Mitigating basis risk in weather index-based crop insurance: harnessing models and big data to enable climate-resilient agriculture in India

降低基于天气指数的农作物保险的基差风险：利用模型和大数据实现印度的气候适应型农业

• 批准号: NE/R014094/1
• 负责人: Timothy Foster
• 金额: $ 36.04万
• 依托单位: University of Manchester
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2018
• 资助国家: 英国
• 起止时间: 2018 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=NE%2FR014094%2F1
• 关键词: Mitigating; basis; risk; weather; index

Livelihoods of millions of smallholder farmers across the developing world are under threat from extreme weather events, such as droughts, floods, and heatwaves, with risks projected to increase significantly in future years due to climate change. Crop insurance protects farmers against financial risks posed by extreme weather events, and has been widely advocated as a tool to help farmer households to escape poverty traps and invest in climate-smart high-productivity agriculture. Yet, to date, the success and uptake of crop insurance schemes across the developing world has been extremely limited. Several reasons can be identified for problems in scaling crop insurance in developing countries. Traditional indemnity-based insurance schemes require time-consuming verification of actual losses experienced by individual farmers resulting in high transaction costs, claims disputes and delays that deter farmers from purchasing insurance. To counteract these issues, governments and insurers seek to develop more cost-effective and reliable tools to determine when, and at what level, insurance should payout to farmers if an extreme weather event occurs. Parametric insurance, for example weather index-based insurance, triggers payouts based on pre-established relationships between meteorological indices and crop yields, removing the need for expensive crop loss assessments. However, a major challenge for current weather index-based insurance is that payouts often are poorly correlated with farmers' actual yield losses, a problem known as 'basis risk', creating a major barrier to use of index insurance for climate risk mitigation.In this context, how can scientists contribute to the design of smarter index insurance products that meet the needs of farmers, insurers, and governments? The overall aim of this project is to improve the current poor performance of index-based crop insurance by using state-of-the-art environmental modelling and big datasets to reduce basis risk and better protect farmers against weather risks. Our proposed research will develop novel weather index-based insurance contracts that reliably and accurately predict weather-related crop yield losses by combining crop growth modelling, satellite and smartphone imagery of crop growth status, and high-resolution gridded estimates of spatial weather variability. Importantly, our work will produce novel tools and approaches that address two stated needs of the index insurance sector: (i) to reduce temporal basis risk by designing weather index triggers that reflect accurately how yield sensitivity to extreme events varies during the growing season, and (ii) to minimise spatial basis risk by exploiting datasets that capture spatial heterogeneity in weather conditions, crop development, field conditions and management practices. Working in collaboration with HDFC ERGO General Insurance, a major provider of weather index-based insurance for smallholder farmers across India, we will apply these approaches to design and test new weather index-based insurance products to protect farmers in the major agricultural states of Punjab and Haryana - the breadbasket of India - against combined production risks from extreme temperature and heavy rainfall events. Leveraging unique field data collected through the recent IFPRI-HDFC Picture-Based Crop Insurance (PBI) Project, we will conduct agro-economic impact evaluations to quantify reductions in basis risk, increases in farmer welfare and demand for insurance from our new contracts relative to both current index insurance products and government area-yield insurance schemes. Our work will contribute directly to improvements in the quality of index insurance for farmers in India, and, more broadly, will provide the scientific foundations for weather index-based insurance to more effectively support climate-smart smallholder agriculture across the developing world.

发展中国家的数百万农民农民的生计受到极端天气事件的威胁，例如干旱，洪水和热浪，由于气候变化，预计未来几年的风险将大大增加。农作物保险保护农民免受极端天气事件带来的财务风险，并被广泛提倡作为一种工具，以帮助农民家庭逃避贫困陷阱并投资于气候智能的高生产力农业。然而，迄今为止，整个发展中国家的农作物保险计划的成功和采用非常有限。可以确定有几个原因的原因是发展中国家的农作物保险问题。传统的基于赔偿的保险计划需要对个人农民造成的实际损失进行耗时的验证，导致交易成本高，索赔纠纷和延误阻止农民购买保险。为了解决这些问题，政府和保险公司试图开发更具成本效益和可靠的工具，以确定何时以及在何种级别上，如果发生极端天气事件，应向农民支付保险。参数保险，例如基于天气指数的保险，基于气象指数和农作物产量之间的预先建立关系的触发支出，从而消除了对昂贵的农作物损失评估的需求。但是，当前基于天气指数的保险的主要挑战是，支出通常与农民的实际收益率损失（一种称为“基础风险”的问题）相关，这为缓解气候风险的索引保险造成了一个主要的障碍。在这种情况下，科学家如何为满足农民，保险公司，保险公司和政府的智能指数保险的设计贡献，并有助于解决气候风险。该项目的总体目的是通过使用最先进的环境建模和大数据集来降低基本风险，并更好地保护农民免受天气风险，从而提高基于指数的作物保险的绩效不佳。我们提出的研究将开发新的基于天气指数的保险合同，这些合同可靠，准确地通过结合作物生长建模，卫星和智能手机图像的作物生长状态以及空间天气可变性的高分辨率网格估算，从而可靠，准确地预测与天气相关的作物产量损失。重要的是，我们的工作将产生新颖的工具和方法来满足指数保险行业的两个陈述需求：（i）通过设计天气指数触发的天气指数来降低时间基础风险，这些天气指数触发对极端事件的产生敏感性在生长季节期间如何产生敏感性，（ii）通过利用空间的练习来捕获空间依据，以捕获空间基础的风险，从而捕获空间练习的情况下，在天气中，情况下，培训习惯和培训习惯。与HDFC Ergo General Insurance合作，这是印度各地小农户的主要提供商保险的主要提供商，我们将把这些方法应用于设计和测试基于天气指数的新保险产品，以保护旁遮普邦和哈里亚纳邦主要农业国家的农民 - 印度面包板 - 印度面包板 - 针对极端温度和重量降雨的共同生产风险。利用最近基于IFPRI-HDFC图片的农作物保险（PBI）项目收集的独特现场数据，我们将进行农业经济影响评估，以量化基本风险中的减少，农民福利的增加以及对当前INDEX保险产品和政府政府领域收益保险计划的新合同的增加和需求。我们的工作将直接有助于改善印度农民的指数保险质量，更广泛地，将为基于天气指数的保险提供科学基础，以更有效地支持整个发展中国家的气候智能小型农业。

项目成果

Weather dataset choice introduces uncertainty to estimates of crop yield responses to climate variability and change

• DOI: 10.1088/1748-9326/ab5ebb
• 发表时间: 2019-12
• 期刊: Environmental Research Letters
• 影响因子: 6.7
• 作者: Ben Parkes;Thomas P. Higginbottom;K. Hufkens;Francisco Ceballos;B. Kramer;T. Foster

Identifying links between monsoon variability and rice production in India through machine learning.

• DOI: 10.1038/s41598-023-27752-8
• 发表时间: 2023-02-10
• 期刊: Scientific reports
• 影响因子: 4.6

Improving the Performance of Index Insurance Using Crop Models and Phenological Monitoring

使用作物模型和物候监测提高指数保险的绩效

• DOI: 10.3390/rs13050924
• 发表时间: 2021
• 期刊: Remote Sensing
• 影响因子: 5
• 作者: Afshar M

Improving performance of index insurance using crop models and phenological monitoring

使用作物模型和物候监测提高指数保险的绩效

• 发表时间: 2020
• 作者: Afshar MH

Monitoring crop phenology using a smartphone based near-surface remote sensing approach

• DOI: 10.1016/j.agrformet.2018.11.002
• 发表时间: 2019-02-15
• 期刊: AGRICULTURAL AND FOREST METEOROLOGY
• 影响因子: 6.2
• 作者: Hufkens, Koen;Melaas, Eli K.;Kramer, Berber
• 通讯作者: Kramer, Berber