EAGER: Bioforecasting: understanding and predicting species persistence in ecological communities under changing environments

生物多样性被定义为所有植物、动物和其他生物物种在特定地点和时间的存在。这些物种共享环境，并以复杂的方式相互作用。科学家们知道，生物多样性对于提供地球健康所必需的许多生态服务至关重要。这些服务的例子包括养分循环、水净化和土壤形成。因此，理解和预测生物多样性的变化（称为“生物预测”）对人类福祉非常重要。然而，由于更大更快的环境变化，知道一个物种是否会在一个给定的地方持续存在已经成为科学家面临的最大的生态挑战之一。主要问题是，几乎不可能知道环境条件将如何、何时、何地发生变化。只有事后看来，这些变化才是显而易见的。这个项目将创造一个解决问题的新方法。目的是估计一个物种在给定地点存活的几率，以及它的存在（或不存在）如何影响其他物种在那里存活的几率。这个项目还将帮助启动科学家和他将指导的学生的职业生涯，因为他们将共同制定一个新的理论。理解和预测物种持久性的主要困难在于了解控制生态群落动态的精确方程，以及关于初始条件、参数值、内在随机性的高度不确定性，更重要的是，变化的外部条件将如何影响所有这些动态。虽然生物预测在生态研究中已经是一项成熟的努力，但目前的生物预测方法需要大量的数据，而且它们的推广往往缺乏实验验证。因此，有人强调需要全新的框架。这些框架的发展涉及很大的风险和许多挑战。该项目将通过将统计力学的集合理论概念与结构稳定性的数学概念相结合，将其应用于种群动力学，以解决当前生物预测中的局限性，从而开辟新的领域。具体地说，不是旨在通过推断作用于它的主要条件来研究系统的未来行为，这个项目将提供一个可测试的方法来估计和解释基于与这种行为的可观察性相容的可能条件的分数的未来行为的概率。研究者将开发模型驱动和数据驱动的方法来估计这些概率，并用文献中已经编制的经验数据来验证它们。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。

Biodiversity is defined as the presence of all species of plants, animals and other living things at a given place and time. These species share their environment and interact in complex ways. Scientists know that biodiversity is critical for providing many ecological services necessary for the planet’s health. Examples of these services include nutrient cycling, water purification, and soil formation. Thus, understanding and predicting changes in biodiversity (called "bioforecasting") is important for human wellbeing. However, because of larger and faster environmental changes, knowing whether a species in a given place will persist has become one of the biggest ecological challenges that scientists face. The main problem is that it’s virtually impossible to know how, where, when and which environmental conditions will change. The changes are obvious only in hindsight. This project will create a new way to solve the problem. The goal is to estimate the chance that a species will persist in a given place and how its presence (or absence) can affect the chance of other species persisting there, too. This project will also help launch the careers of a scientist and the students he will mentor as they together craft a new theory.The main difficulty in understanding and predicting species persistence resides in knowing the exact equations governing the dynamics of ecological communities and the high uncertainty regarding initial conditions, parameter values, intrinsic randomness, and more importantly, how the changing external conditions will affect all of these dynamics. While bioforecasting is already a well-established endeavor in ecological research, current bioforecasting approaches demand extensive amounts of data and their generalization often lacks experimental validation. Thus, it has been emphasized that radically new frameworks are needed. Development of such frameworks involve large risks and many challenges. This project will break new ground by integrating concepts from the ensemble theory of statistical mechanics with the mathematical concepts of structural stability, applied to population dynamics to tackle current limitations in bioforecasting. Specifically, instead of aiming to study the future behavior of a system by inferring the main conditions acting upon it, this project will provide a testable methodology to estimate and interpret the probability of a future behavior based on the fraction of possible conditions compatible with the observability of such behavior. The investigator will develop model-driven and data-driven approaches to estimate such probabilities and validate them with empirical data already compiled in the literature.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.