In northern climates, accurate simulation of thermal and hydrological budgets for farmlands overwinter is crucial for both an accurate prediction of spring flooding and the successful management of nutrient losses. As snow cover influences soil freezing dynamics, it has been hypothesized that reduced snow cover due to warmer winters might intensify soil freezing. The present study was designed to test this hypothesis. Drawing upon observed snow depth and soil temperature data collected from six research farms across Southern Canada over various time spans from 1989 to 2020, the root zone water quality model, integrated with the simultaneous heat and water model (RZ-SHAW), was calibrated and validated. The potential influence of warmer winter on shifts in soil frost dynamics was evaluated by estimating soil freezing dynamics for each farmland site under various air temperature scenarios using the RZ-SHAW model. Soil frozen depth in the eastern Canada sites increased with increasing air temperature in some years but decreased under the highest air temperature increases of 3.5°C. The monthly relationship between snow depth and soil frozen depth was determined through partial correlation analysis. Snow was most effective in alleviating soil freezing in the months of January and February, a period when snow cover depth was least affected by warming air temperatures. This study suggests that the hypothesis of increasing soil frozen depth under global warming-induced snow cover reduction holds true in conditions where soil energy lost through reduced snow cover outweighed the soil energy gained through warmer air temperature.

中文翻译：

---

积雪的长期模拟及其对加拿大南部农田季节性霜冻动态的潜在影响

在北方气候中，准确模拟农田越冬的热量和水文收支对于准确预测春季洪水和成功管理养分流失至关重要。由于积雪会影响土壤冻结动态，因此假设由于冬季温暖而减少的积雪可能会加剧土壤冻结。本研究旨在检验这一假设。利用从 1989 年到 2020 年不同时间跨度从加拿大南部的六个研究农场收集的观测雪深和土壤温度数据，与同时热和水模型 (RZ-SHAW) 相结合的根区水质模型进行了校准和验证。通过使用 RZ-SHAW 模型在各种气温情景下估计每个农田场地的土壤冻结动态，评估了暖冬对土壤霜冻动态变化的潜在影响。加拿大东部地区土壤冻结深度在某些年份随着气温升高而增加，但在最高气温升高 3.5°C 时降低。通过偏相关分析确定了积雪深度与土壤冻结深度的月度关系。1 月和 2 月，雪在缓解土壤冻结方面最有效，这段时间积雪深度受气温升高的影响最小。