Lidar remote sensing of biophysical properties of tolerant northern hardwood forests
Lidar remote sensing of biophysical properties of tolerant northern hardwood forests
复制标题
DOI：
10.5589/m03-025
复制
发表时间：
2003-10-01
影响因子：
2.6
通讯作者：
Green, J
中科院分区：
文献类型：
--
作者：
Lim, K;Treitz, P;Green, J
以前利用飞行时间激光雷达数据进行的森林研究主要侧重于以针叶树为主要树种的森林生态系统。在这项研究中，利用小足迹飞行时间激光雷达数据来估计以成熟糖枫(Acer charcharum Marsh.)为主的耐受阔叶林的生物物理性质。和苏尔街附近土耳其湖流域(TLW)的黄桦(Betula Reaghanisis Britton)。安大略省马里市。2000年7月头两周收集了49块圆形样地的地面参考数据，每块样地的面积为0.04公顷(或400米(2))。激光雷达数据是在2000年8月24日使用Optech Altm 1225(Optech公司，多伦多，安大略省)获取的。每个样地有10个生物物理指标：(1)最大树高(h(Max))，(2)洛雷平均树高(h(Lorey))，(3)平均胸径(DBH)，(4)总断面积(BA)，(5)郁闭度(CO%)，(6)叶面积指数(LAI)，(7)椭圆形郁闭度(CC)，(8)地上总生物量(BIO)，(9)总材积(VOL)，(10)树干密度(SD)。同样，对于每个小区，得出了三个激光高度度量：(1)最大激光高度(Lh(Max))，(2)平均激光高度(Lh(Mean))，以及(3)根据应用于强度返回值(Lh(IR))的阈值过滤的激光雷达回波计算的平均激光高度。结果表明，对于每个具有给定林分结构的森林，存在一个或多个从激光雷达数据获得的激光高度测量，这些测量能够提供对各种生物物理性质的估计。对h(Max)(r(2)=0.79)和h(Lorey)(r(2)=0.87)的估计以1h(Max)最好，对BA(r(2)=0.85)、BIO(r(2)=0.85)和Vol(r(2)=0.87)的估计以1h(IR)最好；对CC(r(2)=0.89)、胸径(r(2)=0.63)、CO%(r(2)=0.76)、LAI(r(2)=0.80)和SD(r(2)=0.86)的估计值最好。这些结果说明了激光高度测量用于估计(I)小区高度和树干密度，(Ii)地上生物量和体积，以及(Iii)与冠层有关的测量的潜力。
Previous forest research using time-of-flight lidar data has primarily focused on forest ecosystems with conifers as the predominant tree type. In this study, small-footprint time-of-flight lidar data were used to estimate biophysical properties of tolerant hardwood forests composed predominantly of mature sugar maple (Acer saccharum Marsh.) and yellow birch (Betula alleghaniensis Britton) in the Turkey Lakes Watershed (TLW) near Sault Ste. Marie, Ontario. Ground reference data were collected during the first two weeks of July 2000 for 49 circular sample plots, each 0.04 ha (or 400 m(2)) in area. Lidar data were acquired on 24 August 2000 using an Optech ALTM 1225 (Optech Incorporated, Toronto, Ont.). Ten biophysical forest metrics were derived for each plot: (1) maximum tree height (h(max)), (2) Lorey's mean tree height (h(Lorey)), (3) mean diameter at breast height (DBH), (4) total basal area (BA), (5) percent canopy openness (CO%), (6) leaf area index (LAI), (7) ellipsoidal crown closure (CC), (8) total aboveground biomass (BIO), (9) total wood volume (VOL), and (10) stem density (SD). Likewise, three laser height metrics were derived for each plot: (1) maximum laser height (Lh(max)), (2) mean laser height (Lh(mean)), and (3) mean laser height calculated from lidar returns filtered based on a threshold applied to the intensity return values (Lh(IR)). The results demonstrate that for each forest with a given stand structure, there exists one or more laser height metrics derived from lidar data that are capable of providing an estimate of various biophysical properties. Lh(max) was the best estimator of h(max) (r(2) = 0.79) and h(Lorey) (r(2) = 0.87); Lh(IR) was the best estimator of BA (r(2) = 0.85), BIO (r(2) = 0.85), and VOL (r(2) = 0.87); and Lh(mean) was the best estimator of CC (r(2) = 0.89), DBH (r(2) = 0.63), CO% (r(2) = 0.76), LAI (r(2) = 0.80), and SD (r(2) = 0.86). The results illustrate the potential for laser height metrics to estimate (i) plot heights and stem densities, (ii) aboveground biomass and volume, and (iii) canopy-related measures.