高级检索

    基于多层融雪模型的融雪期积雪深度模拟

    Simulation of snow depth during melting period based on multi-layer snowmelt model

    • 摘要: 目的为定量描述东北黑土区融雪期积雪消融过程。方法基于能量平衡原理建立多层融雪模型,对梅河口市吉兴小流域2022年积雪消融过程的积雪物理性质、积雪内部能量传递与压实以及积雪深度进行模拟分析与验证。结果模型成功模拟雪面温度,与空气温度的变化趋势较为一致,其中多层融雪模型雪面温度的平均绝对误差(MAE)为0.63 ℃,单层能量平衡模型计算结果的 MAE 为3.88 ℃。雪密度的模拟值为180~274 kg/m³,平均231 kg/m³,实测值为181~280 kg/m³,平均239 kg/m³,模拟的Nash为0.73,MAE为7.66 kg/m³,模型模拟结果与实测值吻合度较高。雪层压力随着积雪深度增加而增大,温度、密度与压力相互作用,共同影响积雪内部结构与深度变化。在积雪深度模拟上,2017年模拟精度偏低,经敏感性分析与参数修正后,2022年Nash由0.87提高至0.97,MAE由0.98 cm降至−0.09 cm,2017年与2018年Nash分别为0.96和0.97,MAE分别为1.01 cm和1.04 cm,修正后适用于不同年份积雪环境,能在更长时间序列、更多样积雪条件下应用。结论研究结果验证了模型的有效性。该模型不仅为融雪期积雪深度的预测提供可靠理论依据,也为其他地区的积雪消融研究提供参考价值

       

      Abstract: ObjectiveThe objective of this study was to quantitatively describe the snow ablation process during the snowmelt period in the Northeast Black Soil Region. Methods To this end, a multilayer snowmelt model was established based on the principle of energy balance. This model was used to simulate, analyze, and validate the physical properties of snow, energy transfer and compaction within the snow, and snow depth during the snowmelt process of the 2022 snowmelt in the Jixing subwatershed of Meihekou city. Results The model accurately simulated snow surface temperature, and the variation trend was generally consistent with that of air temperature. The mean absolute error (MAE) of snow surface temperature was 0.63 °C for the multilayer snowmelt model and 3.88 °C for the single-layer energy balance model. Simulated snow density ranged from 180 to 274 kg/m³ with a mean of 231 kg/m³, while observed values ranged from 181 to 280 kg/m³ with a mean of 239 kg/m³. The Nash–Sutcliffe efficiency coefficient was 0.73 and the MAE was 7.66 kg/m³, indicating good agreement between simulations and observations. Snow layer pressure increased with snow depth, and the interactions among temperature, density, and pressure jointly influenced the internal structure and depth evolution of the snowpack. In snow depth simulations, model performance in 2017 was relatively low. After sensitivity analysis and parameter calibration, the Nash coefficient for 2022 increased from 0.87 to 0.97 and the MAE decreased from 0.98 cm to −0.09 cm. The calibrated results yielded Nash coefficients of 0.96 and 0.97 and MAE values of 1.01 cm and 1.04 cm for 2017 and 2018, respectively. The revised model demonstrated stable applicability under different annual snow conditions and can be extended to longer time series and more diverse snow environments. ConclusionsThese findings substantiate the model's validity. The model provides a reliable theoretical basis for predicting snow depth during the snowmelt period and serves as a reference value for studying snow ablation in other regions.

       

    /

    返回文章
    返回