Peat-MOSS (Peatland Mapping and Observation with Satellite Sensors)

Peat-MOSS（利用卫星传感器进行泥炭地测绘和观测）

• 批准号: 10051596
• 金额: $ 9.49万
• 依托单位: SYLVERA LTD
• 依托单位国家: 英国
• 项目类别: Small Business Research Initiative
• 财政年份: 2022
• 资助国家: 英国
• 起止时间: 2022 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=10051596
• 关键词: Peat; MOSS; Peatland; Mapping; Observation

Peatlands are among the most secure sinks for greenhouse gases, storing 25% of global soil carbon, yet their degradation is responsible for 2-5% of all anthropogenic emissions\[Q1ref-1\]. Understanding the large potential for both storage and emissions, there has been an exponential increase in restoration projects aimed at reducing emissions and enhancing carbon storage in peatlands. However, there is currently no accurate and scalable method for measurement, reporting, and verification (MRV) of peatland restoration projects. At present, the impact of peatland restoration for greenhouse gas reduction (GGR) is simply assumed from a change in peatland classification (e.g., from "drained" to "restored"). Additionally, finding proper sites for further deployment of GGR projects is hindered by the limited knowledge of global peat extent. For example, in tropical regions the total un-mapped peatland area may be three times larger, and four times more voluminous than currently mapped\[Q1ref-2\]. Digital mapping and monitoring of peatlands based on satellite earth observation (EO) data has the potential to scale to meet the global need for accurate MRV and mapping of GGR projects, yet has been limited in three key areas:* Heavy reliance on basic statistical techniques.* Lacking quantification of uncertainty.* Lacking consistency in resolution and data inputs.These limitations have restricted the ability to monitor the effectiveness of peatland projects and identify/prioritise locations for future GGR projects.Sylvera will conquer the limitations of digital peatland mapping and monitoring by combining its expertise in machine learning (ML) and EO data with the research of the UK Centre for Ecology and Hydrology (UKCEH) - leaders in peatland GGR demonstration. Sylvera will lead the next-generation of digital peatland measurements to provide the scalable MRV solution that is desperately needed to bring integrity and confidence to GGR projects. Peat-MOSS will respond by developing an ML-driven method, combining its industry-leading deep-learning models based on EO data with the extensive research from UKCEH. Peat-MOSS will develop maps of peat extent and additional indicators of peat degradation that can be used to determine whether peatlands are emitting or sequestering CO2\. The use cases of Peat-MOSS maps are 1) MRV of peat GGR projects and 2) identification and prioritisation of peatland GGR project sites. Sylvera will leverage EO data and ML technology to provide digital GGR MRV. This MRV solution will ensure the effectiveness, security, and permanence of GGR projects by providing consistent and scalable monitoring of peatlands.

泥炭地是最安全的温室气体汇之一，储存了全球土壤碳的2-5%，但其退化占所有人为排放量的2-5%[Q1 ref-1]。由于认识到储存和排放的巨大潜力，旨在减少排放和提高泥炭地碳储存的恢复项目呈指数级增长。然而，目前还没有一个准确和可扩展的方法来测量，报告和验证（MRV）的泥炭地恢复项目。目前，泥炭地恢复对温室气体减排（GGR）的影响只是从泥炭地分类的变化（例如，从“排水”到“恢复”）。此外，由于对全球泥炭范围的了解有限，为GGR项目的进一步部署寻找合适的地点也受到阻碍。例如，在热带地区，未绘制地图的泥炭地总面积可能比目前绘制的泥炭地大三倍，体积大四倍[Q1 ref-2]。基于卫星地球观测（EO）数据的泥炭地数字制图和监测有可能扩大规模，以满足全球对精确MRV和GGR项目制图的需求，但在三个关键领域受到限制：* 严重依赖基本统计技术。缺乏对不确定性的量化。解决方案和数据输入缺乏一致性。这些局限性限制了监测泥炭地项目的有效性和确定未来GGR项目位置的能力。Sylvera将通过将其在机器学习（ML）和EO数据方面的专业知识与英国生态和水文中心（UKCEH）的研究相结合，克服数字泥炭地测绘和监测的局限性。UKCEH是泥炭地GGR示范的领导者。Sylvera将领导下一代数字泥炭地测量，提供可扩展的MRV解决方案，这是GGR项目迫切需要的完整性和信心。Peat-MOSS将通过开发ML驱动的方法做出回应，将其基于EO数据的行业领先的深度学习模型与UKCEH的广泛研究相结合。泥炭-MOSS将绘制泥炭范围图和泥炭退化的其他指标，这些指标可用于确定泥炭地是否正在排放或吸收CO2。泥炭-MOSS地图的用例是1）泥炭GGR项目的MRV和2）泥炭地GGR项目地点的识别和优先级排序。Sylvera将利用EO数据和ML技术提供数字GGR MRV。该MRV解决方案将通过提供对泥炭地的一致性和可扩展性监控，确保GGR项目的有效性、安全性和持久性。