Introduction

The GloFAS v4.0 hydrological model performance was evaluated in the model calibration context in GloFAS v4 calibration hydrological model performance, using only the stations involved in the calibration and only over the specific periods related to the calibration excersise.

...

This GloFAS v4.0 hydrological model performance assessment presented here, is based on the historical river discharge reanalysis available at https://cds.climate.copernicus.eu/cdsapp#!/dataset/cems-glofas-historical?tab=overview.

Verification period

The verification focused on the whole period with all available river discharge observations, similarly as it is in the GloFAS hydrological model performance layer in the map viewer (GloFAS hydrological model performance web product). 

Performance scores

GloFAS hydrological performance verification is done against river discharge observations available to the GloFAS team. The hydrological model performance analysis was conducted based on the modified Kling–Gupta efficiency metric (KGE'; ideal value is 1):

Image Added, Image Added, Image Added

The three component scores of the KGE' were also used:

• Pearson correlation (r) in KGE' highlights temporal errors through the strength of the linear relationship between simulation and observation time series. It ranges from -1 to 1, with 1 as ideal value.
  
• Bias ratio (β) represents the bias errors, ranging from 0 to +Inf, with 1 as ideal value. Relative bias pbias (ideal value = 0) defined as β-1 (or its absolute value abspbias)
• Variability ratio (γ) shows the variability related errors in the simulation. It ranges from 0 to +Inf, with 1 as optimal value. Relative variability var (ideal value = 0) defined as γ-1 (or its absolute value absvar)

In the GloFAS v4 calibration hydrological model performance and the GloFAS hydrological model performance web product, the KGE' and the three KGE' components (r, β and γ as bias and variability ratios and correlation) were used.

However, in this general GloFAS v4.0 model evaluation, besides the correlation (denoted as pcorr), the β-1 (pbias) and γ-1 (var) were considered as the bias and variability ratio versions, which both have 0 as optimal values instead of 1. In addition, the absolute values of pbias (abspbias) and var (absvar) were also used, which help to show skill differences between models.

Finally, a specific index was also used for measuring timing errors (timing in days; ideal value is 0), which shows the time delay between the simulated and observed river discharge time series (and also the absolute value abstiming). Timing is a time lag (or shift) L that maximises R_xy(L), cross correlation function R_xy(m) with the simulated (x) and observed (y) time series shifted by L days. Positive/negative timing error indicates late/early simulated river discharge. So, for example a timing error of +5 means the simulation needs to be shifted by 5 days backwards (brought earlier) to get to the highest correlation, i.e. the simulation is generally 5-day late predicting the ups and downs in the flow time series. Although this is not directly equivalent to measuring the timing error of the highest flood peaks, it is in very good relation with that and can be used as a simple estimate.