Improving representation of riparian vegetation shading in a regional stream temperature model using LiDAR data
Improving representation of riparian vegetation shading in a regional stream temperature model using LiDAR data
复制标题
DOI：
10.1016/j.scitotenv.2017.12.129
复制
发表时间：
2018-05-15
影响因子：
9.8
通讯作者：
Hannah, David M.
中科院分区：
文献类型：
--
作者：
Loicq, Pierre;Moatar, Florentina;Hannah, David M.
需要在集水区尺度上模拟河流温度，以了解水生群落如何适应当前和预计的气候变化。在中小河流中，河岸植被可以遮荫，大大降低最高水温。因此，重要的是，河流温度模型能够正确地描述这种河岸阴影的影响。在这项研究中，我们描述了一种空间显式方法的使用，该方法使用激光雷达衍生的数据来计算直接和漫射太阳辐射下的河岸遮阳。结果数据被用于T-NET一维河流温度模型，用于模拟2007年8月至2014年7月卢瓦尔河270公里的水温，卢瓦尔河是卢瓦尔河（法国）的间接支流。在沿卢瓦尔河分布的4个温度监测站进行了验证。2009年8月下旬，采用激光雷达方法表征的植被对最高日温度（T-max）的冷却效果从3.0摄氏度（上游）到1.3摄氏度（下游）不等。与其他两种计算密集度较低的河岸遮阳方法相比，我们基于激光雷达的方法在4月至9月期间将T-NET模型模拟的T-max偏差平均提高了0.62摄氏度。然而，在最上游的站点，遮阳程序之间的差异可达2摄氏度（月平均值）。T-max误差的标准差没有得到改善。只要提供准确的植被覆盖数据，使用河段平均参数在每小时时间尺度上计算河岸植被的影响，其结果与基于激光雷达的方法接近。因此，提高河岸植被数据的质量应该是一个优先事项，以提高区域尺度上河流温度模拟的准确性。(C) 2017 Elsevier B.V.版权所有
Modelling river temperature at the catchment scale is needed to understand how aquatic communities may adapt to current and projected climate change. In small and medium rivers, riparian vegetation can greatly reduce maximum water temperature by providing shade. It is thus important that river temperature models are able to correctly characterise the impact of this riparian shading. In this study, we describe the use of a spatially-explicit method using LiDAR-derived data for computing the riparian shading on direct and diffuse solar radiation. The resulting data are used in the T-NET one-dimensional stream temperature model to simulate water temperature from August 2007 to July 2014 for 270 km of the Loir River, an indirect tributary of the Loire River (France). Validation is achieved with 4 temperature monitoring stations spread along the Loir River. The vegetation characterised with the LiDAR approach provides a cooling effect on maximum daily temperature (T-max) ranging from 3.0 degrees C (upstream) to 1.3 degrees C (downstream) in late August 2009. Compared to two other riparian shading routines that are less computationally-intensive, the use of our LiDAR-based methodology improves the bias of T-max simulated by the T-NET model by 0.62 degrees C on average between April and September. However, difference between the shading routines reaches up to 2 degrees C (monthly average) at the upstream-most station. Standard deviation of errors on T-max is not improved. Computing the impact of riparian vegetation at the hourly timescale using reach-averaged parameters provides results close to the LiDAR-based approach, as long as it is supplied with accurate vegetation cover data. Improving the quality of riparian vegetation data should therefore be a priority to increase the accuracy of stream temperature modelling at the regional scale. (C) 2017 Elsevier B.V. All rights reserved.