Extended Data Fig. 5: Further validation metrics of DynQual_Random Forest and DynQual.
From: Climate change drives low dissolved oxygen and increased hypoxia rates in rivers worldwide
(a) Median in Root-Mean-Squared-Errors (RMSEs) per river basin for DynQual_Random Forest (b) RMSEs for DynQual (blue) and DynQual_Random Forest (red) (c) median RMSEs per river basin for DynQual (d) lower 10th percentiles of dissolved oxygen (DO) data: median RMSEs per basin for DynQual_Random Forest (e) RMSEs for DynQual and DynQual_Random Forest (DO data < 10th percentiles) (f) median RMSEs for DynQual (DO data < 10th percentiles) (g) mean annual RMSE values (DynQual_Random Forest) (h) Mean absolute errors (MAE) DynQual and DynQual_Random Forest (i) Mean annual RMSE values for DynQual (j) Frequency of stress days (DO < 5 mg/l) simulated and observed (k) Frequency of hypoxia days (DO < 3 mg/l) simulated and observed. River basin shapefiles in a, c, d and f adapted with permission from ref. 60, Wiley.
