SBIR Phase I: Climate analysis at scale through mapping biomass, water restoration, and soil carbon sequestration

这项小企业创新研究（SBIR）第一阶段项目的更广泛影响/商业潜力是扩大基于自然的气候缓解。第一阶段项目将创建和管理综合数据集和模型，这些数据集和模型代表当前和拟议的未来生态系统健康状况，以森林地上生物量增长、土壤健康和碳固存以及通过恢复水文功能在干旱或其他退化环境中恢复地下水来衡量。通过其开源和面向社区的方法，该项目旨在为各组织提供宝贵的资源，以减轻因采掘土地使用做法而造成的生物多样性、土壤健康、森林覆盖和地下水的损失。这项技术提供的信息将对土地所有者、政府当局、研究人员和金融机构有用。该项目将利用遥感、云计算基础设施和机器学习方面的最新进展，预测最有效的陆地干预措施，以管理和加强水文循环、保护森林和加大再造林力度，并通过土壤有机碳封存建立土壤健康和弹性。影响最大的水干预措施将通过使用包括孔隙度、径流速率和质地在内的土壤模型改进的随机森林算法来确定。以森林为基础的保护和增加地上生物量的干预措施将以通过卫星图像监测的历史土地利用变化和干扰制度为基础，并加上通过恢复和改善森林管理实现的潜在固存率。除了按土地管理类型研究土壤固存率外，还将利用80年期间的历史土壤样本数据和相关的环境协变量来完成土壤健康和碳固存模型。这将使训练随机的基于森林的模型能够预测对提高或恢复土壤有机碳影响最大的土地利用干预措施。这些模型将组合在一起创建一个交互式的陆地查看器。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to scale nature-based climate mitigation. The Phase 1 project will create and manage comprehensive data sets and models representing the current and proposed future state of ecosystem health as measured by aboveground biomass growth in forests, the health and carbon sequestration in soils, and the restoration of groundwater in arid or otherwise degraded environments through the restoration of hydrological function. Through its open source and community-oriented approach, the project intends to be a valuable resource for organizations mitigating loss of biodiversity, soil health, forest cover, and groundwater due to extractive land use practices. The information provided by this technology will be useful for landowners, governmental authorities, researchers, and financial institutions.The project will utilize recent advances in remote sensing, cloud compute infrastructure, and machine learning to predict the highest leverage land-based interventions for managing and enhancing the hydrological cycle, protecting forests and increasing reforestation efforts, and building soil health and resiliency through soil organic carbon sequestration. The highest impact water interventions will be determined using random forest algorithms improved through the use of soil models that include porosity, rates of runoff, and texture. The forest-based interventions for conserving and increasing above-ground biomass will be based on historic land use change and disturbance regimes as monitored through satellite imagery and coupled with potential sequestration rates achieved through restoration and improved forest management. Soil health and carbon sequestration modeling will be accomplished using historic soil sample data over an 80 year period and the associated environmental covariates in addition to research into rates of soil sequestration by land management type. This will enable the training of a random forest-based model to predict the highest impact land use interventions to enhance or restore soil organic carbon. These models will be combined to create an interactive land viewer.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.