Beyond Landsat: a comparison of four satellite sensors for detecting burn severity in ponderosa pine forests of the Gila Wilderness, NM, USA
Beyond Landsat: a comparison of four satellite sensors for detecting burn severity in ponderosa pine forests of the Gila Wilderness, NM, USA
复制标题
超越陆地卫星：四个卫星传感器的比较，用于检测美国新墨西哥州吉拉荒野黄松林的烧伤严重程度
DOI：
--
复制
发表时间：
2010
期刊：
影响因子：
--
通讯作者：
L. Vierling
中科院分区：
文献类型：
--
作者：
Z. Holden;P. Morgan;A. Smith;L. Vierling
需要远程测量烧伤严重程度的方法来评估大型偏远野火对生态和环境的影响。与Landsat项目相关的挑战突出了评估替代传感器以表征火灾后影响的必要性。我们比较了2003年吉拉荒野干湖火灾的55个复合燃烧指数野外图与四个卫星传感器不同空间和光谱分辨率的光谱指数之间的统计相关性。在频谱可行的情况下，使用差分增强植被指数（dEVI）、差分归一化植被指数（dNDVI）和差分归一化烧伤比率（dNBR）来评估烧伤严重程度。与Landsat Thematic Mapper数据得到的dNBR相比，Quickbird数据得到的dEVI和ASTER数据得到的dNBR显示出相似或略有改善的相关性（R分别为0.82、0.84和0.78）。modis衍生的相对较粗的ndv图像与地面数据的相关性较弱（R = 0.38）
Methods of remotely measuring burn severity are needed to evaluate the ecological and environmental impacts of large, remote wildland fires. The challenges that were associated with the Landsat program highlight the need to evaluate alternative sensors for characterising post-fire effects. We compared statistical correlations between 55 Composite Burn Index field plots and spectral indices from four satellite sensors varying in spatial and spectral resolution on the 2003 Dry Lakes Fire in the Gila Wilderness, NM. Where spectrally feasible, burn severity was evaluated using the differenced Enhanced Vegetation Index (dEVI), differenced Normalised Difference Vegetation Index (dNDVI) and the differenced Normalised Burn Ratio (dNBR). Both the dEVI derived from Quickbird and the dNBR derived from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) showed similar or slightly improved correlations over the dNBR derived from Landsat Thematic Mapper data (R 2 ¼0.82, 0.84, and 0.78 respectively). The relativelycoarseresolutionMODIS-derivedNDVIimagewasweaklycorrelatedwithgrounddata(R 2 ¼0.38).Ourresults