SP3: Novel Earth Observation techniques for forest structure analyses and multi-scale characterization of forests

对地观测是获取大尺度地球表面空间清晰信息和监测其随时间变化的有力手段。与森林有关的遥感应用有各种各样。这是基于主动传感器系统（如激光雷达和合成孔径雷达）主要对森林结构（水平和垂直）的敏感性，以及被动-主要是光学-系统主要对冠层的生物化学特性的敏感性（如多光谱或高光谱传感器）。然而，最近的科学工作表明，无人机上的高空间分辨率无源光学传感器有能力量化地面（树冠层以下）的森林特征。尽管有这些新的发现，但并没有一个适合所有人的传感器。虽然存在越来越多地组合对特定性质敏感的不同传感器系统的趋势，（例如，小尺度激光雷达测量作为当地尺度森林结构的代用指标，大覆盖范围的星载多光谱图像将当地信息转移到更大尺度），几乎没有任何详尽的研究说明所有传感器类型及其空间，在SP3中，我们希望缩小这一差距，因此我们的目标是探索具有异构传感器特性的传感器系统的巨大覆盖范围，包括无人机（厘米级极高空间分辨率、多光谱、热成像和激光雷达）、机载（高空间分辨率<5米、超光谱）和星载（中等空间分辨率10米至1公里、多光谱、超光谱、合成孔径雷达）。我们的目标是在RU的控制实验中对森林干扰的时间性（结构β复杂性增强（ESBC））进行分类，即如果，在何种条件下以及在何种程度上可以检测到各种传感器的干扰。我们的主要目标是确定来自遥感数据的参数，这些参数对干扰敏感，因此对森林结构敏感，对野外的分类记录敏感，并使用这些参数来绘制更大尺度的β和γ多样性。因此，我们的研究结果将弥合森林结构，生物多样性和功能，精细尺度近距离遥感（无人机类型，机载）和星载卫星遥感的现场测量的规模。因此，我们期望能够更好地了解以不同光谱、空间和时间分辨率对多传感器数据进行的联合分析是否以及如何有助于对生物多样性和生态系统功能进行实地评估。

Earth observation is a powerful means to derive spatially explicit information about the Earth’s surface at large scales and monitor its change over time. Various remote sensing applications related to forests exist. This is based on the sensitivity of active sensor systems (such as LIDAR and SAR) mainly to forest structure (horizontal & vertical), and of passive – predominantly optical – systems mainly to biochemical properties of the canopy (e.g. multi- or hyperspectral sensors). However, recent scientific work demonstrated the capability of very high spatial resolution passive optical sensors onboard UAVs to quantify forest features on the ground (below-canopy). Despite these new findings, there is no one-fits-all sensor in place. While there is a trend of increasingly combining different sensor systems sensitive to specific properties (e.g. small-scale LIDAR measurements as proxy for forest structure at local scale and wide-coverage spaceborne multispectral images to transfer local information to larger scales), there is hardly any exhaustive study that accounts for all sensor types and their spatial, spectral and temporal resolution characteristics.Within SP3 we want to close this gap, and hence our objective is to explore a huge coverage of sensor systems with heterogeneous sensor characteristics, including UAV (very high spatial resolution at cm-range, multispectral, thermal & LIDAR), airborne (high spatial resolution <5 m, hyperspectral) and spaceborne (medium spatial resolution 10 m to 1 km, multispectral, hyperspectral, SAR). We aim at categorizing the temporality of forest disturbances (Enhancement of Structural Beta Complexity (ESBC)) within the controlled experiments of the RU, i.e. if, under which conditions and how long disturbances are detectable for the various sensors to which extent. Our main goal is to identify parameters derived from the remote sensing data that are sensitive to the disturbances, hence forest structure, and to the taxa records in the field, and use those parameters to map β- and γ-diversity at larger scales. Consequently, our findings will bridge the scales of in-situ measurements of forest structure, biodiversity and functionality, fine-scale near-distance remote sensing (UAV type, airborne) and spaceborne satellite remote sensing. As a result, we expect improved understanding if and how the joint analyses of multi-sensor data at different spectral, spatial and temporal resolutions can contribute to in-situ assessments of biodiversity and ecosystem functions.