Spruce budworm (Choristoneura fumiferana; SBW) is the most destructive forest pest of northeastern Canada and United States. SBW occurrence as well as the extent and severity of its damage are highly dependent on the characteristics of the forests and the availability of host species namely, spruce (Picea sp.) and balsam fir (Abies balsamea (L.) Mill.). Remote sensing satellite imagery represents a valuable data source for seamless regional-scale mapping of forest composition. This study developed and evaluated new models to map the distribution and abundance of SBW host species at 20 m spatial resolution using Sentinel-1 synthetic aperture radar (SAR) and Sentinel-2 multispectral imagery in combination with several site variables for a total of 191 variables in northern New Brunswick, Canada using the Random Forest (RF) algorithm. We found Sentinel-2 multi-temporal single spectral bands and numerous spectral vegetation indices (SVIs) yielded the classification of SBW host species with an overall accuracy (OA) of 72.6% and kappa coefficient (K) of 0.65. Incorporating Sentinel-1 SAR data with Sentinel-2 variables coupled with elevation, only marginally improved the performance of the model (OA: 73.0% and K: 0.66). The use of Sentinel-1 SAR data with elevation resulted in a reasonable OA of 57.5% and K of 0.47. These spatially explicit up-to-date SBW host species maps are essential for identifying susceptible forests, monitoring SBW defoliation, and minimizing forest losses from insect impacts at landscape scale in the current SBW outbreak in the region.

云杉芽虫（Choristoneura fumiferana； SBW）是加拿大东北部和美国破坏力最大的森林害虫。SBW的发生以及其破坏的程度和严重程度在很大程度上取决于森林的特征以及寄主物种（即云杉（Picea sp。）和香脂冷杉（Abies balsamea）的可用性（L.）Mill。）。遥感卫星图像代表着宝贵的数据来源，可用于无缝区域区域森林组成的制图。这项研究开发并评估了新模型，该模型使用Sentinel-1合成孔径雷达（SAR）和Sentinel-2多光谱成像技术结合20个空间变量，绘制了SBW宿主物种在20 m空间分辨率下的分布和丰度，共计191个变量使用随机森林（RF）算法在加拿大新不伦瑞克省北部。我们发现Sentinel-2多时相单光谱带和众多光谱植被指数（SVI）得出了SBW宿主物种的分类，总准确度（OA）为72.6％，kappa系数（K）的0.65。将Sentinel-1 SAR数据与Sentinel-2变量以及海拔高度结合在一起，只能稍微改善模型的性能（OA：73.0％，K：0.66）。使用具有高度的Sentinel-1 SAR数据可得出合理的OA为57.5％，K为0.47。这些空间明确的最新SBW宿主物种图对于识别易感森林，监测SBW的落叶以及在该地区当前SBW爆发时以景观尺度使昆虫影响的森林损失最小化至关重要。