基于INFORM模型与GEE的根河火烧迹地冠层含水量反演研究

【目的】提出一种结合辐射传输模型与遥感云平台反演火烧迹地冠层含水量(Canopy water content,CWC)的新方法,弥补目前对火烧迹地恢复阶段植被含水量的监测,为定量监测植被水分与火灾预警提供理论参考。【方法】以内蒙古自治区根河市火烧迹地为研究对象,基于INFORM辐射传输模型,使用查找表的方法反演森林冠层含水量,并结合Google Earth Engine(GEE)大数据遥感平台与Mann-kendall模型分析了火烧迹地的冠层含水量时序性变化,最后绘制根河2018年8月森林冠层含水量分布图。【结果】1)基于样地的CWC反演精度较高(R2为0.79);2)大范围的归一化水分指数(Normalized difference water index, NDWI)和CWC呈指数关系(R2为0.77),但CWC比NDWI的饱和点更高;3)CWC可作为火烧迹地恢复的生态指标,获得了34 a的CWC时序性反演结果,表明火灾后CWC明显降低,并基于Mann-kendall模型得到各样地CWC和LAI的恢复速率。【结论】联合INFORM模型与GEE反演并监测火烧迹地冠层含水量,方法通用且高效。火烧迹地在恢复过程中面临再次发生火烧和虫害病害的风险,研究结果可为森林防火和森林病虫害监测提供技术支持,对该区域森林火灾与森林虫害的预警有一定意义。

Inversion of canopy water content of fire scars in Genhe county based on INFORM model and Google Earth Engine

【Objective】Based on radiative transfer model and cloud-based platform, this paper intends to provide a new method to estimate the canopy water content(CWC) of fire scars. This method builds up monitoring vegetation water content in fire scars and provides a basis for quantitative monitoring the vegetation water content and forest fires.【Method】The look-up table was used to estimate the CWC of the fire scars in Genhe county based on the INFORM model. Additionally, Google Earth engine and Mann-kendall model were used to analysis the temporal change of canopy water content in fire scars. In the end, the map of forest canopy water content in August 2018 of Genhe county was made.【Result】1) The inversion results of canopy water content with higher accuracy are obtained(R~2=0.79). 2) CWC inversion still increases when normalized difference water index(NDWI) reaches at the saturation value. And overall inversion of CWC and NDWI shows an exponential growth relationship(R~2=0.77), which proves that the precision on inversion of CWC by INFORM model is high. 3) The temporal change research of canopy water content in fire scars shows that CWC decreases significantly after the fire, and recovery rates in various plots are different.【Conclusion】This research proves that it is universal and efficient in inverting and monitoring canopy water content of burned areas by INFORM model and GEE. Forest fires affect the stability of forest ecosystem. These areas are at risk of fire and insect pests. The research method and results can provide technical supports for forest fire protection and forest pest control.