Evaluation of random forests for short-term daily streamflow forecasting in rainfall- and snowmelt-driven watersheds
Evaluation of random forests for short-term daily streamflow forecasting in rainfall- and snowmelt-driven watersheds
复制标题
DOI：
10.5194/hess-25-2997-2021
复制
发表时间：
2021-06
影响因子：
6.3
通讯作者：
Leo T. Pham;L. Luo;A. Finley
中科院分区：
文献类型：
--
作者：
Leo T. Pham;L. Luo;A. Finley
抽象的。在过去的几十年中，数据驱动的机器学习（ML）模型已成为短期水流预测的有前途的工具。除其他优点外，机器学习模型在此类应用中的受欢迎是因为它们相对容易实施、不太严格的分布假设以及具有竞争力的计算和预测性能。尽管取得了令人鼓舞的结果，但大多数用于径流预测的机器学习应用仅限于降雨是径流主要来源的流域。在本研究中，我们评估了随机森林 (RF)（一种流行的 ML 方法）在太平洋西北部 86 个流域进行 1 天的水流预测的潜力。这些流域涵盖不同的气候条件和地形环境，并表现出降雨和融雪对其水流的不同贡献。根据年流量中心的时间，流域被分为三种水文状况：降雨主导型、瞬态型和融雪主导型。 RF 性能以朴素模型和多元线性回归 (MLR) 模型为基准，并使用四个标准进行评估：确定系数、均方根误差、平均绝对误差和克林古普塔效率 (KGE)。模型评估分数表明，与降雨驱动的流域相比，RF 在融雪驱动的流域中表现更好。与基准模型相比，预测的最大改进出现在降雨驱动的流域中。射频性能随着流域坡度和土壤沙质的增加而恶化。我们注意到两种流行的射频变量重要性衡量标准之间存在分歧，并建议将这些衡量标准与正在研究的物理过程结合考虑。这些结果和其他结果为有效应用基于射频的水流预测提供了新的见解。
Abstract. In the past decades, data-driven machine-learning (ML) models have emerged as promising tools for short-term streamflow forecasting. Among other qualities, the popularity of ML models for such applications is due to their relative ease in implementation, less strict distributional assumption, and competitive computational and predictive performance. Despite the encouraging results, most applications of ML for streamflow forecasting have been limited to watersheds in which rainfall is the major source of runoff. In this study, we evaluate the potential of random forests (RFs), a popular ML method, to make streamflow forecasts at 1 d of lead time at 86 watersheds in the Pacific Northwest. These watersheds cover diverse climatic conditions and physiographic settings and exhibit varied contributions of rainfall and snowmelt to their streamflow. Watersheds are classified into three hydrologic regimes based on the timing of center-of-annual flow volume: rainfall-dominated, transient, and snowmelt-dominated. RF performance is benchmarked against naïve and multiple linear regression (MLR) models and evaluated using four criteria: coefficient of determination, root mean squared error, mean absolute error, and Kling–Gupta efficiency (KGE). Model evaluation scores suggest that the RF performs better in snowmelt-driven watersheds compared to rainfall-driven watersheds. The largest improvements in forecasts compared to benchmark models are found among rainfall-driven watersheds. RF performance deteriorates with increases in catchment slope and soil sandiness. We note disagreement between two popular measures of RF variable importance and recommend jointly considering these measures with the physical processes under study. These and other results presented provide new insights for effective application of RF-based streamflow forecasting.