基于无人机遥感的古银叶树群落健康快速诊断

Rapid Diagnosis of Ancient Heritiera littoralis Community Health Using UAV Remote Sensing

采用无人机低空遥感与地面调查相结合的方法对邻海陆地、远海陆地和盐生沼泽生境的古银叶树群落健康进行评价,利用冠层高度、林窗特征、光合有效辐射截面比、归一化植被指数(NDVI)、氮素反射指数(NRI)、黄色波段指数(YI)以及森林健康指数(FHI)等遥感指标指征古银叶树群落的健康状况。空-地结合的调查结果表明:1)盐生沼泽生境的古树由于树龄高,其树洞大且数量多,生境内生物多样性最低,邻海陆地生境的生物多样性最高。2)盐生沼泽生境的冠层高度最低;林窗面积最大,数量最少,形状复杂度最低;光合有效辐射截面比最小。以上指标在邻海陆地和远海陆地间差异不明显。NDVI、NRI、YI以及FHI的数值均表现出盐生沼泽小于远海陆地和邻海陆地的趋势,而在远海陆地和邻海陆地间的差异较小。3)无人机遥感的评价结果与地面调查结果契合度较高,客观地反映了不同生境古银叶树的健康状态。基于无人机遥感的评价体系在针对具体植物群落修改完善后,可以作为一种快速、无损和定量化的古树群落健康诊断方法。

We evaluated the health of ancient Heritiera littoralis communities in landward terrestrial, seaward terrestrial, and halophytic marsh habitats using UAV remote sensing combined with ground surveys. Remote sensing indicators such as canopy height, forest window characteristics, photosynthetic effective radiation cross section ratio, Normalized Difiference Vegetation Index (NDVI), Nitrogen Reflection Index (NRI), Yellow band Index (YI)S and Forest Health Index (FHI) were used to indicate the health of ancient H. littoralis communities. The sky-ground combined survey indicated that: 1) Ancient H. littoralis tree holes were larger and more common in the halophytic marsh habitat due to their age. Moreover, the halophytic marsh habitat had the lowest biodiversity, with the highest biodiversity found in the seaward terrestrial habitat. 2) The lowe就 canopy height was found in the halophytic marsh habitat, with few large canopy gaps and the lowest shape complexity and photosynthetic effective radiation cross section ratio. The above indexes displayed little difference between the landward and seaward terrestrial habitats. NDVI, NRI, YI, and FHI values all tended to be lower in halophytic marshes than in landward and seaward terrestrial habitats, with no significant difference between the landward and seaward terrestrial habitats. 3) The UAV remote sensing survey results exhibited good correlation with the ground survey results, objectively reflecting the health of ancient H. littoralis communities in different habitats. After further improvements for specific communities, the evaluation system based on UAV remote sensing could be a rapid, nondestructive, and quantitative method for diagnosing ancient tree health.