Fig. 2: Relative accuracy differences by air temperature for the machine learning methods compared to the benchmarks. | Nature Communications

Fig. 2: Relative accuracy differences by air temperature for the machine learning methods compared to the benchmarks.

From: Machine learning shows a limit to rain-snow partitioning accuracy when using near-surface meteorology

Fig. 2

Dashed lines show the comparison relative to the best benchmark and the solid lines are relative to the average benchmark value. The three machine learning phase partitioning methods (PPM) are the artificial neural network (ANN), random forest (RF) and XGBoost (XG) applied to the crowdsourced (left panel) and synoptic datasets (right panel). Source data are provided as a Source Data file.

Back to article page