Climate change has already made significant alterations to various elements of the hydrologic cycle. One relatively less attended hydrologic impact of climate change is on the landscape Freeze-Thaw (FT), which largely affects surface and sub-surface hydrology, phenology, and land-atmospheric interactions, particularly in cold regions. Understanding the impacts of climate change on FT patterns, however, is not trivial due to sparse networks of in-situ measurements as well as limitations in current physically-based modeling schemes, aiming at continuous simulations of FT states. Here, we propose shifting the focus of FT modeling from continuous simulations in time and space, and move toward statistical representations of FT characteristics at larger temporal and spatial scales. We accordingly suggest using copulas to formally describe the impacts of temperature alterations on FT characteristics using conditional probabilities. To showcase the application of this framework in practice, we pair datasets of satellite-based FT with gridded temperature over Québec, Canada. The results show strong and rather unique dependencies between temperature and FT characteristics across different regions and/or timeframes. Our findings demonstrate copulas as effective tools to capture such dependencies and to reconstruct marginal FT characteristics. Through a set of impact assessments, it is shown that a similar change in temperature conditions can result in different regional responses in landscape FT. These responses are often nonsymmetric, meaning that the magnitude of change in FT conditions can be different under warming and cooling conditions. In addition, we highlight intensifications in FT responses to a similar magnitude of change in temperature under more recent years, which is linked to alterations in dependencies between temperature and FT. This study provides another line of evidence for complex responses of landscape FT to climate change.

气候变化已经对水文循环的各个要素产生了重大改变。气候变化的一个相对较少关注的水文影响是对景观冻融 (FT) 的影响，它在很大程度上影响地表和次表层水文、物候和陆地 - 大气相互作用，特别是在寒冷地区。然而，由于原位测量的稀疏网络以及当前基于物理的建模方案的局限性，旨在连续模拟 FT 状态，因此了解气候变化对 FT 模式的影响并非易事。在这里，我们建议将 FT 建模的重点从时间和空间上的连续模拟转移到更大的时间和空间尺度上的 FT 特征的统计表示。因此，我们建议使用 copula 使用条件概率来正式描述温度变化对 FT 特性的影响。为了展示该框架在实践中的应用，我们将基于卫星的 FT 数据集与加拿大魁北克的网格温度配对。结果表明，不同地区和/或时间范围内的温度和 FT 特性之间存在强烈且相当独特的相关性。我们的研究结果表明，copula 是捕获这种依赖关系和重建边缘 FT 特征的有效工具。通过一组影响评估，表明温度条件的类似变化会导致景观 FT 中的不同区域响应。这些反应通常是非对称的，这意味着 FT 条件的变化幅度在升温和降温条件下可能不同。此外，我们强调了近年来 FT 对类似温度变化幅度的反应加剧，这与温度和 FT 之间的依赖性变化有关。这项研究为景观 FT 对气候变化的复杂响应提供了另一条证据。