdong, Hainan, and south of Yunnan, while that for the other five species was predominantly distributed in Hunan, Jiangxi, Guangxi, Fujian, Zhejiang, and Guangdong; some favorable region for F. obovata Tanaka even extended to Chongqing. The main climatic factors affecting the growth of Fortunella include minimum temperature, precipitation, and isothermality. The major climatic factors affecting F. hindsii Swingle distribution included isothermality and precipitation in April and June, while for F. margarita Swingle such climatic factors comprised precipitation in April and June, minimum temperature in February and July, and isothermality and precipitation in the driest month. Similarly, geographical distribution of F. japonica Swingle was influenced by precipitation in April and June, minimum temperature in July, isothermality and precipitation in the driest month, and precipitation in the warmest quarter. Likewise, the major climatic factors affecting natural distribution of F. polyandra Tanaka comprised precipitation in July and in December, minimum temperature in February and June, isothermality, and mean precipitation in the warmest quarter. Similar analytical results demonstrated that the dominant climatic factors affecting the distribution of both F. crassifolia Swingle and F. obovata Tanaka were in terms of precipitation in June, and isothermality and mean precipitation of the warmest quarter. The area under the curve (AUC) value of the MaxEnt model for all six species exceeded 0.9 and their actual distribution areas were also integrated, which indicated that the predicted distribution range of this study was highly accurate. Consequently, these results are expected to provide scientific guidance for regional investigation and protection of wild species of Fortunella and promotion of its varieties.

     

/

返回文章
返回