基于MaxEnt模型的红鳍笛鲷全球分布格局

Global distribution pattern of Lutjanus erythropterus based on the MaxEnt model

  • 摘要:
    目的 基于最大熵模型(MaxEnt)对红鳍笛鲷在全球海域的潜在适生区分布进行预测,以期为该物种的增养殖选址与渔业资源管理提供科学依据。
    方法 整合350个红鳍笛鲷有效分布点和22个相对独立的海洋环境变量,利用MaxEnt模型开展分布预测分析,通过Jacknife检验和SHAP分析筛选关键环境因子,并通过响应曲线反应其生态阈值,以受试者工作特征曲线下面积(AUC值)评估模型精度。
    结果 训练及测试AUC均值均大于0.9,表明模型预测具有极高的准确性和可靠性。综合SHAP分析和Jacknife检验,筛选水深、光合有效辐射和夏季温度(5—10月)作为影响红鳍笛鲷生存的关键海洋环境变量。响应曲线结果显示红鳍笛鲷对水深的高度适宜范围为78 m以内;对光合有效辐射的高度适宜范围为46.2~47.5 E/(m2·d);对夏季温度(5—10月)的高度适宜范围为26.8~28.8 ℃。全球高度适生区总面积约67.45万km2,集中分布于澳大利亚北部沿海,并碎片化分布于东南亚沿岸、红海及东非沿海等地。模型预测结果与实际分布基本一致,但红海等区域因物种互作或历史原因未见实际分布记录。
    结论 MaxEnt模型适用于红鳍笛鲷适生区的预测,结果可靠。本次研究明确了红鳍笛鲷全球适生区分布特征及关键环境变量(水深、光合有效辐射和夏季温度),基于此,建议在澳大利亚北部、东南亚沿岸等模型识别出的高、中适生区,优先发展增养殖工作,以期为该重要经济物种的资源养护与可持续利用提供科学依据。

     

    Abstract: Lutjanus erythropterus is a commercially important marine fish species with undefined global potential suitable habitats. To clarify the distribution of the global potential suitable habitats of this species and provide a theoretical basis for optimizing the site selection of its stock enhancement and aquaculture, this study conducted a predictive analysis of potential suitable habitats in global marine based on the MaxEnt model, by integrating 350 distribution points and 22 marine environmental variables. Key environmental variables were screened through jacknife test and SHAP analysis, their ecological suitability thresholds were determined in combination with response curves, and the model accuracy was evaluated using AUC values. The results showed that the mean AUC values of both the training and test sets of the model were exceeded 0.9, indicating extremely high prediction accuracy and reliability. Water depth, photosynthetically active radiation, and summer temperature (May-Oct.) were identified as the key environmental variables regulating the distribution of this species. The highly suitable habitat conditions for L. erythropterus were water depth within 78 m, photosynthetically active radiation of 46.2-47.5 E/(m2·d), and summer temperature of 26.8-28.8 ℃. The total area of its global highly suitable habitats was approximately 674 500 km2, mainly concentrated along northern Australian coasts and fragmentally distributed along the coasts of Southeast Asia, the Red Sea and East Africa. Model predictions were generally consistent with the actual distribution of L. erythropterus, yet no actual distribution records were found in predicted areas such as the Red Sea, presumably due to interspecific interactions and historical dispersal. This study confirmed that the MaxEnt model is applicable to the prediction of suitable habitats for L. erythropterus with reliable results. It clarified the distribution characteristics of the global suitable habitats of this species and the key influencing variables, and recommended that the high and moderate suitable habitats such as northern Australian and Southeast Asian coasts be designated as priority areas for stock enhancement and aquaculture.

     

/

返回文章
返回