热红外遥感对马尾松林松材线虫病早期响应研究

松材线虫病由于传播速度快、致死率高等特点也被称“松树的癌症”,早期监测非常重要。基于Landsat 8 TIRS热红外遥感数据评估绿色攻击阶段的植被温度的异常变化，并探究其监测潜力。应用2018年江西省吉安市万安县的Landsat 8 TIRS遥感数据，通过单窗算法反演地表温度(LST)。以30个100 m×100 m样地的受害马尾松林的枯梢率为指标，分别开展5月(绿色阶段)、7月(绿色攻击阶段)及11月(红褐色-灰色阶段)3个不同月份的LST与枯梢率(SDR)的相关性分析。对比分析热红外和Sentinel-2光学数据在早期监测能力上的差异。结果表明，1)Landsat 8反演的地表温度LST分布合理，对虫害区域有响应；2)LST与枯梢率在5月(0.01

Early Response of Thermal Infrared Remote Sensing to Pine Wood Nematode in Pinus massoniana Forest

Pine wood nematode disease has the characteristics of fast transmission speed and high mortality.Known as "the cancer of pines",early detecting is very important.In this paper,Landsat 8 TIRS thermal infrared remote sensing data were adopted to evaluate the monitoring potential of vegetation temperature anomalies during the green attack phase.Firstly,based on Landsat 8 TIRS remote sensing data of Wan’an County,Ji’an City,Jiangxi Province in 2018,land surface temperature (LST) was retrieved by mono-window algorithm.Then the correlation between LST and shoot dieback rate (SDR) were analyzed in three different months (May,July and November) based on the shoot dieback rate of 20 100 m×100 m sample plots of masson pine forest in Ji’an City.Finally,the differences in early monitoring capabilities were analyzed between thermal infrared and Sentinel-2.The results showed that 1) the land surface temperature LST retrieved by Landsat 8 had a reasonable distribution and responds to the pest area.2) There was a good correlation between LST and dieback rate in May (0.01<P<0.05),July (P<0.01) and November (P<0.01).Combined spectral index could improve the accuracy of SDR inversion results.By comparing the accuracy results,it was found that the model established by NDVI and LST had the highest accuracy.3) In addition to the NDVI (P<0.01),the relationship between spectral index and dieback rate was not significant in May and July.It is possible to monitor the early environmental changes of pine wood nematode disease through thermal infrared remote sensing,about 20 days earlier than optical band.