From raw signals to decomposed insights: boosting models enhanced by EMD variants for multi-depth soil temperature and moisture prediction
TL;DR
The obtained results revealed that, for soil temperature prediction, the EMD, EEMD, and CEEMDAN significantly contributed to the improvement of prediction accuracy, and the XGBoost model exhibited high performance with root-mean-square error (RMSE), mean absolute error (MAE), Nash–Sutcliffe efficiency (NSE), and correlation coefficient (R) of 1.2.