Advances in continental domain hydrology: Improving simulations of cold region hydrological processes over North America

对水循环的陆地部分进行建模仍然是一项关键的科学挑战。虽然全球水文模型在过去的几十年里确实有所改进，但这样的大区域模型仍然难以在数据稀疏的环境中表示复杂的寒区水文过程。当可用于描述景观特征(地形、植被、土壤、地质)的数据有限，并且可用于评估过程算法的保真度的数据有限时，模拟冰川、季节性雪、永久冻土、湖泊和湿地以及河流的动态以及它们的生物地球化学循环尤其困难。这些挑战在加拿大具有一定的紧迫性，因为加拿大的变暖幅度远远大于全球平均水平，寒冷地区水文过程的变化也是显而易见的。这项研究的长期目标是大幅提高模拟大陆区域水文过程的能力。我将通过实施系统的模型分析和改进方法来实现这一长期研究战略，这些方法既利用多种假设建模工具，也利用诊断模型评估方法。我有三个短期目标：目标1：改进寒冷地区主要水文过程的机械模型模拟。为了实现这一目标，我将改进机械性多假设总和模型(统一多个模拟备选方案的结构)中对寒冷地区水文过程的算法描述。目标2：基于过程的模型评估和诊断的高级能力。为实现这一目标，我将开发一个多管齐下的模型评价和基准制度，其中包括：(A)土地模型的功能单元测试，例如理想化测试用例的解析解，以确保模型方程得到正确实施；(B)多尺度和多变量模型评价的诊断工具，扩大国际土地模型基准项目的能力；(C)正式的基准方法，以评估模型评价数据中的信息内容。目的3：评估不同复杂性的水文过程的替代参数化的价值。在前两项任务的基础上，我将完成一组结构化的模型实验，以评估水文过程替代参数化的价值。我将系统地探索过程表示的复杂性(过程复杂性)和空间离散化的复杂性(空间复杂性)，并评估替代参数化在无数空间尺度上表示主导水文过程的能力。这项研究将有利于一系列与社会相关的应用，即水安全评估和水文预测，有助于提高社会对气候多变和变化的适应能力。

Modelling the terrestrial component of the water cycle remains a critical scientific challenge. While global hydrological models have certainly improved over the past several decades, such large-domain models still struggle with representing complex cold region hydrological processes in data sparse environments. It is especially difficult to simulate the dynamics of glaciers, seasonal snow, permafrost, lakes and wetlands, and rivers, along with their biogeochemical cycles, when there is limited data available to prescribe landscape characteristics (topography, vegetation, soils, geology), and there is limited data available to evaluate the fidelity of process algorithms. These challenges have some urgency in Canada where warming is much larger than the global average and the changes in cold region hydrological processes are palpable. The long-term goal of this research is to substantially advance capabilities to simulate hydrological processes over continental domains. I will accomplish this long-term research strategy by implementing systematic approaches to model analysis and improvement that take advantage of both multiple hypothesis modeling tools as well as diagnostic model evaluation methods. I have three short-term objectives: Objective 1: Improve mechanistic model simulations of dominant hydrological processes in cold regions. To achieve this objective, I will improve algorithmic descriptions of cold region hydrological processes in the mechanistic multi-hypothesis SUMMA model (the Structure for Unifying Multiple Modeling Alternatives). Objective 2: Advance capabilities for process-based model evaluation and diagnostics. To achieve this objective, I will develop a multi-pronged model evaluation and benchmarking system, including (a) functional unit tests for land models, e.g., analytical solutions for idealized test cases, to ensure that the model equations are implemented correctly; (b) diagnostic tools for multiscale and multivariate model evaluation, extending capabilities in the international Land Model Benchmarking project (iLAMB); and (c) formal benchmarking methods to evaluate the information content in the model evaluation data. Objective 3: Evaluate the value of alternative parameterizations of hydrological processes of varying complexity. Building on the first two tasks, I will complete a set of structured model experiments to evaluate the value of alternative parameterizations of hydrological processes. I will systematically explore the complexity in process representation (process complexity) and the complexity in spatial discretization (spatial complexity), and evaluate how well alternative parameterizations represent dominant hydrologic processes across a myriad of spatial scales. This research will provide benefits for a range of societally-relevant applications, namely water security assessments and hydrological predictions, which can help increase societal resilience to climate variability and change.