CAJ | 학술논문

为提高东、黄海鲐鱼渔场预报准确率、降低渔业生产成本，研究提出了一种基于提升回归树的渔场预报模型。研究采用2003—2010年我国大型灯光围网渔捞日志数据，以有网次记录的小渔区为渔场，以渔捞日志未记录的区域作为背景场随机选择假定非渔场数据，以海表水温等环境因子作为预测变量构建东、黄海鲐鱼渔场预报模型并以2011年的实际作业记录对预报模型进行精度验证。验证计算得到预报模型的 AUC（area under receiver operating curve）值为0．897，表明模型的预报精度较高。模型的空间预测结果表明，预报渔场与实际作业位置基本吻合，其位置移动也与实际情况相符。这表明基于提升回归树的渔场预报模型可以用来进行东、黄海鲐鱼渔场的预报。
위제고동、황해태어어장예보준학솔、강저어업생산성본，연구제출료일충기우제승회귀수적어장예보모형。연구채용2003—2010년아국대형등광위망어로일지수거，이유망차기록적소어구위어장，이어로일지미기록적구역작위배경장수궤선택가정비어장수거，이해표수온등배경인자작위예측변량구건동、황해태어어장예보모형병이2011년적실제작업기록대예보모형진행정도험증。험증계산득도예보모형적 AUC（area under receiver operating curve）치위0．897，표명모형적예보정도교고。모형적공간예측결과표명，예보어장여실제작업위치기본문합，기위치이동야여실제정황상부。저표명기우제승회귀수적어장예보모형가이용래진행동、황해태어어장적예보。
To improve the accuracy of fishing ground forecasting of chub mackerel (Scomber japonicus )in the Yellow and East China Sea,and reduce the fishery production cost,a new fishing ground forecasting model based on boosted regression trees was proposed in this study.Model was fitted with data extracted from electronic logbooks of Chinese mainland large-type lighting purse seine fishery for chub mackerel,with a range from 2003 to 2010.The fishing area with fishing effort was identified as fishing ground and the pseudo non fishing ground data was ran-domly collected from background field,which is the fishing areas with no records in the logbooks.The predictive variables were sea surface temperature and other environmental factors.The performance of prediction of the model was evaluated with the testing dataset consist of actual fishing locations of year 2011.The results of the evaluation showed that the prediction model had a high prediction performance with an AUC value of 0.897.The results of spatial prediction showed that the predicted fishing ground and its shifting were coincided with the actual fishing lo-cations,which indicated that the forecasting model based on boosted regression trees can be used to forecasting the fishing ground of chub mackerel in the Yellow and East China Sea.