GeoTERM: Geospatial Toolkit for Enhanced River Management

河流系统的可持续管理涉及平衡多个目标。这些措施包括减轻洪水灾害及其对人民和重要资产构成的风险，同时通过支持健康的生物群落和加强生境多样性，促进良好的生态状况。在向河流提供大量粗沙的地区，管理往往涉及解决与河道内砾石逐渐堆积有关的问题。这种淤积会使河床水位升高，导致洪水能力下降，进而可能导致洪水发生的可能性增加，并降低与现有防御工事有关的保护标准。管理这种危险的一种方法是通过提取河床砾石来恢复洪水能力，纠正河流的排列，防止河岸侵蚀，减少灾难性河道变化的威胁。提取的河卵石也并非没有价值，是建筑行业骨料的重要来源。因此，靠近城市地区的砾石河床通常被认为是现成沉积物的理想矿山。因此，这种情况可以被视为一种潜在的双赢游戏。只要砾石开采与自然发生的上游沉积物供应相平衡，就可以设计一种适应性管理制度，在产生关键商业资源的同时保持洪水容量。然而，现在已经确定的是，不正确地估计这种平衡和过度提取砾石会降低河床，使河道坡度变陡，导致堤岸侵蚀加剧，并通过破坏现有的防洪措施而矛盾地降低防洪能力。此外，去除通常粗糙的河床表层砾石会改变河床沉积物的组成，产生大量的细沉积物，使无脊椎动物和鱼类的栖息地退化。因此，疏浚河流以提高防洪能力的计划，必须建立在谨慎的、科学的、循证的战略基础上，以计划、实施和审查干预措施。最近在萨默塞特水位发生的事件使这些计划得到了显著推广。传统上，沉积物管理计划是基于河流断面稀疏网络的数据。这为通过定期勘测监测地层水平趋势提供了基础。所得数据也可用于确定形态砂砾输运速率和估计泥沙供应的背景速率。最近的研究表明，基于断面数据的河流水位和砾石运输估计，有效地忽略了断面之间的河流形态，可能存在显著的偏差，导致用于驱动管理策略的关键数据产生2-3个数量级的不确定性。遥感技术的进步提供了一种解决方案，通过比较三维高程模型随时间的变化来估计航道变化，从而减轻了这种偏见。这些模式之间的差异提供了可靠的高程变化测量，可以综合起来评估区域趋势。从历史上看，获取密集地形数据以创建这些模型的高成本阻碍了它们用于常规监测。持续的发展，尤其是在摄影测量方法方面，最近大大降低了这些数据的成本，并消除了阻碍其采用的瓶颈。在这个项目中，我们将与来自英国和新西兰的国家和地方政府的一组利益相关者合作，开发一种软件工具，可以使用这些新的密集3D地形数据流支持常规渠道监测。由此产生的工具将促进简化的工作流程，可由机构和当局工作人员轻松实施，并用于在统计不确定性框架内呈现结果，该框架可解释底层地形数据中的错误。

Sustainable management of river systems involves balancing multiple objectives. These include alleviating flood hazards and the risks they pose to people and critical assets whilst promoting good ecological status by supporting healthy biological communities and enhancing habitat diversity. In regions with high rates of coarse sediment supply to rivers, management often involves addressing the issues associated with the progressive accumulation of gravel within channels. Such sedimentation can raise riverbeds levels, resulting in reduced flood capacity which in turn may result in an increased probability of flooding and a reduction in the standards of protection associated with existing defences. One approach to manage this hazard is through the extraction of riverbed gravels to restore flood capacity, correct river alignments, prevent bank erosion and reduce the threat of catastrophic course changes. The extracted river gravels are also not without value and represent an important source of aggregate for the construction industry. So much so, that gravel-bed rivers close to urban areas are often considered ideal mines of readily available sediment. This situation can, therefore, be presented as a potential win-win game. As long as gravel extraction is balanced against the naturally occurring upstream sediment supply, an adaptive management regime can be devised to maintain flood capacity whilst generating a key commercial resource. However, it is now well-established that estimating this balance incorrectly and over-extracting gravels can lower the riverbed, steepen the channel gradient, leading to enhanced bank erosion and paradoxically reduce flood protection by destabilizing existing flood control measures. Additionally, removal of the typically coarse surface layer of riverbed gravels can alter the bed sediment composition creating a flush of fine sediment that degrades invertebrate and fish habitat.Plans to dredge rivers to enhance flood capacity, so prominently popularized by the recent events in the Somerset Levels, must therefore be based on cautious, scientifically-informed and evidence-led strategies to plan, implement and review interventions. Traditionally, sediment management plans have been based on data from sparse networks of river cross-sections. These provide a basis for monitoring trends in bed levels through periodic resurveys. The resulting data can also be used to determine a morphological gravel transport rate and estimate the background rate of sediment supply. Recent research has shown that the river level and gravel transport estimates based on section data, which is effectively blind to the river morphology between sections, can incorporate significant bias giving rise of 2-3 order of magnitude uncertainties the key data used to drive management strategies.Advances in remote sensing offer a solution to alleviate this bias by estimating channel changes through the comparison of 3D elevation models through time. Differences between these models provide reliable measures of elevation change and can be integrated to assess regional trends. Historically, the high costs of acquiring dense topographic data to create these models has prohibited their use for routine monitoring. Continuing developments, most notably in photogrammetry methods have recently and dramatically reduced the cost of these data and removed a bottleneck preventing their adoption. In this project we will work with a group of stakeholders from national and local government in the UK and NZ to develop a software tool that can support routine channel monitoring using these new streams of dense 3D topographic data. The resulting tool will facilitate simplified workflows that can be easily implemented by agency and authority staff and used to present the results within a statistical uncertainty framework that accounts for errors in the underlying topographic data.