Abstract: Predicting the risk of citrus Huanglongbing(HLB)is an effective measure to control the spread of the disease.In order to achieve precise monitoring and prevention of HLB in a large area,a comprehensive multi‑indicator HLB risk assessment at the plot scale was carried out.Taking the citrus plot in Xinxu Town,Mengshan County,Guangxi as the object,based on multi‑temporal NDVI(normalized difference vegetation index),LSWI(land surface water index)and MTCI(MERIS terrestrial chlorophyll index)remote sensing indicators,a citrus HLB monitoring model was established to obtain HLB‑affected plots.Comprehensive migration distance,slope,citrus growth,elevation and average temperature indicators were used to calculate the weight of indicators using the analytic hierarchy process method,and the comprehensive model for evaluating the risk of citrus HLB infection was constructed to assess the risk level of citrus plots infected with HLB.The results showed that the accuracy rate of the citrus HLB monitoring model was 80.31%,and the leakage rate was 10.53%,which effectively reduced the field investigation of citrus HLB.The weights of migration distance,slope,citrus growth,elevation and average temperature were 0.404,0.289,0.186,0.120 and 0.086,respectively.The ability of migration and diffusion of psyllid was the dominant factor in the spread of the disease. The high‑risk area of HLB infection in Xinxu Town was 475.93 ha,the medium‑risk area was 552.00 ha,and the low‑risk area was 1 188.87 ha,accounting for 21.47%,24.90%,and 53.63% of the citrus area,respectively.The risk of further spread of HLB infection is high,and it is suggested that the relevant departments should strengthen the prevention and control of high‑risk areas to prevent the spread of the disease.

Key words: Citrus HLB, Remote sensing technology, Plot scale, Risk assessment, Monitoring

摘要： 柑橘黄龙病受灾风险预判是控制病情蔓延的有效措施，为实现大区域黄龙病精准监测防控，综合多指标开展地块尺度黄龙病风险评估，以广西蒙山县新圩镇柑橘地块为对象，基于多时相归一化植被指数（NDVI）、地表水分指数（LSWI）、MERIS陆地叶绿素指数（MTCI）遥感指标，建立柑橘黄龙病监测模型，获取黄龙病受灾现状地块；综合迁飞距离、坡度、柑橘长势、高程和均温指标，利用层次分析法（AHP）计算指标权重，构建柑橘黄龙病受灾风险评估综合模型，评估柑橘地块感染黄龙病风险等级。结果表明，柑橘黄龙病监测模型准确率为80.31%、漏分率为10.53%，有效减少柑橘黄龙病外业调查；迁飞距离、坡度、柑橘长势、高程和均温评估指标权重分别为0.404、0.289、0.186、0.120、0.086，木虱迁飞扩散能力是病情蔓延主导因素。新圩镇感染黄龙病的高风险区475.93 hm²、中风险区552.00 hm²、低风险区1 188.87 hm²，分别占柑橘面积的21.47%、24.90%、53.63%，黄龙病进一步扩散感染风险性较高，建议有关部门加强对高风险区的防控力度，防止病情蔓延。

关键词: 柑橘黄龙病, 遥感技术, 地块尺度, 风险评估, 监测