Figure 3 | Scientific Reports

Figure 3

From: Explainable AI approach with original vegetation data classifies spatio-temporal nitrogen in flows from ungauged catchments to the Great Barrier Reef

Figure 3

Additive Performance Criteria for pseudo-ungauged catchment DIN simulations. Dimensionless graph shows the additive scores for the best MSE, R2, NSE, d, RMSE for each data discrimination scenario evaluated in the case study. Zero is the centrally located white star and represents the target/observed data. Discrimination of data to spatio-temporal regime (i.e., shown as crosses) and to the matched catchment improves results the most. Inclusion of Ecounit data improves simulation performance where training datasets are not discriminated to classified catchment or category.

Back to article page