The paper "Decomposition of the Mean Squared Error & NSE Performance Criteria: Implications for Improving Hydrological Modelling" by Hoshin V. Gupta, Harald Kling, Koray K. Yilmaz, and Guillermo F. Martinez-Baquero explores the limitations of using the Nash-Sutcliffe Efficiency (NSE) and Mean Squared Error (MSE) as criteria for calibrating and evaluating hydrological models. The authors decompose NSE into three components: correlation, conditional bias, and unconditional bias, and discuss how these components interact, leading to potential issues in model calibration. They propose an alternative criterion, the Kullback-Leibler Efficiency (KGE), which equally weights these components, and demonstrate its effectiveness through a case study using a simple precipitation-runoff model calibrated on Austrian basins. The results show that while NSE may lead to underestimation of flow variability and systematic underestimation of runoff peaks, KGE provides better calibration and more consistent performance during an independent evaluation period. The study highlights the importance of considering multiple criteria in hydrological modeling to improve model performance and diagnostic analysis.The paper "Decomposition of the Mean Squared Error & NSE Performance Criteria: Implications for Improving Hydrological Modelling" by Hoshin V. Gupta, Harald Kling, Koray K. Yilmaz, and Guillermo F. Martinez-Baquero explores the limitations of using the Nash-Sutcliffe Efficiency (NSE) and Mean Squared Error (MSE) as criteria for calibrating and evaluating hydrological models. The authors decompose NSE into three components: correlation, conditional bias, and unconditional bias, and discuss how these components interact, leading to potential issues in model calibration. They propose an alternative criterion, the Kullback-Leibler Efficiency (KGE), which equally weights these components, and demonstrate its effectiveness through a case study using a simple precipitation-runoff model calibrated on Austrian basins. The results show that while NSE may lead to underestimation of flow variability and systematic underestimation of runoff peaks, KGE provides better calibration and more consistent performance during an independent evaluation period. The study highlights the importance of considering multiple criteria in hydrological modeling to improve model performance and diagnostic analysis.