GloFAS v4, the 0.05 degrees quasi-global (-180,180,90,-60) implementation of the LISFLOOD model, was calibrated using in-situ discharge gauge stations with a minimum drainage area of 500 km²  and at least 4-years-long time series of measurements more recent than 01 January 1980.

1996 quality checked time series were used for calibration. More specifically, 212 calibration points were in Europe, 250 in Asia, 61 in Oceania, 420 in Africa, 617 in Centre-North America, and 436 in South America.

Catchments for which in situ discharge data could be used for calibration entailed 47.5 % of the quasi-global domain (Figure 6, blue area). For these catchments, the Distributed Evolutionary Algorithm for Python (DEAP, Fortin et al. 2012) was used to explore the parameter space and identify the parameter set leading to the highest value of the modified Kling Gupta Efficiency (KGE', Gupta et al., 2009).  

Parameters of the catchments for which in situ discharge data were not available (Figure 5, yellow area) were estimated by parameter regionalization.

Figure 6- Subdivision GloFAS domain into LISFLOOD calibration inter-catchments.

Figure 5 – Catchments with available discharge data (gauged catchments) in blue.

Parameter estimation for gauged catchments

The Distributed Evolutionary Algorithm for Python (DEAP, Fortin et al. 2012) was used to explore the parameter space and identify the parameter set leading to the highest value of the modified Kling Gupta Efficiency (KGE, Gupta et al., 2009), as implemented by the open-source calibration tool. 

When multiple calibration points are available in one basin, the calibration protocol follows a top-down approach from head-catchments to downstream catchments; each segmentation of the area is called inter-catchment. Figure 6 shows the fragmentation of the area with available discharge observations into inter-calibration catchments.

Figure 6- Subdivision GloFAS domain into LISFLOOD calibration inter-catchments.

The size of the inter-catchments is mainly driven by data availability. The largest inter-catchment is located in the Congo basin: the scarcity of stations in this basin (especially along the main river) led to an inter-catchment with drained area larger than 2.500.000 km2. Figure 7 shows the distribution of the area of the calibration inter-catchments, the median value is 14.000 km2.

Figure 7- Distribution of the area of the LISFLOOD calibration inter-catchments. Note the logarithmic scale of the x-axis.

The overall calibration period was from 01/01/1982 to 31/12/2019. Each calibration station uses a different calibration period, depending on length of available discharge observations, but a minimum of 4 years is always used for calibration. A spin-up period of three year was added to each model run to ensure that model's variable are correctly initialised (please see par XXX for the exceptions to this criteria).

If a period longer than eight years was available, the discharge record was split in two for calibration and verification purposes. The most recent period was used for the calibration because the most recent forcing data and discharge observations are expected to have lower uncertainty and to provide a closer representation of the climatic and hydrological conditions of the forecast period. Figure 8 shows the distribution of the length of the time series used for calibration.

Figure 8 –calibration stations, number of years used in calibration

14 LISFLOOD parameters were simultaneously calibrated for each catchment, with the purpose to optimize the modelling of snow melt, water infiltration into the soil, surface water flow, groundwater flow, lakes and reservoirs dynamics. Feasible parameter ranges were defined for each parameter used in calibration to obtain more physically realistic calibrated parameters.

Table 1 - LISFLOOD calibration parameters for GloFAS 4.

Parameter regionalization

Catchments for which in situ discharge data was not available entailed 52.5% of the quasi-global domain (Figure 5, yellow area). A parameter regionalization approach was implemented to estimate the parameters that control snow melt, infiltration, runoff, groundwater, and routing for those catchments (11 parameters in total, see Table 1). In the implemented regionalization approach, calibrated parameter values were transferred from calibrated catchments (donors) to uncalibrated catchments. For each uncalibrated catchment, the donor catchment was identified according to a proximity criterion accounting for both climatic similarity and the geographical proximity (Parajka et al 2005, Beck et al. 2016). Parameters controlling the local behavior of lakes and reservoirs cannot be transferred from donor catchments to target catchments: default lakes and reservoirs parameter values were used in target catchments.

A leave-one-out cross validation approach for a subset of 739 gauged catchments was used to verify the performances of the pragmatic regionalization approach. Moreover, stations with observation time series shorter than 4 years (hence excluded from calibration) were used to verify the benefit of using the regionalized parameters over the default parameter values. As an example, Figure 9 allows to compare the observed discharge (black line) with the model results generated using the regionalized parameters (red line) and the default parameters (blue line) for one small catchment on the Black Sea (top) and one small catchment in Siberia (bottom).

Finally, it is here noted coastal and endorheic catchments with drainage area smaller than 500 km2 are modelled using the default parameter values (LINK or Table 1).

Figure 9 – Observed (black) and modelled discharge using regionalized parameters (red) and the default parameter set (blue) for two not calibrated catchments located on the Black Sea (top) and in Siberia (bottom)