Belmont Forum Collaborative Research: Scenarios of Biodiversity and Ecosystem Service

需要对社会经济和全球环境趋势对生物多样性和生态系统服务的复杂相互作用进行创新研究，以帮助开发更多信息丰富的情景，以应对环境和人类发展挑战。为了克服这些挑战，需要结合自然-人类系统的方法和分析。这些方案提供了生物多样性和生态系统服务的改进情景，将土地利用、入侵物种、过度开发、生物多样性、环境变化和污染等直接和间接驱动因素的产出结合起来。由此产生的模型提供了一种最先进的方法，可以实现更准确的定量评估、更好的土地利用和更有效的生态系统服务。利用这一方法，本研究项目侧重于解决气候突变引起的土地利用变化及其对物种灭绝、范围转移和物候变化的影响。更好地了解这一领域至关重要，因为物种丧失会破坏食物网的稳定，从而对人类福祉产生负面影响，并对多种生态系统功能和相关服务产生意想不到的后果。因此，预测生物多样性对全球变化的响应已成为一个重要的研究领域。目前旨在预测未来生物多样性和生态系统服务的模型因忽视诸如生物相互作用和营养水平之间的联系（例如，捕食者-猎物关系）等输入而受到批评。本研究将使用最近创建的复杂的生物多样性建模代码，该代码集成了预测性生物地理学、地理学、生物统计学和营养网生态学，以得出全球变化对生物多样性、生态系统功能和生态系统服务提供的影响的情景。作为一个测试案例，该项目将结合来自利益相关者的信息和脊椎动物的统计数据，生成一套可能受气候变化影响的泛欧脊椎动物生物多样性模型。利益相关方的信息将被用于项目，因此可以制定/设置方法选择、情景和指标，从而使结果对保护规划和决策支持有用。这项工作的更广泛影响包括美国研究人员与六个欧洲（芬兰、英国、荷兰、德国、瑞士、法国和意大利）科学家之间的国际合作，以及为改善涉及整个欧洲大陆脊椎动物物种的生态系统服务的政策、决策和保护规划制定方案。在这个研究人员联盟中，每个国家资助自己的科学家和他们参与的项目。这项工作的其他更广泛的影响包括研讨会和暑期学校，分别向利益相关者团体提供信息，并培训/教育学生和博士后；为欧盟委员会编写报告；通过一个与社交媒体、在线讨论小组和论坛相连的项目网站进行公众宣传。在美国部分，将培养一名跨学科、国际科学和团队合作方面的博士后学者。该奖项支持参与由来自23个国家的26个资助机构组成的联盟通过贝尔蒙特论坛征集“生物多样性和生态系统服务情景”提案竞争性选择的项目的美国研究人员。该呼吁是一项多边倡议，旨在支持有助于开发情景、模型和决策支持工具的研究项目，以了解和解决我们星球面临的关键问题。竞赛的目标是改进和应用参与式情景方法，以提高研究的相关性和接受度，并解决影响驱动因素和政策干预建模方法方面的差距。它还开发和交流与模型相关的不确定性水平，以改善数据可访问性并填补知识空白。利用这一方法，本研究建立在新开发的广义联合属性模型（Generalized Joint Attribute Model）的基础上，该模型解决了预测群落变化的两大挑战：(1)在亲缘关系较远的分类类群中建立相互依赖物种响应模型的能力；(2)从基于样地的植物丰度到个体和样地水平的食草动物损害的尺度上将物种联系起来的能力。这些进展允许在考虑对环境的共同反应和物种之间的共同变异的同时，对整个群落中单个物种和物种群的分布进行建模。项目中使用的建模代码采用灵活的分层框架来模拟响应多种环境变化的物种群落。除了建模之外，该项目还将扩展现有建模平台的功能，以便解决捕食者-猎物和猎物竞争的相互作用。它还将允许结合基于“谁吃谁”网络中通过分层实现的关系的物种之间预期关系的先验信息。物种预测将包括对不确定性和对环境参数的反应的分析，以及物种对环境的共同反应和与其他物种的相互作用。通过将系统保护规划原则与多营养群落和生态系统服务的整体预测相结合，模型结果将用于确定优先保护的区域。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。

Innovative research on the complex interaction of socio-economic and global environmental trends on biodiversity and ecosystem services is needed to help develop more informative scenarios for addressing environmental and human development challenges. To overcome these challenges coupled natural-human systems approaches and analyses are needed. These provide improved scenarios of biodiversity and ecosystem services that couple the outputs of direct and indirect drivers such as land use, invasive species, overexploitation, biodiversity, environmental change, and pollution. The resulting models provide a methodological state-of-the art that results in more accurate quantitative assessments, better land use, and more effective ecosystem services. Employing this methodology, this research project focuses on addressing land-use changes due to abrupt climate change and their impact on species extinctions, range shifts, and phenological changes. Better knowledge in this area is critical because species loss destabilizes food webs which can negatively impact human well-being and have unexpected consequences for multiple ecosystem functions and associated services. Predicting the response of biodiversity to global change has thus become an important area of research. Present models intended to anticipate future biodiversity and ecosystem services have been criticized for ignoring inputs such as biotic interactions and links between trophic levels (e.g., predator-prey relationships). This research will use a recently created, sophisticated, biodiversity modeling code that integrates predictive biogeography, geography, biostatistics, and trophic web ecology to derive scenarios of the impact of global change on biodiversity, ecosystem functioning, and the provision of ecosystem services. As a test case, the program will be combined with information from stakeholders and statistics on vertebrates to generate a suite of likely models of pan-European vertebrate biodiversity as they might be influenced by climate change. Stakeholders information will be used to inform the project so methodological choices, scenarios, and indicators can be made/set in such a way that results are useful for conservation planning and decision-making support. Broader impacts of the work include international collaboration between US investigators and six European (Finnish, United Kingdom, Dutch, German, Swiss, French, and Italian) scientists and the generation of scenarios for improving policy, decision-making, and conservation planning of ecosystem services involving vertebrate species across the European continent. In this coalition of investigators, each country funds it own scientists and their part of the project. Additional broader impacts of the work include workshops and summer schools to inform stakeholder groups and train/education students and postdocs, respectively; the generation of reports for the European Commission; and public outreach through a project website linked to social media and online discussion groups and forums. For the US component of the work, there will be the training of a postdoctoral scholar in transdisciplinary, international science and teamwork.This award supports US researchers participating in a project competitively selected by a coalition of 26 funding agencies from 23 countries through the Belmont Forum call for proposals on "Scenarios of Biodiversity and Ecosystem Services". The call was a multilateral initiative designed to support research projects that contribute to the development of scenarios, models, and decision-support tools for understanding and solving critical issues facing our planet. The goal of the competition was to improve and apply participatory scenario methods to enhance research relevance and its acceptance and to address gaps in methods for modelling impact drivers and policy interventions. It was also to develop and communicate levels of uncertainty associated with the models, to improve data accessibility and fill gaps in knowledge. Using this methodology, this research builds on the newly developed Generalized Joint Attribute Model, a generative model that solves two big challenges for predicting community change: (1) the ability to model responses of interdependent species across distantly related taxonomic groups and (2) the ability to link species on scales from plot-based plant abundance to individual- and plot-level herbivore damage. These advances allow modeling of the distribution of individual species and species groups across entire communities, while considering shared responses to the environment and co-variation between species. The modeling code used in the project employs a flexible hierarchical framework to model species communities responding to multiple environmental changes. In addition to the modeling, the project will work to expand the capabilities of the present modeling platform so it can address predator-prey and competition-for-prey interactions. It will also allow incorporation of prior information on expected relationships between species based on relationships within the "who eats whom" network via hierarchical implementation. Species predictions will include an analysis of uncertainties and responses to environmental parameters as well as shared species responses to the environment and interactions with other species. Model results will be used to identify areas of conservation priority by combining principles of systematic conservation planning with ensemble projections of multi-trophic communities and ecosystem services.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.