Evaluation of SVM, ELM and four tree-based ensemble models for predicting daily reference evapotranspiration using limited meteorological data in different climates of China
Evaluation of SVM, ELM and four tree-based ensemble models for predicting daily reference evapotranspiration using limited meteorological data in different climates of China
复制标题
基于有限气象数据的中国不同气候下的SVM、ELM和四种基于树的集合模型预测日参考蒸散量的评估
DOI：
10.1016/j.agrformet.2018.08.019
复制
发表时间：
2018-12-15
影响因子：
6.2
通讯作者：
Xiang, Youzhen
中科院分区：
文献类型：
--
作者：
Fan, Junliang;Yue, Wenjun;Xiang, Youzhen
准确估算参考蒸散量（ET0）对于区域水资源规划和灌溉调度设计具有重要意义。建议采用FAO-56 Penman-Monteith模型作为预测ET0的参考模型，但其应用通常因全球许多地区缺乏完整的气象数据而受到限制。本研究评估了机器学习模型的潜力，特别是四种相对简单的基于树的组装算法（例如随机森林（RF）、M5模型树（M5Tree）、梯度提升决策树（GBDT）和极限梯度提升（XGBoost）），使用K折交叉验证方法利用有限的气象数据估计每日ET0。为了评估基于树的模型的预测精度、稳定性和计算成本，这些模型进一步与其相应的支持向量机（SVM）和极限学习机（ELM）模型进行比较。利用中国不同气候的8个代表性气象站1961-2010年的气象数据，以T-max、T-min和R-a为基础数据集，考虑了日最高和最高气温（T-max和T-min）、相对湿度（H-r）、风速（U-2）、全球和地外太阳辐射（R-s和R-a）四种输入组合。结果表明，在缺乏完整气象数据的情况下，利用T-max、T-min、H-r、U-2和R-a的机器学习模型在中国温带大陆、山地高原和温带季风区获得了令人满意的ET0估计值（RMSE < 0.5 mm d(-1)）。然而，具有 T-max、T-min 和 R-s 三个输入参数的模型对于热带和亚热带地区的每日 ET0 预测更为优越。 ELM 和 SVM 模型提供了预测准确性和稳定性的最佳组合。简单的基于树的 XGBoost 和 GBDT 模型表现出与 SVM 和 ELM 模型相当的准确性和稳定性，但计算成本却低得多。考虑到所研究模型的复杂程度、预测精度、稳定性和计算成本，推荐使用XGBoost和GBDT模型用于中国不同气候带以及世界上类似气候的其他地区的每日ET0估算。
Accurate estimation of reference evapotranspiration (ET0) is of great importance for the regional water resources planning and irrigation scheduling design. The FAO-56 Penman-Monteith model is recommended as the reference model to predict ET0, but its application is commonly restricted by lack of complete meteorological data at many worldwide locations. This study evaluated the potential of machine learning models, particularly four relatively simple tree-based assemble algorithms (Le. random forest (RF), M5 model tree (M5Tree), gradient boosting decision tree (GBDT) and extreme gradient boosting (XGBoost)), for estimating daily ET0 with limited meteorological data using a K-fold cross-validation method. For assessment of the tree-based models in terms of prediction accuracy, stability and computational costs, these models were further compared with their corresponding support vector machine (SVM) and extreme learning machine (ELM) models. Four input combinations of daily maximum and maximum temperature (T-max and T-min), relative humidity (H-r), wind speed (U-2), global and extra-terrestrial solar radiation (R-s and R-a) with T-max, T-min and R-a as the base dataset were considered using meteorological data during 1961-2010 from eight representative weather stations in different climates of China. The results showed that, when lack of complete meteorological data, the machine learning models using T-max, T-min, H-r, U-2 and R-a obtained satisfactory ET0 estimates in the temperate continental, mountain plateau and temperate monsoon zones of China (RMSE < 0.5 mm d(-1)). However, models with three input parameters of T-max, T-min and R-s were superior for daily ET0 prediction in the tropical and subtropical zones. The ELM and SVM models offered the best combination of prediction accuracy and stability. The simple tree-based XGBoost and GBDT models showed comparable accuracy and stability to the SVM and ELM models, but exhibited much less computational costs. Considering the complexity level, prediction accuracy, stability and computational costs of the studied models, the XGBoost and GBDT models have been recommended for daily ET0 estimation in different climatic zones of China and maybe elsewhere with similar climates around the world.