Predictive ecology in a world with dynamic data and models

点击翻译按钮获取中文摘要

An overarching objective of many ecological research groups is to forecast ecosystems, species or processes. Indeed, for many researchers, large-scale ecological forecasts form the backbone of some of the most challenging ecological and conservation problems of today, such as building conservation strategies for wide-ranging species. One of the key challenges for large-scale studies, however, is that processes, and how their relative importance varies in space, are changing through time. Changing future conditions make it critical to understand key ecological mechanisms within a system, as empirical correlations will fail sooner without some mechanistic understanding. The long term objective is to build a new vision and reality for ecological simulation models. This involves building and using open systems based around R, building new -- and adapting our and other's existing -- ecological models that are open and interchangeable. Once these are done, we will evaluate emergent phenomena, such as stability, resilience, and feedbacks to inform land management at continental scales.The specific objectives here are 1) to build a new, large-scale, plant-functional-trait succession model, 2) test this in a wide array of systems, and 3) evaluate ecological consequences and trade-offs between vegetation management (e.g., carbon sequestration) and other management practices (e.g., caribou, fire, pest, and logging).This research will have 3 different levels of impact. First, the tools we will develop will be useful for the scientific community and its need to build new models. Second, the new models we will develop will be available for use in the new and open/interchangeable modeling and decision making community. Third, we will produce web apps that will allow the public to become involved in ecological models, allowing a transparency of decision making that has rarely been available. This will allow for a new level of outreach and connection to a much wider audience than previously envisioned for ecological models and modelers.