激光雷达回波模型辅助的坡地森林冠层高度反演

大光斑激光雷达数据已广泛应用于森林冠层高度提取,但通常仅限于地形坡度小于20°的平缓地区。在地形坡度大于20°的陡峭山区,地形引起的波形展宽使得地面回波和植被回波信息混合在一起,给森林冠层高度提取带来巨大挑战。本文利用激光雷达回波模型和地形信息,提出了一种模型辅助的坡地森林冠层高度反演算法。该方法以激光雷达回波信号截止点为参考,定义了波形高度指数H50和H75,使用激光雷达回波模型与已知地形信息模拟裸地的激光雷达回波,将裸地回波信号截止点与森林激光雷达回波信号截止点对齐,利用裸地回波计算常用的波形相对高度指数RH50和RH75,对森林冠层高度进行反演。并与高斯波形分解法和波形参数法的反演结果进行了比较。研究结果表明:(1)利用所提取的波形指数RH50和RH75对胸高断面积加权平均高(Lorey’s height)进行了估算,在坡度小于20°时,高斯波形分解法、波形参数法和模型辅助法的估算结果与实测值线性拟合的相关系数(R2)分别为0.70,0.78和0.98,对应的均方根误差(RMSE)分别为2.90 m,2.48 m和0.60 m,模型辅助法略优于其他两种方法;(2)在坡度大于20°时,高斯波形分解法、波形参数法和模型辅助法的R2分别为0.14,0.28和0.97,相应的RMSE分别为4.93 m,4.53 m和0.81 m,模型辅助法明显优于其他两种方法;(3)在0°—40°时,模型辅助法对Lorey’s height估算结果与实测值的R2为0.97,RMSE为0.80 m。本研究提出的模型辅助法具有更好的地形适应性,在0°—40°的坡度范围内具备对坡地森林冠层高度反演的潜力。     

A novel approach to estimate canopy height using ICESat/GLAS data: A case study in the New Forest National Park,UK

The Geoscience Laser Altimeter System (GLAS) aboard Ice, Cloud and land Elevation Satellite (ICESat) is a spaceborne LiDAR sensor. It is the first LiDAR instrument which can digitize the backscattered waveform and offer near global coverage. Among others, scientific objectives of the mission include precise measurement of vegetation canopy heights. Existing approaches of waveform processing for canopy height estimation suggest Gaussian decomposition of the waveform which has the limitation to properly characterize significant peaks and results in discrepant information. Moreover, in most cases, Digital Terrain Models (DTMs) are required for canopy height estimation. This paper presents a new automated method of GLAS waveform processing for extracting vegetation canopy height in the absence of a DTM. Canopy heights retrieved from GLAS waveforms were validated with field measured heights. The newly proposed method was able to explain 79% of variation in canopy heights with an RMSE of 3.18 m, in the study area. The unexplained variation in canopy heights retrieved from GLAS data can be due to errors introduced by footprint eccentricity, decay of energy between emitted and received signals, uncertainty in the field measurements and limited number of sampled footprints.Results achieved with the newly proposed method were encouraging and demonstrated its potential of processing full-waveform LiDAR data for estimating forest canopy height. The study also had implications on future full-waveform spaceborne missions and their utility in vegetation studies.     

激光雷达森林参数反演研究进展

激光雷达通过发射激光能量和接收返回信号的方式,来获取高精度的森林空间结构和林下地形信息。全波形激光雷达通过记录返回信号的全部能量,得到亚米级植被垂直剖面;离散回波激光雷达记录的单个或多个回波,表示来自不同冠层的回波信号。星载激光雷达一般采用全波形或光子计数激光剖面系统,仅能获取卫星轨道下方的单波束或多波束数据,用于区域/全球范围的森林垂直结构及变化观测。机载激光雷达多采用离散回波或全波形激光扫描系统,能够获取飞行轨迹下方特定视场范围内的扫描数据,用于林分/区域范围的森林结构观测。地基激光雷达多采用离散回波激光扫描系统,获取以测站为中心的球形空间内扫描数据,用于单木/样地范围的森林结构观测。激光雷达单木因子估测方法可分为CHM单木法、NPC单木法和体元单木法3类。CHM单木法通过局部最大值识别树冠顶点,采用区域生长或图像分割算法识别树冠边界或树冠主方向,NPC单木法一般通过空间聚类或形态学算法识别单木,体元单木法在3维体元空间采用区域生长或空间聚类算法识别树冠。根据激光雷达冠层高度分布可以估测林分因子,冠层高度分布特征来自于离散点云或全波形。多时相激光雷达可用于森林生长量、生物量变化等监测,以及森林采伐、灾害等引起的结构变化监测。随着激光雷达技术的发展,它将在森林调查、生态环境建模等生产与科学研究领域中得到更为广泛的应用。     

Forest canopy height estimation using ICESat/GLAS data and error factor analysis in Hokkaido,Japan

Spaceborne light detection and ranging (LiDAR) enables us to obtain information about vertical forest structure directly, and it has often been used to measure forest canopy height or above-ground biomass. However, little attention has been given to comparisons of the accuracy of the different estimation methods of canopy height or to the evaluation of the error factors in canopy height estimation. In this study, we tested three methods of estimating canopy height using the Geoscience Laser Altimeter System (GLAS) onboard NASA’s Ice, Cloud, and land Elevation Satellite (ICESat), and evaluated several factors that affected accuracy. Our study areas were Tomakomai and Kushiro, two forested areas on Hokkaido in Japan. The accuracy of the canopy height estimates was verified by ground-based measurements. We also conducted a multivariate analysis using quantification theory type I (multiple-regression analysis of qualitative data) and identified the observation conditions that had a large influence on estimation accuracy. The method using the digital elevation model was the most accurate, with a root-mean-square error (RMSE) of 3.2 m. However, GLAS data with a low signal-to-noise ratio (⩽10.0) and that taken from September to October 2009 had to be excluded from the analysis because the estimation accuracy of canopy height was remarkably low. After these data were excluded, the multivariate analysis showed that surface slope had the greatest effect on estimation accuracy, and the accuracy dropped the most in steeply sloped areas. We developed a second model with two equations to estimate canopy height depending on the surface slope, which improved estimation accuracy (RMSE = 2.8 m). These results should prove useful and provide practical suggestions for estimating forest canopy height using spaceborne LiDAR.     

中国南方森林冠顶高度Lidar反演—以江西省为例

激光雷达(Lidar)与光学遥感的有效结合对中国南方区域森林冠顶高度反演意义重大,而国产卫星将为中国森林生态研究提供新的数据源。本文联合利用大脚印激光雷达GLA和国产MERSI数据,在实现GLAS波形数据处理和不同地形条件下森林冠顶高度反演算法基础上,建立了区域尺度不同森林类型林分冠顶高度GLAS+MERSI联合反演关系模型,进行了江西地区森林冠顶高度反演。总体上,GLAS激光雷达森林冠顶高度估算精度较高;且在与MERSI 250 m数据的联合反演模型中,针叶林模型精度较好(R²=0.7325);阔叶林次之(R²=0.6095);混交林较差(R²=0.4068)。分析发现,考虑了光学遥感生物物理参数的GLAS+MERSI联合关系模型在区域森林冠顶高度估算中有较高精度,且在空间分布上与土地覆盖数据分布特征非常一致。     

基于ICESatGLAS的云南省森林地上生物量反演

结合机载、星载激光雷达对GLAS(地球科学激光测高系统)光斑范围内的森林地上生物量进行估测,并利用MODIS植被产品以及MERIS土地覆盖产品进行了云南省森林地上生物量的连续制图。机载LiDAR扫描的260个训练样本用于构建星载GLAS的森林地上生物量估测模型,模型的决定系数(R2)为0.52,均方根误差(RMSE)为31Mg/ha。研究结果显示,云南省总森林地上生物量为12.72亿t,平均森林地上生物量为94Mg/ha。估测的森林地上生物量空间分布情况与实际情况相符,森林地上生物量总量与基于森林资源清查数据的估测结果相符,表明了利用机载LiDAR与星载ICESatGLAS结合进行大区域森林地上生物量估测的可靠性。     

多源遥感数据支持下的区域性森林冠层高度估测

结合机载LiDAR数据,提出了一种改进的GLAS光斑点冠层高度地形校正模型,以校正后的GLAS光斑点作为输入样本,结合MODIS遥感影像,利用支持向量回归（SVR）的方法对研究区森林冠层高度进行分生态区估测,并利用野外调查数据和机载LiDAR冠层高度结果对估测结果进行验证。结果显示:研究区的坡度等级直接影响GLAS光斑点森林冠层高度估测精度,改进的地形校正模型可以较好的减小坡度对GLAS光斑点森林冠层高度估测的影响,模型精度RMSE稳定在3.25~3.48 m;不同生态分区的SVR模型估测精度较为稳定,其RMSE=6.41~7.56 m;与算数平均高相比,样地的Lorey's高与制图结果拟合最好,不同生态分区平均估测精度为80.3%。机载LiDAR冠层高度结果的验证平均精度为79.5%,和Lorey's高验证结果呈现较好的一致性。     

吉林长白山森林冠顶高度激光雷达与MERSI联合反演

将激光雷达与光学遥感相结合进行区域林分冠顶高度联合反演,提出了大脚印激光雷达GLAS脚点波形数据处理和不同地形条件下的森林冠顶高度反演算法,并建立了区域尺度不同森林类型林分冠顶高度GLAS＋MERSI联合反演模型,制作了长白山地区森林冠顶高度图。     

Waveform LiDAR concepts and applications for potential vegetation phenology monitoring and modeling:a comprehensive review

Researchers continually demonstrated through published literature how LiDAR could create unparalleled measurements of ecosystem structure and forest height. There are a number of studies conducted utilizing waveform LiDAR products for terrestrial monitoring, but those that deal specifically with the assessment of space-borne waveform LiDAR for monitoring and modeling of phenology is very limited. This review highlights the waveform LiDAR system and looks into satellite sensors that could link waveform LiDAR and vegetation phenology, such as the proposed NASA's Global Ecosystem Dynamics Investigation (GEDI) and the Japanese Experimental Module (JEM)-borne LiDAR sensor named MOLI (Multi-footprint Observation LIDAR and Imager). Further, this work examines the richness and utility of the waveform returns and proposes a spline-function-derived model that could be exploited for estimating the leaf-shooting date. The new approach may be utilized for ecosystem-level phenological studies.     

GLAS星载激光雷达和Landsat/ETM+数据的森林生物量估算

基于大脚印激光雷达数据和野外观测数据,该文提出一种获取脚印点内森林生物量的新思路,并结合陆地卫星数据应用于长白山地区森林地上生物量估算。首先,基于3种森林类型(针叶林、阔叶林和针阔混交林),采用多元逐步回归方法建立激光雷达波形指数与脚印点内实测平均树高的回归模型,估算全部脚印点内的平均树高;然后根据脚印点内样方的野外观测数据(平均树高和平均胸径)以及它们与样方生物量的拟合方程估算没有野外调查数据对应的脚印点的生物量;最后对3种森林类型的脚印点森林生物量在各森林覆盖度条件下进行分层分区统计得到生物量等级图。验证比较遥感估算的生物量与野外调查数据推算的生物量,总体误差在0~30(t·hm~(-2))之间,均方根误差为14.66(t·hm~(-2))。     

基于遥感的区域尺度森林地上生物量估算研究

森林是陆地生态系统最大的碳库,精确估算森林生物量是陆地碳循环研究的关键。首先从机载LiDAR数据中提取高度和密度统计量,采用逐步回归模型进行典型样区生物量估算;然后利用机载LiDAR数据估算的生物量作为样本数据,与多光谱遥感数据Landsat8 OLI的波段反射率及植被指数建立回归模型,实现区域尺度森林地上生物量估算。实验结果显示,机载LiDAR数据估算的鼎湖山样区生物量与地面实测生物量的相关性R2达0.81,生物量RMSE为40.85 t/ha,说明机载LiDAR点云数据的高度和密度统计量与生物量存在较高的相关性。以机载LiDAR数据估算的生物量为样本数据,结合多光谱遥感数据Landsat8 OLI估算粤西北地区的森林地上生物量,精度验证结果为:R2为0.58,RMSE为36.9 t/ha;针叶林、阔叶林和针阔叶混交林等3种不同森林类型生物量的估算结果为:R2分别为0.51(n=251)、0.58(n=235)和0.56(n=241),生物量RMSE分别为24.1 t/ha、31.3 t/ha和29.9 t/ha,估算精度相差不大。总体上看,利用遥感数据可以开展区域尺度的森林地上生物量估算,为森林固碳监测提供有力的参考数据。     

高分二号异轨立体数据的森林高度提取

森林植被碳储量的空间分布格局及其动态变化是陆地生态系统碳收支核算的基础。作为森林地上生物量的重要指示因子,森林高度的精确估算是提高森林植被碳储量估算精度的关键。现有研究已证明,由专业星载摄影测量系统获取的立体观测数据可用于森林高度提取,但光学遥感数据最大的问题是受云雨等天气因素的影响严重。区域森林地上生物量产品的生产需要充分挖掘潜在数据源。国产高分二号卫星(GF-2)虽然不是为获取立体观测数据而设计的专业星载摄影测量系统,但其获取的图像空间分辨率可达0.8 m,且具备±35°的的侧摆能力,在重复观测区域可构成异轨立体观测。本文以分别获取于2015年6月20日和2016年7月19的GF-2数据作为立体像对,其标称轨道侧摆角分别为0.00118°和20.4984°,以激光雷达数据获取的林下地形(DEM)和森林高度(CHM)为参考,对利用GF-2立体观测数据进行森林高度提取进行了研究。通过对立体处理得到的摄影测量点云的栅格化得到DSM,以激光雷达数据提供的DEM作为林下地形,得到了GF-2的CHM。结果表明GF-2提取的CHM与激光雷达CHM空间分布格局较为一致,两者之间存在明显的相关性,像素对像素的线性相关性(R2)达到0.51,均方根误差(RMSE)为3.6 m。研究结果表明,在林下地形已知的情况下,GF-2立体观测数据可用于森林高度估算。     

Assessment of terrain elevation derived from satellite laser altimetry over mountainous forest areas using airborne lidar data

Gaussian decomposition has been used to extract terrain elevation from waveforms of the satellite lidar GLAS (Geoscience Laser Altimeter System), on board ICESat (Ice, Cloud, and land Elevation Satellite). The common assumption is that one of the extracted Gaussian peaks, especially the lowest one, corresponds to the ground. However, Gaussian decomposition is usually complicated due to the broadened signals from both terrain and objects above over sloped areas. It is a critical and pressing research issue to quantify and understand the correspondence between Gaussian peaks and ground elevation. This study uses ～2000 km² airborne lidar data to assess the lowest two GLAS Gaussian peaks for terrain elevation estimation over mountainous forest areas in North Carolina. Airborne lidar data were used to extract not only ground elevation, but also terrain and canopy features such as slope and canopy height. Based on the analysis of a total of ～500 GLAS shots, it was found that (1) the lowest peak tends to underestimate ground elevation; terrain steepness (slope) and canopy height have the highest correlation with the underestimation, (2) the second to the lowest peak is, on average, closer to the ground elevation over mountainous forest areas, and (3) the stronger peak among the lowest two is closest to the ground for both open terrain and mountainous forest areas. It is expected that this assessment will shed light on future algorithm improvements and/or better use of the GLAS products for terrain elevation estimation.     

机器学习算法在森林地上生物量估算中的应用

森林地上生物量是森林生产力的重要评价指标,对其进行高效监测对维持全球碳平衡和保护生态系统具有重要意义。本文首先基于冠层高度模型数据,通过分水岭分割算法得到单木冠幅边界;然后在单木冠幅范围内提取23个LiDAR变量,结合佩诺布斯科特试验森林的87组实测数据,利用随机森林和支持向量机建立森林地上生物量估算模型;最后对样地模型估算的结果进行了比较,讨论了预测结果及其精度。结果表明:本文选用的随机森林模型和支持向量机模型在估算森林地上生物量的应用中获得了较高的精度;并且,随机森林模型在基于机载雷达数据估测森林地上生物量中的估算精度更高,模型泛化能力更强,制图精度也更好,具有更好的适用性。     

大光斑激光雷达数据的森林冠层高度反演

针对传统的LM波形分解算法在GLAS大光斑波形数据处理中容易陷于局部最优解,限制了GLAS大光斑激光雷达数据在森林结构参数反演方面应用的问题,该文结合GLAS大光斑数据特征,引进优化后的EM算法对大光斑全波形数据进行分解,获取波形参数最优值。结合波形前缘长度和波形后缘长度,建立树高反演模型,并与LM分解算法建立的模型进行对比分析。研究结果表明,通过改进的EM算法对GLAS大光斑激光雷达数据进行处理,波形特征参数的获取更为精确,达到了较高的树高反演精度。     

Lidar-aided analysis of boreal forest backscatter at Ku band

This paper analyzes the backscatter of the microwave signal in a boreal forest environment based on a Ku -band airborne Frequency-Modulated Continuous Waveform (FMCW) profiling radar—Tomoradar. We selected a half-managed boreal forest in the southern part of Finland for a field test. By decomposing the waveform collected by the Tomoradar, the vertical canopy structure was achieved. Based on the amplitude of the waveform, the Backscattered Energy Ratio of Canopy-to-Total (BERCT) was calculated. Meanwhile, the canopy fraction was derived from the corresponding point cloud recorded by a Velodyne VLP-16 LiDAR mounted on the same platform. Lidar-derived canopy fraction was obtained by counting the number of the first/ the strongest returns versus the total amount of returns. Qualitative and quantitative analysis of radar-derived BERCT on lidar-derived canopy fraction and canopy height are investigated. A fitted model is derived to describe the Ku-band microwave backscatter in the boreal forest to numerically analyze the proportion contributed by four factors: lidar-derived canopy fraction, radar-derived canopy height, the radar-derived distance between trees and radar sensor and other factors, from co-polarization Tomoradar measurements. The Root Mean Squared Error (RMSE) of the proposed model was 0.0958, and the coefficient of determination R2 was 0.912. The fitted model reveals that the correlation coefficient between radar-derived BERCT and lidar-derived canopy fraction is 0.84, which illustrates that lidar surface reflection explains the majority of the profiling /waveform radar response. Thus, vertical canopy structure derived from lidar can be used for the benefit of radar analysis.     

Estimation of forest leaf water content through inversion of a radiative transfer model from LiDAR and hyperspectral data

The accurate estimation of leaf water content (LWC) and knowledge about its spatial variation are important for forest and agricultural management since LWC provides key information for evaluating plant physiology. Hyperspectral data have been widely used to estimate LWC. However, the canopy reflectance can be affected by canopy structure, thereby introducing error to the retrieval of LWC from hyperspectral data alone. Radiative transfer models (RTM) provide a robust approach to combine LiDAR and hyperspectral data in order to address the confounding effects caused by the variation of canopy structure. In this study, the INFORM model was adjusted to retrieve LWC from airborne hyperspectral and LiDAR data. Two structural parameters (i.e. stem density and crown diameter) in the input of the INFORM model that affect canopy reflectance most were replaced by canopy cover which could be directly obtained from LiDAR data. The LiDAR-derived canopy cover was used to constrain in the inversion procedure to alleviate the ill-posed problem. The models were validated against field measurements obtained from 26 forest plots and then used to map LWC in the southern part of the Bavarian Forest National Park in Germany. The results show that with the introduction of prior information of canopy cover obtained from LiDAR data, LWC could be retrieved with a good accuracy (R² = 0.87, RMSE = 0.0022 g/cm², nRMSE = 0.13). The adjustment of the INFORM model facilitated the introduction of prior information over a large extent, as the estimation of canopy cover can be achieved from airborne LiDAR data.     

A new 500-m resolution map of canopy height for Amazon forest using spaceborne LiDAR and cloud-free MODIS imagery

In the present study, we aimed to map canopy heights in the Brazilian Amazon mainly on the basis of spaceborne LiDAR and cloud-free MODIS imagery with a new method (the Self-Organizing Relationships method) for spatial modeling of the LiDAR footprint. To evaluate the general versatility, we compared the created canopy height map with two different canopy height estimates on the basis of our original field study plots (799 plots located in eight study sites) and a previously developed canopy height map. The compared canopy height estimates were obtained by: (1) a stem diameter at breast height (D) – tree height (H) relationship specific to each site on the basis of our original field study, (2) a previously developed D–H model involving environmental and structural factors as explanatory variables (Feldpausch et al., 2011), and (3) a previously developed canopy height map derived from the spaceborne LiDAR data with different spatial modeling method and explanatory variables (Simard et al., 2011). As a result, our canopy height map successfully detected a spatial distribution pattern in canopy height estimates based on our original field study data (r = 0.845, p = 8.31 × 10⁻³) though our canopy height map showed a poor correlation (r = 0.563, p = 0.146) with the canopy height estimate based on a previously developed model by Feldpausch et al. (2011). We also confirmed that the created canopy height map showed a similar pattern with the previously developed canopy height map by Simard et al. (2011). It was concluded that the use of the spaceborne LiDAR data provides a sufficient accuracy in estimating the canopy height at regional scale.     

Assessing the potential for leaf-off LiDAR data to model canopy closure in temperate deciduous forests

Estimates of canopy closure have many important uses in forest management and ecological research. Field measurements, however, are typically not practical to acquire over expansive areas or for large numbers of locations. This problem has been addressed, in recent years, through the use of airborne light detection and ranging (LiDAR) technology which has proven effective in modeling canopy closure remotely. The techniques developed to use LiDAR for this purpose have been designed and evaluated for datasets acquired during leaf-on conditions. However, a large number of LiDAR datasets are acquired during leaf-off conditions since their primary purpose is to generate bare-earth Digital Elevation Models. In this paper, we develop and evaluate techniques for leveraging small-footprint leaf-off LiDAR data to model leaf-on canopy closure in temperate deciduous forests.We evaluate three techniques for modeling canopy closure: (1) the canopy-to-total-return-ratio (CTRR), (2) the canopy-to-total-pixel-ratio (CTPR), and (3) the hemispherical-viewshed (HV). The first technique has been used widely, in various forms, and has been shown to be effective with leaf-on LiDAR datasets. The CTRR technique that we tested uses the first-return LiDAR data only. The latter two techniques are new contributions that we develop and present in this paper. These techniques use Canopy Height Models (CHM) to detect significant gaps in the forest canopy which are of primary importance in estimating closure.The techniques we tested each showed good promise for predicting canopy closure using leaf-off LiDAR data with the CTPR and HV models having particularly high correlations with closure estimates from hemispherical photographs. The CTRR model had performance on par with results from previous studies that used leaf-on LiDAR, although, with leaf-off data the model tended to be negatively biased with respect to species having simple and compound leaf types and positively biased for coniferous species. The CTPR and HV models also showed some slight negative biases for compound-leaf species. The biases for the CTPR and HV models were mitigated when the CHM data were smoothed to fill in small gaps. The CHM-based models were robust to changes in the CHM model resolution which suggests that these methods may be applicable to a variety of small-footprint LiDAR datasets. In this research, the new CTPR and HV methods showed a strong ability to predict canopy closure using leaf-off data, however, future work will be needed to test the applicability of the models to variations in LiDAR datasets, forest types, and topography.     

High-resolution mapping of forest canopy height using machine learning by coupling ICESat-2 LiDAR with Sentinel-1, Sentinel-2 and Landsat-8 data

Forest canopy height is an important indicator of forest carbon storage, productivity, and biodiversity. The present study showed the first attempt to develop a machine-learning workflow to map the spatial pattern of the forest canopy height in a mountainous region in the northeast China by coupling the recently available canopy height (H_canopy) footprint product from ICESat-2 with the Sentinel-1 and Sentinel-2 satellite data. The ICESat-2 H_canopy was initially validated by the high-resolution canopy height from airborne LiDAR data at different spatial scales. Performance comparisons were conducted between two machine-learning models – deep learning (DL) model and random forest (RF) model, and between the Sentinel and Landsat-8 satellites. Results showed that the ICESat-2 H_canopy showed the highest correlation with the airborne LiDAR canopy height at a spatial scale of 250 m with a Pearson’s correlation coefficient (R) of 0.82 and a mean bias of -1.46 m, providing important evidence on the reliability of the ICESat-2 vegetation height product from the case in China’s forest. Both DL and RF models obtained satisfactory accuracy on the upscaling of ICESat-2 H_canopy assisted by Sentinel satellite co-variables with an R-value between the observed and predicted H_canopy equalling 0.78 and 0.68, respectively. Compared to Sentinel satellites, Landsat-8 showed relatively weaker performance in H_canopy prediction, suggesting that the addition of the backscattering coefficients from Sentinel-1 and the red-edge related variables from Sentinel-2 could positively contribute to the prediction of forest canopy height. To our knowledge, few studies have demonstrated large-scale vegetation height mapping in a resolution ≤ 250 m based on the newly available satellites (ICESat-2, Sentinel-1 and Sentinel-2) and DL regression model, particularly in the forest areas in China. Thus, the present work provided a timely and important supplementary to the applications of these new earth observation tools.