University College London
ERFA risks of ecological change for global river flows.xlsx (108.3 kB)

ERFA risks of ecological change for global river flows

Download (108.3 kB)
posted on 2021-03-03, 17:16 authored by Julian ThompsonJulian Thompson, Simon Gosling, Jamal Zaherpour, Cedric LaizeCedric Laize

These data summarise the risks of ecological change due to climate change-induced modifications to river flow for 321 major global river basins. The basins are a subset of those within the DDM30 global river network (Döll & Lehner, 2002) and are identified by Global Runoff Data Centre (GRDC) ID, river and gauging station name. The dominant hydrobelt according to Meybeck et al. (2013) of each basin is specified.

Risk of ecological change in high and low flows is based on the Ecological Risk due to Flow Alteration (ERFA) approach as described by Laizé and Thompson (2019). ERFA was used to compare simulated river flows from global hydrological models (GHM) for a baseline period (1980–2010) with those for periods associated with 1.0°C, 1.5°C, 2.0°C and 3.0°C increases in global mean temperature. Simulated river flows were from nine GHMs:










Climate projections for each of the four warming scenarios were provided from five global climate models (GCM):






In this way, for a given warming scenario, a total of 45 baseline-scenario times series were used (i.e. 9 GHM x 5 GCMs) for each river basin. Across the 321 basins and four warming scenarios, this equates to 57,780 baseline-scenario pairs.

ERFA assigns the risk of change in high and low flows to one of four classes (no risk, low risk, medium risk and high risk). This assignment is based on the number of indicators used to characterize the river flow time series that differ significantly between the scenario and baseline (see Laizé and Thompson 2019). This dataset summarizes the resulting risks of change defined in this way across the 57,780 baseline-scenario pairs.

Risk classes

These data comprise the number of baseline-scenario pairs for each river basin that are assigned to each of the four classes (i.e. no risk, low risk, medium risk and high risk). Results are grouped initially by high and low flows and then within these two flow extremes the four warming scenarios (i.e. 1.0°C, 1.5°C, 2.0°C and 3.0°C increases in global mean temperature).

Total percentage risk score

An alternative, more concise, summary of the risks of ecological change is provided by the total percentage risk score. This metric is based on first cumulating the overall risk scores (i.e. 0–3 for no–high risk) for each basin and a given warming scenario. This is undertaken for both high and flow flows. These totals are then expressed as a percentage of the possible highest score of 135 - i.e. if all 45 (9 GHMs × 5 GCMs) baseline-scenario pairs for a basin under a given warming scenario were classified as high risk (score 3). In this way, a single value for both high and low flows representing risks across the ensemble of GHMs and GCMs is provided for each of the four warming scenarios.


Döll, P., & Lehner, B. (2002). Validation of a new global 30-min drainage direction map. Journal of Hydrology, 258, 214–231.

Laizé, C. L. R., & Thompson, J. R. (2019). R implementation of the Ecological Risk due to Flow Alteration (ERFA) method. Wallingford: NERC Environmental Information Data Centre.

Meybeck, M., Kummu, M., & Dürr, H. H. (2013). Global hydrobelts and hydroregions: improved reporting scale for water-related issues? Hydrology and Earth System Sciences, 17(3), 1093–1111.


International - Translation of Environmental Flow Research in Cambodia (TEFRIC)

Natural Environment Research Council

Find out more...

water 581 SCenarios for Europe and NEighbouring States" (SCENES - held by CEH)