A composite indicator for assessing habitat quality of riparian forests derived from Earth observation data
A composite indicator for assessing habitat quality of riparian forests derived from Earth observation data
复制标题
DOI：
10.1016/j.jag.2014.09.006
复制
发表时间：
2015-05
期刊：
影响因子：
--
通讯作者：
B. Riedler;L. Pernkopf;T. Strasser;S. Lang;Geoff Smith
中科院分区：
文献类型：
--
作者：
B. Riedler;L. Pernkopf;T. Strasser;S. Lang;Geoff Smith
河岸森林是珍贵、复杂的栖息地，可促进高度生物多样性，其中对栖息地质量的有效监测尤为重要。我们提出了一个综合指标，称为河岸森林综合指标：关注结构（RFI_S），用于评估栖息地质量和识别需要采取保护行动的“热点”区域。 RFI_S 由源自甚高分辨率 (VHR) 卫星图像和 LiDAR 数据的七个指标组成，并按斑块级别计算。这些指标评估河岸森林质量的四个重要属性：（1）树种组成，（2）垂直森林结构，（3）水平森林结构和（4）水情。对于 RFI_S 的聚合，应用了两种不同的加权方案：基于专家的加权和统计加权。具有高累积 RFI_S 值的森林斑块代表栖息地质量良好的斑块。这些斑块主要沿着水体发现，反映了水体对于结构复杂性、最佳水情和树种组成的重要性。对于栖息地质量低下的森林斑块，RFI_S 有助于设计适当的措施，通过其可分解为基础指标来改善栖息地质量状况。检验RFI_S稳健性的敏感性分析表明，地形粗糙度指标方差对综合指标的影响最强。最后，与现有的基于专家的保护状况地图的比较揭示了对研究地点的栖息地质量进行补充定量评估的潜力。因此，我们得出的结论是，RFI_S 具有很强的能力来支持可持续森林管理，并补充定期收集的实地数据。
Riparian forests are precious, complex habitats fostering high biodiversity where effective monitoring of habitat quality is particularly important. We present a composite indicator, referred to as Riparian Forest composite Indicator: focus on Structure (RFI_S), for the assessment of habitat quality and identification of ‘hot-spot’ areas where conservation actions need to be taken. The RFI_S is composed of seven indicators derived from very high resolution (VHR) satellite imagery and LiDAR data, calculated on patch level. These indicators assess four important attributes of riparian forest quality: (1) tree species composition, (2) vertical forest structure, (3) horizontal forest structure and (4) water regime. For the aggregation of the RFI_S, two different weighting schemes, expert-based and statistical weighting, are applied. Forest patches with high cumulative RFI_S values represent patches of good habitat quality. These patches are primarily found along water bodies, reflecting the importance of water bodies for the structural complexity, an optimum water regime and tree species composition. For forest patches of low habitat quality the RFI_S helps to design suitable measures to improve habitat quality status through its decomposability into the underlying indicators. A sensitivity analysis to test the robustness of the RFI_S shows that the indicator variance in terrain roughness has the strongest influence on the composite indicator. Finally, a comparison with an existing expert-based map on conservation status reveals the potential of a complementary quantitative assessment of habitat quality in the study site. We hence conclude that the RFI_S has a high capability to support sustainable forest management complementing regularly gathered in situ data.