Landslide Multi-Hazard Risk Assessment, Preparedness and Early Warning in South Asia: Integrating Meteorology, Landscape and Society

印度约12.6%的陆地容易发生山体滑坡，由于气候、地貌和地质原因，喜马拉雅和西高止地区尤其容易发生山体滑坡。降雨和地震是这些山体滑坡的主要诱因。糟糕的土地管理做法(例如，砍伐森林、砍伐和焚烧耕作、随意采矿和农业重耕细作)，再加上开发的增加和居住地点的恶劣，使这些地区的社区更容易受到山体滑坡的影响。山体滑坡对人民、商业、文化和遗产的影响可能是巨大和广泛的，包括死亡、农田和基础设施的损失以及对生态系统的破坏。为了在这些脆弱的社区中建立对滑坡的复原力(剪切的一个关键目标)，需要对人类与易发生滑坡的环境的相互作用进行彻底的评估，并改进对“物理”过程的了解。开发集成天气、景观和社会动力模型的方法是建立有效的水文控制滑坡预警系统(EWS)的基础。山体滑坡将通过建立在印度、英国和意大利现有科学研究的基础上，并使用跨学科的方法和观点来开发新的见解。由于滑坡发生的复杂环境条件和触发过程，其时空尺度的广度和变异性意味着滑坡在场地、坡度、流域和区域空间尺度以及小时尺度到十年尺度的时间尺度上都难以预测和管理。滑坡将通过研究了解触发滑坡的天气状况(以前在S亚洲没有做过)和降雨特征，以及可以增强滑坡易感性的地貌/地质控制因素来解决这个问题。我们还将利用开发和管理滑坡清单和影响数据库的广泛联盟经验，对历史滑坡发生的地点、时间和环境条件进行整理和分析。我们在滑坡方面应对的一个创新挑战是，斜坡和特定地点的EWS如何为国家滑坡EWS提供更广泛的信息，以及来自中期预报的早期预警信息如何补充和加强短期(日至周)预报方法。滑坡的另一个创新方面是通过整合从“人类”(即社交媒体)和物理传感器(由Amrita部署的遥感和预先存在的特定地点无线网络)收集的社会动态信息来提高EWS的有效性。山体滑坡将开发利用这些信息来源的方法，以补充现有的库存，并为决策者提供更多的电子战信息。我们的方案将与公共和私营部门的决策者、学者和非营利机构合作，以实现一个总体目标，即在印度的多灾害框架内促进更好的山体滑坡风险评估和早期预警，旨在提高复原力和减少损失。将与当地科学家、决策者和社区一起开发侧重于网络地图界面的工具和服务，以提高印度水文控制的山体滑坡的复原力，特别是利用两个试点研究区：喜马拉雅的大吉岭-东锡金和西高止的尼尔吉里斯。我们将通过评估如何将知识远程应用于阿富汗，确保将知识转让给其他脆弱社区。通过跨学科科学的进步和实践中的应用，该联盟的集体抱负是在多灾害框架内促进更好的滑坡风险评估和预警，并通过与社区合作，从区域空间尺度和日常到季节时间尺度，更好地准备应对以水文控制的山体滑坡和斜坡上的相关灾害。

About 12.6% of Indian land mass is prone to landslides, with the Himalaya and Western Ghats regions particularly prone due to climate, geomorphology & geology. Rainfall and earthquakes are the main triggers of these landslides. Poor land management practices (e.g., deforestation, slash & burn cultivation, haphazard mining and heavy tilling in agriculture), coupled with increased development and poor settlement location have increased vulnerability of communities in these areas to landslides. The impact of landslides on people, business, culture and heritage can be considerable and wide-ranging, including fatalities, loss of agricultural land and infrastructure, and damage to ecosystems. To build resilience to landslides in these vulnerable communities (a key aim of SHEAR), a root and branch evaluation of human interactions with landslide prone environments, and improved knowledge of the 'physical' processes is required. Developing approaches to integrate weather, landscape and social-dynamic models is fundamental to building an effective hydrologically-controlled landslide early warning system (EWS). LANDSLIP will develop new insights by building on existing scientific research in India, the UK and Italy and using interdisciplinary methodologies and perspectives. Due to complex environmental conditions and triggering processes that cause landslides, the extent and variability of spatial & temporal scales means that landslides are inherently difficult to forecast and manage at site, slope, catchment and regional spatial scales and hourly to decadal temporal scales. LANDSLIP will address this by doing research to understand weather regimes (previously not done in S Asia) and rainfall characteristics that trigger landslides and geomorphological/geological control factors that can enhance landslide susceptibility. Knowledge of where and when historic landslides have occurred and under what environmental conditions, will also be collated and analysed, drawing on extensive consortium experience of developing and managing landslide inventories and impact libraries.An innovative challenge we address in LANDSLIP is how slope and site specific EWS inform wider catchment to national landslide EWS and how early warning information from medium-range forecasts supplement and enhance short-term (day to a week) forecasting approaches. A further innovative aspect of LANDSLIP is improving EWS effectiveness through integrating social dynamics information gathered from both 'Human' (i.e. social media) and physical sensors (remote sensing and pre-existing site-specific wireless networks deployed by AMRITA). LANDSLIP will develop ways of utilising these sources of information to supplement existing inventories and enhance EW information for decision makers. Our programme will operate in partnership with decision makers, in public and private sectors, academics and non-for profit agencies to achieve an overarching aim of contributing to better landslide risk assessment and early warning, in a multi-hazard framework in India, aiming to increase resilience and reduce loss. Tools and services, focussed on a web map interface, will be developed in conjunction with local scientists, decision makers and communities to improve resilience to hydrologically-controlled landslides in India, specifically using two pilot study areas; Darjeeling-East Sikkim in the Himalaya and Nilgiris in the Western Ghats. We will ensure knowledge transfer to other vulnerable communities by assessing how they can be applied, remotely, in Afghanistan.Through advances in interdisciplinary science and application in practise, the collective ambition of this consortium is to contribute to better landslide risk assessment and early warning in a multi-hazard framework, and, by working with communities, better preparedness for hydrologically controlled landslides and related hazards on a slope to regional spatial scale and daily to seasonal temporal scale.