Abstract Integrated groundwater management is a major challenge for industrial, agricultural and domestic activities. In some agricultural production systems, optimized water table management represents a significant factor to improve crop yields and water use. Therefore, predicting water table depth (WTD) becomes an important means to enable real-time planning and management of groundwater resources. This study proposes a decision-tree-based modelling approach for WTD forecasting as a function of precipitation, previous WTD values and evapotranspiration with applications in groundwater resources management for cranberry farming. Firstly, two decision-tree-based models, namely Random Forest (RF) and Extreme Gradient Boosting (XGB), were parameterized and compared to predict the WTD up to 48 -h ahead for a cranberry farm located in Quebec, Canada. Secondly, the importance of the predictor variables was analyzed to determine their influence on WTD simulation results. WTD measurements at three observation wells within a cranberry field, for the growing period from July 8, 2017 to August 30, 2017, were used for training and testing the models. Statistical parameters such as the mean squared error, coefficient of determination and Nash-Sutcliffe Efficiency coefficient were used to measure models performance. The results show that the XGB model outperformed the RF model for all predictions of WTD and was, accordingly, selected as the optimal model. Among the predictor variables, the antecedent WTD was the most important for water table depth simulation, followed by the precipitation. Based on the most important variables and optimal model, the prediction error for entire WTD range was within ±5 cm for 1-, 12-, 24-, 36- and 48 -h predictions. The XGB models can provide useful information on the WTD dynamics and a rigorous simulation for irrigation planning and management in cranberry fields.

中文翻译：

---

使用两种决策树建模方法进行蔓越莓田地下水位深度预测

摘要 地下水综合管理是工业、农业和家庭活动的主要挑战。在一些农业生产系统中，优化的地下水位管理是提高作物产量和用水量的重要因素。因此，预测地下水位深度（WTD）成为实现地下水资源实时规划和管理的重要手段。本研究提出了一种基于决策树的建模方法，将 WTD 预测作为降水、先前 WTD 值和蒸散的函数，并在蔓越莓种植的地下水资源管理中应用。首先，对两个基于决策树的模型，即随机森林 (RF) 和极限梯度提升 (XGB) 进行参数化并进行比较，以预测位于加拿大魁北克的蔓越莓农场最多 48 小时的 WTD。其次，分析了预测变量的重要性，以确定它们对 WTD 模拟结果的影响。2017年7月8日至2017年8月30日期间，蔓越莓田内三个观测井的WTD测量值用于训练和测试模型。统计参数如均方误差、决定系数和 Nash-Sutcliffe 效率系数用于衡量模型性能。结果表明，XGB 模型在 WTD 的所有预测方面都优于 RF 模型，因此被选为最佳模型。在预测变量中，前因 WTD 对地下水位深度模拟最重要，其次是降水。基于最重要的变量和最优模型，对于 1、12、24、36 和 48 小时的预测，整个 WTD 范围的预测误差在 ±5 厘米以内。XGB 模型可以提供有关 WTD 动态的有用信息以及蔓越莓田灌溉规划和管理的严格模拟。