Skip to content
Switch branches/tags
Go to file
1 contributor

Users who have contributed to this file


Aqueduct 3.0 Water Risk Atlas Metadata

Creative Commons 4.0 License

Suggested Citation: Hofste, R., S. Kuzma, S. Walker, E.H. Sutanudjaja, et. al. 2019. “Aqueduct 3.0: Updated Decision- Relevant Global Water Risk Indicators.” Technical Note. Washington, DC: World Resources Institute. Available online at:

For the technical note, please click here

This document helps you understand the downloaded Aqueduct water risk atlas data. For questions, check out our FAQ page. The methodology and a full description of the indicators can be found in the Aqueduct 3.0 technical note and future projections technical note (todo: add link when ready).

There are two ways to download Aqueduct water risk atlas data:

  1. The full database, available here
  2. Site specific data by using the location analyzer in the Water Risk Atlas online tool. (Coming soon)
    These files contain a few extra columns explained here

New to GIS data or need help opening the files? See the "how to open" section.


The Aqueduct Water Risk Atlas features multiple water-related risk datasets:

  1. Baseline Annual
  2. Baseline Monthly
  3. Future Projections.

Baseline Annual

This dataset contains the main Aqueduct framework. This is the dataset that most users are interested in.

The columns (attributes or fields) in the Annual baseline results are grouped into "identifiers", "indicators" and "grouped water risk".


Column Name Data Type Description
string_id (string) contains a unique string for each geometry. Geometries are the union of hydrological basins, provinces and groundwater aquifers. The string_id is a combination of pfaf_id-gid_1-aqid. See the description of those columns.
aq30_id (integer) unique identifier in numeric format.
pfaf_id (integer) six digit Pfafstetter code for the hydrological basins.
gid_1 (string) identifier for sub-national units based on the GADM dataset. It contains the Iso A3 country code, followed by numeric values separated by underscores for each sub-national unit.
aqid (integer) identifier for groundwater Aquifers based on WHYMAP.
gid_0 (string) ISO A3 country code based on GADM.
name_0 (string) National or political entity name based on GADM.
name_1 (string) Sub-national or political entity name based on GADM.
area_km2 (double) area of the geometry in km2 (union of sub-basin, province and groundwater aquifer).


For each of the 13 indicators the columns contain the indicator abbreviation plus the type {indicator}_{type}, e.g.:
"bws_raw" is baseline water stress, raw value. The indicator abbreviations and types are listed below.

Physical risk quantity:

Short Full
bws Baseline water stress
bwd Baseline water depletion
iav Interannual variability
sev Seasonal variability
gtd Groundwater table decline
rfr Riverine flood risk
cfr Coastal flood risk
drr Drought risk

Physical risk quality:

Short Full
ucw Untreated connected wastewater
cep Coastal eutrophication potential

Regulatory and reputational risk:

Short Full
udw Unimproved/no drinking water
usa Unimproved/no sanitation
rri Peak RepRisk country ESG risk index


Type Data Type Description
_raw (double) raw value. Units depend on the indicator. See the technical note.
_score (double) each indicator is mapped to a [0-5] scale.
_label (string) A label explaining the category of the indicator including threshold. e.g. "Extremely High (more than 1 in 100)".
_cat (integer) integer for each category [-1,4], can be used for visuals.

Grouped water risk

see the technical note for a description of aggregating the 13 indicators into sub-groups and an overall water risk score using the composite index approach. The grouped water risk scores use the follwing format:


w_awr, stand for weighted aggregated water risk. Mainly used to keep them separate from the remaining indicators.

e.g. w_awr_min_rrr_score is the aggregated score using the mining weighting scheme for the regulatory and reputational risk group.

Weighting Scheme

Short Full
def Default
agr Agriculture
che Chemicals
con Construction Materials
elp Electric Power
fnb Food & Beverage
min Mining
ong Oil & Gas
smc Semiconductor
tex Textile


Short Full
qan Physical risk quantity
qal Physical risk quality
rrr Regulatory and reputational risk
tot Total, Overall water risk.


Type Data Type Description
_raw (double) raw value on 0-5 scale. Result of weighted composite approach
_score (double) score [0-5], result of applying a quantile approach to raw values. See below
_label (string) A label explaining the category of the grouped water risk.
_cat (integer) integer for each category [-1,4], can be used for visuals.
_weight_fraction (double) the fraction [0-1] of the group towards the overall water risk score. NoData is excluded from the weights and therefore the fractions can be lower than 1 depending on data availability. See the technical note for the weights per industy and indicator.

Quantile linear interpolation.

Aggregating multiple scores will change the distribution of the data. Therefore we calculate the quantiles and use linear interpolation to remap the raw values to 0-5 scores. The quantiles can be obtained from the data but are also listed below.

percentile score quantity quality regulatory and reputational overall
0 0 0.00 0.00 0.00 0.00
0.2 1 0.72 1.45 0.30 1.01
0.4 2 1.09 2.20 1.39 1.61
0.6 3 1.60 2.92 2.81 2.10
0.8 4 2.34 3.83 3.93 2.68
1 5 5.00 5.00 5.00 5.00

Baseline monthly


pfaf_id, (integer), six digit Pfafstetter code for the hydrological basins.
month, (integer), Month of the year.


bws, Baseline water stress,
bwd, Baseline water depletion,
iav, Interannual variability,


_raw, double, raw value. Units depend on the indicator. See the technical note.
_score, double, each indicator is mapped to a [0-5] scale.
_label, string, A label explaining the category of the indicator includin threshold. e.g. "Extremely High (more than 1 in 100)"
_cat, integer, integer for each category [-1,4], can be used for visuals.

Future Projections

The columns (attributes or fields) in the future projections results are grouped into "identifiers" and "indicators". Methodology document


BasinID, integer, Sub-basin identifiers.
dwnBasinID, integer, Next downstream sub-basin.
Area_km2, double, Area of sub-basin in square kilometer.
Shape_Leng, double, Perimeter of the sub-basin in kilometer.


There are four indicators, three target years, three scenarios, three data types.

The indicators use the following format:

II, indicator code
YY, year code
SS, scenario code
T, data type code
X, suffix

{II} Indicator codes

ws water stress
sv seasonal variability
ut water demand
bt water supply

{YY} Year codes

20 2020
30 2030
40 2040

{SS} Scenario codes

24 ssp2 rcp45 (optimistic)
28 ssp2 rcp85 (business as usual)
38 ssp3 rcp85 (pessimistic)

{T} Data types

c change from baseline
t future value
u uncertainty value (available for seasonal variablity and water supply)

{X} Suffixes

l label string
r raw value


Units depend on the combination of tha above options. Units are [m/y] for ut, bt future value and dimensionless for ws and sv respectively. Other units are self-explanatory.

For example the layer {ws4028cl} is "projected change in water stress by the year 2040 under a business as usual (ssp2 rcp85) scenario"

Extra columns for Location Analyzer

These columns are only added to your data if you analyzed your locations in the online tool. Users that download the entire dataset will not have these.

Column Name Data Type Description
location_name (string) the user-defined name for your location.
input_address (string) a copy of the address as specified by the user. Only if using addresses as input file.
match_address (string) the address as matched by our geocode API. You can check if the match was succesful by comparing with the input_address. Only if using addresses as input file.
latitude (double) location latitude (y) in decimal degres. Either specified by the user or derived from the geocoder API.
longtitude (double) location longitude (x) in decimal degrees. Either specified by the user or derived from the geocoder API.
major_basin_name (string) Name of the major river basin name.
minor_basin_name (string) Name of the minor river basin name.
aquifer_name (string) Name of the groundwater aquifer system.

How to use

Aqueduct 3.0 is available in various GIS data formats. The entire database comes in a zipped folder. Unzip the folder first. For QGIS, simply open the QGIS project files (.qgz). For use of the data in ArcMap, simply click the project files(.mxd) files and the data and styles should load correctly.

For other programs we recommend opening the GeoPackage (.gpkg) and shapefiles (.shp).

The baseline data is not available in shapefile format because the number of columns exceeds the maximum of this format.

We also provide tabular data (without the geometry) in Excel (.csv UTF-8) format.

ArcMap 10.x and above
QGIS 3.4.x and above


This script, can help you generate the column names programatically.

Issues or suggestions ?

Pleasse check our FAQ page and technical notes first. If the issues is unresolved, please open an issue to this github repo.