Optimizing sampling design, data integration, and methods for understanding population dynamics and predicting ecological changes

我的长期目标是评估，开发和应用分析方法来量化描述人口和生态系统动态的生物，非生物和人为驱动因素之间相互作用的生态过程，特别是在北方地区。考虑到当前剧烈的生态变化，迫切需要提高我们预测生态系统动态的能力。开发可靠的预测工具对于更好地预测气候对生态系统的影响是必不可少的。我的短期研究目标是：1)改进采样设计和方法，以便更好地整合多个数据源，以模拟人口和生态系统动态的驱动因素；2)培养新的方法，将土著知识和科学知识结合起来，以增强我们对北方生态变化的理解；3)开发模型，以产生近期（季节性/年度）迭代预测，以更好地预测人口和生态系统的变化，并实时适应管理行动。我的程序是嵌入在近期迭代预测（NTIF）框架。NTIF使用综合了不同来源和时间尺度（日、季、年）数据的模型来描述生态系统的响应。然后，它产生每日至年度规模的预测，并在获得新数据时进行更新。在几项正在进行的长期研究的基础上，包括一些已经收集了关于状态变量和驱动因素的大量数据的研究，我的小组将在达到有用的生态预测所需的所有步骤中推进知识。我们将评估结合新的替代方法（相机陷阱、头发DNA个体识别、公民科学采样）的研究设计如何减少估计的不确定性，并提高大时空尺度下数据收集的效率。建立面向过程的模型，结合多种数据来源，揭示驱动生态变化的机制。例如，我们将测试气候或人为过程是否调节黑熊的种群动态。我们将开发强大的分析方法来模拟定性数据，使我们能够利用本土知识来验证气候诱导机制同步北极关键物种种群动态的假设。我们将开发NTIF模型并评估其用于预测未来生态系统状态的局限性。这些模型将研究气候对冬季蜱虫（quemassbec）和落叶蛾（挪威）向北传播的影响，以及它们各自对驼鹿种群和白桦林动态（死亡率、再生）的影响。我的项目将培养创新的HQP，他们将加速科学学习，从目前对测量影响的关注转向一种自适应的方法，在这种方法中，监测数据被迭代地用于完善假设，以更好地理解机制。这些方法将改进近期生态变化的预测，以更好地预测影响和不希望的未来状态。

My long-term objective is to evaluate, develop and apply analytical methods to quantify ecological processes describing the interplay among biotic, abiotic, and anthropogenic drivers of population and ecosystem dynamics, particularly in northern regions. Given the current drastic ecological changes, improvements in our ability to predict the dynamics of ecological systems are urgently needed. The development of reliable predictive tools is indispensable to better anticipate climate-induced impacts on ecosystems. My short-term research objectives seek to 1) improve sampling designs and methods for better integrating multiple data sources to model drivers of population and ecosystem dynamics, 2) foster new methods to combine indigenous and scientific knowledge to enhance our understanding of ecological changes in the North, and 3) develop models to produce near-term (seasonal/annual) iterative forecasts to better anticipate population and ecosystem changes and adapt management actions in real time. My program is embedded within the Near-Term Iterative Forecasting (NTIF) framework. NTIF uses models integrating data from diverse sources and temporal scales (day, season, year) to describe ecosystem responses. It then produces daily to annual scale forecasts that are updated as new data become available. Building upon several ongoing long-term studies, including some that have already collected extensive data on state variables and drivers, my group will advance knowledge at all steps required to reach useful ecological forecasts. We will assess how study designs combining new alternative methods (camera-traps, individual identification from hair DNA, Citizen Science sampling) can reduce the uncertainty of estimates and improve the efficiency of data collection at large spatiotemporal scales. We will develop process-oriented models, combining multiple data sources to unravel which mechanisms drive ecological changes. For instance, we will test whether climatic or anthropogenic processes regulate black bear population dynamics. We will develop robust analytical methods to model qualitative data, allowing us to use indigenous knowledge to test the hypothesis that climate-induced mechanisms synchronize population dynamics among key Arctic species. We will develop NTIF models and evaluate their limitations for predicting future ecosystem states. These models will examine climate impacts on the northward spread of winter ticks (Québec) and defoliating moths (Norway), and their respective consequences on moose populations and birch forests dynamics (mortality, regeneration). My program will train innovative HQP who will accelerate scientific learning by shifting from the current focus on measuring impacts towards an adaptive approach where monitoring data is used iteratively to refine hypotheses to better understand mechanisms. These methods will improve predictions of near-term ecological changes to better anticipate impacts and undesirable future states.