Flood Mapping Tools for Disaster Preparedness and Emergency Response Using Satellite Data and Hydrodynamic Models: A Case Study of Bagmathi Basin, India
Flood Mapping Tools for Disaster Preparedness and Emergency Response Using Satellite Data and Hydrodynamic Models: A Case Study of Bagmathi Basin, India
复制标题
DOI：
10.1007/s40010-017-0461-7
复制
发表时间：
2017-12-01
影响因子：
0.9
通讯作者：
Yoshimoto, Shuhei
中科院分区：
文献类型：
--
作者：
Amarnath, Giriraj;Matheswaran, Karthikeyan;Yoshimoto, Shuhei
比哈尔邦北部是印度主要的洪水易发地区之一，由于季风降雨而经常发生洪水，影响了数千人的生命和生计。虽然不可能阻止极端洪水事件的发生，但灾害规划可以帮助减轻其不利影响。在洪水灾害发生后利用卫星观测监测洪水范围是快速应急响应流程的核心组成部分，除了向决策者和规划者提供近乎实时的洪水信息外，它还使应急救援队伍能够优先在关键地区开展工作，以拯救生命和保护健康。本研究的主要目的是展示数据密集程度较低但同样稳健的水动力模型的实用性，以结合不同尺度的免费遥感图像来开发洪水范围图。 MODIS TERRA 卫星数据用于绘制 2001 年至 2016 年整个比哈尔邦的洪水范围。两个水力模型，即 FLDPLN 和 RRI，应用于巴格马蒂盆地来评估我们的目标。这两个模型都具有不同的复杂性，但可以使用最少量的输入数据生成洪水范围模式。所提出的方法适用于绘制洪水范围图，以便在无法获得详细的水力模型和卫星数据集时近实时（h）提供输入信息。使用 Landsat-7 和 MODIS TERRA 导出的洪水范围对 FLDPLN 和 RRI 模型的洪水淹没范围进行了模型性能验证。结果显示模型预测的洪水范围与 Landsat-7 观测到的洪水范围之间具有可接受的空间一致性，表明这些工具在数据稀缺环境中洪水测绘应用中的实用性。
Northern Bihar is one of the major flood prone region in India affecting thousands of human lives and livelihoods during the recurrent floods occurring due to the monsoonal rains. While it is impossible to prevent the occurrence of extreme flood events, disaster planning can help in mitigating its detrimental effects. Monitoring flood extent using satellite observations just after the flood disasters is a core component of rapid emergency response process, which enables the emergency rescue teams to prioritize their efforts in critical areas to save lives and protect health, in addition to providing near real-time flooding information to the decision makers and planners. The main objective of this study is to demonstrate the utility of less data intensive, but equally robust hydrodynamic models to develop flood extent maps in conjunction with freely available remote sensing imageries at different scales. MODIS TERRA satellite data was used to map flood extent from 2001 to 2016 for entire Bihar. Two hydraulic models namely FLDPLN and RRI applied for the Bagmathi basin to evaluate our objectives. Both these models are of varying complexity but generate flood extent patterns with minimum amount of input data. The proposed approach is suited for mapping flood extents to provide an input information in near real time (h) when there is no availability to detailed hydraulic models and satellite datasets. Flood inundation extents from FLDPLN and RRI models were validated with Landsat-7 and MODIS TERRA derived flood extents for model performance. The results show acceptable spatial agreement between model predicted and Landsat-7 observed flood extents, denoting the utility of these tools for flood mapping application in data scarce environments.