Environment Database - Freshwater abstractions (million m3)
< < >-< OECD.Stat
Open all groups and itemsClose all groups and itemsSend link via emailPrintOpen in stand alone windowClose this window
Click to expand Data Characteristics
Click to collapse Data Characteristics
Click to expand Variables collected
Click to collapse Variables collected
This dataset shows the state and changes over time in the abstractions of freshwater resources.

Water abstractions are a major pressure on freshwater resources, particularly from public water supplies, irrigation, industrial processes and cooling of electric power plants. It has significant implications for issues of quantity and quality of water resources.

This dataset shows water abstractions by source (surface and ground water) and by major uses. Water abstractions refer to water taken from ground or surface water sources and conveyed to the place of use. If the water is returned to a surface water source, abstraction of the same water by the downstream user is counted again in compiling total withdrawal.

When interpreting those data, it should be borne in mind that the definitions and estimation methods employed by Member countries may vary considerably among countries

Last updated: July 2017

Contact: ENV.Stat@oecd.org

Disclaimer: Data available to the OECD Secretariat by April 2017, and subject to possible slight revisions

Australia: Abstraction by agricultural sector: 2001-07 data are estimates based on national reports; due to changes in methodology, data may not be comparable with previous years. 1985: public supply refers to the domestic sector

Austria: Partial totals including estimates (gross abstractions in 1990-1995)

Canada: Agriculture 1980-1995: data exclude forestry and fishing

Chile: Only data on public supply is available. This data is based on the information that the sanitary companies provide for the areas where they provide their services

Czech Republic: Up to 1995 water abstractions for fishponds were included (change of methodology - break in series). The drop in total abstractions in 2013 is explained by the lower water abstraction for cooling for electricity production. For more information see http://voda.gov.cz/portal/en/

Denmark: 2009: partial totals for public water supply and manufacturing industry. Irrigation includes fish farming. A rough estimate for abstractions for aquaculture is 180 million m3/year, but data is not published because deemed unreliable. For more information : http://www.dst.dk/ ; http://www.geus.dk/ ; www.statistikbanken.dk/vandind

Estonia: Production of electricity: up to 2001 data refer to total abstractions for electricity production (ISIC 35.1 Rev.4). Since 2001 data refer to the NACE activity 40.1, which means that part of the cooling water is allocated to the "other" category. Public water supply (surface water, 2000-2001): data include a high share of water use by manufacturing enterprises

France: Total abstractions in 1975: estimates based on four basins out of six. Data refers only to metropolitan France (including Corsica); data for overseas territories is available since 2012 but is not included here (they represent 1.5% of metropolitan France abstractions, including electricity cooling, and 3.5% excluding it). Data are estimated using basin agencies' calculations on the licence fees. Most of water abstracted for agriculture is used for irrigation, but there is no available time series. Data on manufacturing industry includes mining and quarrying, construction and services (the manufacturing industry represents around 80% of this total). Abstractions for drinkable water since 1990 have first increased then stabilised and finally decreased thanks to improvements in equipment and in behaviour. However, the reduction of leakages in the distribution system (which has a rate of performance of 80%) remains an opportunity for water savings. In the energy sector, water is mostly used for nuclear power plants, and the improvements in the cooling systems have stabilised water abstractions. The industrial sector has managed to reduce its consumption by a third since 2003. Water abstractions for irrigation vary with the weather conditions: depending on the years, tensions can arise locally and cause restrictions and usage conflicts. For more information see http://www.statistiques.developpement-durable.gouv.fr/ ; http://geoidd.developpement-durable.gouv.fr/  ; http://www.hydro.eaufrance.fr/  ; http://www.statistiques.developpement-durable.gouv.fr/lessentiel/s/ressources-eau.html

Germany: Totals up to 1998 exclude all agricultural uses except irrigation. From 2001 totals include mining and quarrying. Before 1991 data refer to western Germany only. Some of the water used for the production of electricity might be obtained from other industrial sectors or from public water supply. Total surface water abstraction and total ground water abstraction may not add up to total freshwater abstraction because abstraction by private households cannot be broken down to ground and surface water (only the total abstracted amount can be estimated).

Greece: Partial totals excluding agricultural uses besides irrigation (agriculture includes only irrigation)

Hungary: Break in series in 2000: change in data source ("Water resources fee" database instead of the "Report on industrial water uses"). The large share of freshwater abstracted for electricity cooling is due to a nuclear power plant

Ireland: 1994:  estimates including 1980 data for electrical cooling. Agriculture includes only irrigation. Break in series in 2005. The methodology has changed and there is no data for sectors. Abstractions are based upon the amount of water supplied by treatment plants. Figures cannot be verified e.g. against metering data, yet. The data is supplied by local authorities and the national water utility. There is ongoing work to review and update the dataset in order to improve the accuracy of the data. Any detectable trend of increased abstraction may be due to a more complete dataset for 2014 rather than to any true increase in abstractions. Annual drinking water report available at http://www.epa.ie/pubs/reports/water/drinking . Historic information on drinking water monitoring results and water supply details for each county (dating back to the year 2000) is available on the EPA's SAFER (Secure Archive for Environmental Research Data) web-page at http://erc.epa.ie/safer/resourcelisting.jsp?oID=10206&username=EPA%20Drinking%20Water 

Iceland: Public supply includes the domestic use of geothermal water. Since 1995 fish farming is a major user of abstracted water, explaining the change in the relative contribution of other sectors.

Japan: Public supply: data refer to households and service sector. Agriculture excludes forestry and fishing. Manufacturing industry includes electricity cooling

Korea: Public supply refers to the domestic sector only (households and the commercial sector, excluding the agricultural and industrial sectors). Agriculture includes only irrigation. Break in series in 2013: for groundwater the Jeju-do area has been included, and so data is not comparable to earlier years. For more information see: National Groundwater Information Center ( http://www.gims.go.kr/), Water Resources Management Information System (http://www.wamis.go.kr/).

Luxembourg: 1989: estimates including 1983 data for industry and electrical cooling. Agriculture includes only irrigation in 1995-1999. Further references: http://www.eau.public.lu/

Mexico: 2001 onwards: volumes of water granted in concessions (administrative figures, not collected data); prior data are estimates. "Manufacturing industry" figures refer to all industries (not only manufacturing) and include some services. Hydroelectricity generation includes in-situ uses (figures for individual uses are not available). For more information: http://www.conagua.gob.mx/ , http://www.gob.mx/semarnat/acciones-y-programas/sistema-nacional-de-informacion-ambiental-y-de-recursos-naturales

Netherlands: Before 1980 data include marine waters. After 1980 data exclude underground flows (estimated at 2 billion m3). Partial totals excluding all agricultural uses before 1990. Manufacturing industry in 1976 includes mining and quarrying. For more information: http://statline.cbs.nl/Statweb/publication/?DM=SLNL&PA=82883NED&D1=a&D2=0,23,6,39,41&D3=a&VW=T , http://www.cbs.nl/nl-NL/menu/themas/natuur-milieu/publicaties/milieurekeningen/publicaties/archief/2014/2014-environmental-accounts-of-the-netherlands-2013-pub.htm

New Zealand: Data exclude storage water (dams and lakes). 2010 figures are based on an estimated water abstraction of 50% of water allocations. This is based on the average consumption in all regions excepting Southland, where the 16,000 m3 allocated to hydroelectricity - 60% of the total national allocation - skews the national average

Norway: Freshwater abstractions: since 1996 data include water abstractions for aquaculture. Totals include estimates.

Agriculture and irrigation: these series have been discontinued in 2007. Manufacturing industry: this series has been discontinued in 2010

Poland: Totals include mining and construction water discharged without use. Data include abstractions for agriculture, which refer to aquaculture (areas over 10 ha) and irrigation (arable land and forest areas greater than 20 ha). Animal production and domestic needs of rural inhabitants are not covered (self-supply). Aquaculture: water abstraction for filling fishponds exceeding 10 ha. Industry cooling includes marine water. For more information, see: http://www.kzgw.gov.pl/ , http://www.pgi.gov.pl/ , http://www.psh.gov.pl/ , http://www.imgw.pl/

Portugal: Data do not include the Azores and Madeira. Total freshwater abstractions in 2013 include 5,6 million m3, and 1.5 million in 2014, of water abstracted from an unknown source. Irrigation: data estimated using mathematical models for calculating soil water balance. It is uncertain how much is abstracted from surface and from groundwater. The estimates reflect the characteristics of the soil and climate, reflected by specific coefficients of the mathematical model, and therefore care needs to be used in making comparisons. The recent decrease in public water abstractions is due to the effects  of the economic crisis and the consequent need for water savings and for a more efficient use. For more information see http://www.ine.pt/ (Statistics Portugal) and http://www.ersar.pt/  (Water and Waste Services Regulation Authority).

Slovak Republic: For more information see https://www.enviroportal.sk/indicator/detail?id=1761 and https://www.enviroportal.sk/spravy/index

Slovenia: Agriculture includes only irrigation. The production of electricity includes Gas and Steam and hot water supply activities. Break in series in 2012: previous values are under-reported (they did not take into account some reporting units)

Spain: Before 2000 agriculture includes only irrigation. After 2000 it excludes the fishing sector. Totals also exclude the fishing sector. 1995 data do not refer to 1995, but represent an average hydrological year, except for those data referring to cooling of thermal and nuclear plants

Sweden: Irrigation: 1985-2004 estimates for dry year. For more information: http://www.scb.se/

Switzerland: Partial totals excluding all agricultural uses. In 2006 they exclude rainwater use. Public supply includes total industry (ISIC 5-43 rev. 4), which totals 215 million m3 in 1994, and other activities (101 million m3 in 1994). In 2006 they are 158 million m3 (total industry) and 84 million m3 (other activities). Production of electricity (cooling) includes non-consumptive water use (cooling on rivers). Data for 2012 contains estimates

Turkey: Break in series in 2008 for public water supply (before 2008 data refer to urban areas only). Totals: Estimates including data based on partial inventories, excluding agricultural uses besides irrigation and, until 1993, electrical cooling. Agriculture: irrigation only. Manufacturing industry and electrical cooling: data are based on partial surveys. 1993 data include abstractions by industry. 1991: partial totals for public supply include manufacture and industry cooling, whereas the grand total excludes them. 2005 and 2008: partial data including estimates based on previous year data for abstraction by industry and electrical cooling.

England and Wales: Series break in 1991 and 1999 (significant changes in reporting methods and classifications). Figures are therefore not strictly comparable with those in previous/intervening years. Data for 2014 refers to England only. Public water supply includes estimations. Agriculture and irrigation: Under the Water Act 2003 abstraction of less than 20m3/day became exempt from the requirement to hold a licence as of 1 April 2005. As a result over 22,000 licences were deregulated, mainly for agriculture or private water supply purposes. However, due to the small volumes involved this has had minimal effect on the estimated licensed and actual abstraction totals Estimates for irrigation are solely for spray irrigation. Return requirements, which set out the actual abstraction information requested from abstractors, were standardised across England & Wales from 1 April 2008. Returns are now requested on a financial year basis. Prior to April 2008 returns were either by calendar year or financial year. To account for this, data collection for 2008 was the subject of two requests. At the end of the period January 2008 to March 2008 and at the end of the period April 2008 to March 2009. This may have had the effect of underestimating actual abstraction.

OECD Total: ross freshwater abstractions per capita, as percentage of total renewable resources and as percentage of internal resources are OECD Secretariat estimates. Data exclude Chile and the United Kingdom refers to England and Wales only

Colombia: The data presented for the different years are not comparable, given the different methodology applied and the database used. Production of electricity refers to the water demand for cooling by the thermoelectric sector, which does not include water demand for the generation of hydroelectricity. Amounts of water abstracted from surface and groundwater sources are estimated. Groundwater demand is estimated at 15% of total demand. For more information: http://www.ideam.gov.co/ , http://documentacion.ideam.gov.co/openbiblio/bvirtual/023080/023080.html

Costa Rica: Data includes only legally authorised abstraction. There is a significant amount of water that is illegally abstracted and is not included. It is possible that some concessions that were withdrawn are not included. Concessions granted in previous periods are also not included. Public supply includes (authorised) water abstracted by private households. There might be manual errors in the compilation of the source database on water abstraction concessions. Data should therefore be interpreted with great caution. The volumes of water concession for hydroelectric generation represents between 80% and 90% of water granted. For more information see National Institute of Statistics and Census: http://www.inec.go.cr/, Directorate of Water: http://www.da.go.cr/, and National Meteorological Institute: http://www.imn.ac.cr/.

Latvia: Data are estimates

Lithuania: The large decrease in total abstracted water in 2010 is due to the decrease in the production of power stations: a power plant, using large amounts of water to cool its atomic reactors, has been closed. Public water supplied to small enterprises is reported as "total water used by industry" (NACE 10-45), and it cannot be broken down by sector

Brazil: The category "industry" refers to ISIC 10-33 categories

Environment Database - Freshwater abstractions (million m3)Variables collected
This dataset shows the state and changes over time in the abstractions of freshwater resources.

Water abstractions are a major pressure on freshwater resources, particularly from public water supplies, irrigation, industrial processes and cooling of electric power plants. It has significant implications for issues of quantity and quality of water resources.

This dataset shows water abstractions by source (surface and ground water) and by major uses. Water abstractions refer to water taken from ground or surface water sources and conveyed to the place of use. If the water is returned to a surface water source, abstraction of the same water by the downstream user is counted again in compiling total withdrawal.

When interpreting those data, it should be borne in mind that the definitions and estimation methods employed by Member countries may vary considerably among countries

Last updated: July 2017

Contact: ENV.Stat@oecd.org

Disclaimer: Data available to the OECD Secretariat by April 2017, and subject to possible slight revisions

Australia: Abstraction by agricultural sector: 2001-07 data are estimates based on national reports; due to changes in methodology, data may not be comparable with previous years. 1985: public supply refers to the domestic sector

Austria: Partial totals including estimates (gross abstractions in 1990-1995)

Canada: Agriculture 1980-1995: data exclude forestry and fishing

Chile: Only data on public supply is available. This data is based on the information that the sanitary companies provide for the areas where they provide their services

Czech Republic: Up to 1995 water abstractions for fishponds were included (change of methodology - break in series). The drop in total abstractions in 2013 is explained by the lower water abstraction for cooling for electricity production. For more information see http://voda.gov.cz/portal/en/

Denmark: 2009: partial totals for public water supply and manufacturing industry. Irrigation includes fish farming. A rough estimate for abstractions for aquaculture is 180 million m3/year, but data is not published because deemed unreliable. For more information : http://www.dst.dk/ ; http://www.geus.dk/ ; www.statistikbanken.dk/vandind

Estonia: Production of electricity: up to 2001 data refer to total abstractions for electricity production (ISIC 35.1 Rev.4). Since 2001 data refer to the NACE activity 40.1, which means that part of the cooling water is allocated to the "other" category. Public water supply (surface water, 2000-2001): data include a high share of water use by manufacturing enterprises

France: Total abstractions in 1975: estimates based on four basins out of six. Data refers only to metropolitan France (including Corsica); data for overseas territories is available since 2012 but is not included here (they represent 1.5% of metropolitan France abstractions, including electricity cooling, and 3.5% excluding it). Data are estimated using basin agencies' calculations on the licence fees. Most of water abstracted for agriculture is used for irrigation, but there is no available time series. Data on manufacturing industry includes mining and quarrying, construction and services (the manufacturing industry represents around 80% of this total). Abstractions for drinkable water since 1990 have first increased then stabilised and finally decreased thanks to improvements in equipment and in behaviour. However, the reduction of leakages in the distribution system (which has a rate of performance of 80%) remains an opportunity for water savings. In the energy sector, water is mostly used for nuclear power plants, and the improvements in the cooling systems have stabilised water abstractions. The industrial sector has managed to reduce its consumption by a third since 2003. Water abstractions for irrigation vary with the weather conditions: depending on the years, tensions can arise locally and cause restrictions and usage conflicts. For more information see http://www.statistiques.developpement-durable.gouv.fr/ ; http://geoidd.developpement-durable.gouv.fr/  ; http://www.hydro.eaufrance.fr/  ; http://www.statistiques.developpement-durable.gouv.fr/lessentiel/s/ressources-eau.html

Germany: Totals up to 1998 exclude all agricultural uses except irrigation. From 2001 totals include mining and quarrying. Before 1991 data refer to western Germany only. Some of the water used for the production of electricity might be obtained from other industrial sectors or from public water supply. Total surface water abstraction and total ground water abstraction may not add up to total freshwater abstraction because abstraction by private households cannot be broken down to ground and surface water (only the total abstracted amount can be estimated).

Greece: Partial totals excluding agricultural uses besides irrigation (agriculture includes only irrigation)

Hungary: Break in series in 2000: change in data source ("Water resources fee" database instead of the "Report on industrial water uses"). The large share of freshwater abstracted for electricity cooling is due to a nuclear power plant

Ireland: 1994:  estimates including 1980 data for electrical cooling. Agriculture includes only irrigation. Break in series in 2005. The methodology has changed and there is no data for sectors. Abstractions are based upon the amount of water supplied by treatment plants. Figures cannot be verified e.g. against metering data, yet. The data is supplied by local authorities and the national water utility. There is ongoing work to review and update the dataset in order to improve the accuracy of the data. Any detectable trend of increased abstraction may be due to a more complete dataset for 2014 rather than to any true increase in abstractions. Annual drinking water report available at http://www.epa.ie/pubs/reports/water/drinking . Historic information on drinking water monitoring results and water supply details for each county (dating back to the year 2000) is available on the EPA's SAFER (Secure Archive for Environmental Research Data) web-page at http://erc.epa.ie/safer/resourcelisting.jsp?oID=10206&username=EPA%20Drinking%20Water 

Iceland: Public supply includes the domestic use of geothermal water. Since 1995 fish farming is a major user of abstracted water, explaining the change in the relative contribution of other sectors.

Japan: Public supply: data refer to households and service sector. Agriculture excludes forestry and fishing. Manufacturing industry includes electricity cooling

Korea: Public supply refers to the domestic sector only (households and the commercial sector, excluding the agricultural and industrial sectors). Agriculture includes only irrigation. Break in series in 2013: for groundwater the Jeju-do area has been included, and so data is not comparable to earlier years. For more information see: National Groundwater Information Center ( http://www.gims.go.kr/), Water Resources Management Information System (http://www.wamis.go.kr/).

Luxembourg: 1989: estimates including 1983 data for industry and electrical cooling. Agriculture includes only irrigation in 1995-1999. Further references: http://www.eau.public.lu/

Mexico: 2001 onwards: volumes of water granted in concessions (administrative figures, not collected data); prior data are estimates. "Manufacturing industry" figures refer to all industries (not only manufacturing) and include some services. Hydroelectricity generation includes in-situ uses (figures for individual uses are not available). For more information: http://www.conagua.gob.mx/ , http://www.gob.mx/semarnat/acciones-y-programas/sistema-nacional-de-informacion-ambiental-y-de-recursos-naturales

Netherlands: Before 1980 data include marine waters. After 1980 data exclude underground flows (estimated at 2 billion m3). Partial totals excluding all agricultural uses before 1990. Manufacturing industry in 1976 includes mining and quarrying. For more information: http://statline.cbs.nl/Statweb/publication/?DM=SLNL&PA=82883NED&D1=a&D2=0,23,6,39,41&D3=a&VW=T , http://www.cbs.nl/nl-NL/menu/themas/natuur-milieu/publicaties/milieurekeningen/publicaties/archief/2014/2014-environmental-accounts-of-the-netherlands-2013-pub.htm

New Zealand: Data exclude storage water (dams and lakes). 2010 figures are based on an estimated water abstraction of 50% of water allocations. This is based on the average consumption in all regions excepting Southland, where the 16,000 m3 allocated to hydroelectricity - 60% of the total national allocation - skews the national average

Norway: Freshwater abstractions: since 1996 data include water abstractions for aquaculture. Totals include estimates.

Agriculture and irrigation: these series have been discontinued in 2007. Manufacturing industry: this series has been discontinued in 2010

Poland: Totals include mining and construction water discharged without use. Data include abstractions for agriculture, which refer to aquaculture (areas over 10 ha) and irrigation (arable land and forest areas greater than 20 ha). Animal production and domestic needs of rural inhabitants are not covered (self-supply). Aquaculture: water abstraction for filling fishponds exceeding 10 ha. Industry cooling includes marine water. For more information, see: http://www.kzgw.gov.pl/ , http://www.pgi.gov.pl/ , http://www.psh.gov.pl/ , http://www.imgw.pl/

Portugal: Data do not include the Azores and Madeira. Total freshwater abstractions in 2013 include 5,6 million m3, and 1.5 million in 2014, of water abstracted from an unknown source. Irrigation: data estimated using mathematical models for calculating soil water balance. It is uncertain how much is abstracted from surface and from groundwater. The estimates reflect the characteristics of the soil and climate, reflected by specific coefficients of the mathematical model, and therefore care needs to be used in making comparisons. The recent decrease in public water abstractions is due to the effects  of the economic crisis and the consequent need for water savings and for a more efficient use. For more information see http://www.ine.pt/ (Statistics Portugal) and http://www.ersar.pt/  (Water and Waste Services Regulation Authority).

Slovak Republic: For more information see https://www.enviroportal.sk/indicator/detail?id=1761 and https://www.enviroportal.sk/spravy/index

Slovenia: Agriculture includes only irrigation. The production of electricity includes Gas and Steam and hot water supply activities. Break in series in 2012: previous values are under-reported (they did not take into account some reporting units)

Spain: Before 2000 agriculture includes only irrigation. After 2000 it excludes the fishing sector. Totals also exclude the fishing sector. 1995 data do not refer to 1995, but represent an average hydrological year, except for those data referring to cooling of thermal and nuclear plants

Sweden: Irrigation: 1985-2004 estimates for dry year. For more information: http://www.scb.se/

Switzerland: Partial totals excluding all agricultural uses. In 2006 they exclude rainwater use. Public supply includes total industry (ISIC 5-43 rev. 4), which totals 215 million m3 in 1994, and other activities (101 million m3 in 1994). In 2006 they are 158 million m3 (total industry) and 84 million m3 (other activities). Production of electricity (cooling) includes non-consumptive water use (cooling on rivers). Data for 2012 contains estimates

Turkey: Break in series in 2008 for public water supply (before 2008 data refer to urban areas only). Totals: Estimates including data based on partial inventories, excluding agricultural uses besides irrigation and, until 1993, electrical cooling. Agriculture: irrigation only. Manufacturing industry and electrical cooling: data are based on partial surveys. 1993 data include abstractions by industry. 1991: partial totals for public supply include manufacture and industry cooling, whereas the grand total excludes them. 2005 and 2008: partial data including estimates based on previous year data for abstraction by industry and electrical cooling.

England and Wales: Series break in 1991 and 1999 (significant changes in reporting methods and classifications). Figures are therefore not strictly comparable with those in previous/intervening years. Data for 2014 refers to England only. Public water supply includes estimations. Agriculture and irrigation: Under the Water Act 2003 abstraction of less than 20m3/day became exempt from the requirement to hold a licence as of 1 April 2005. As a result over 22,000 licences were deregulated, mainly for agriculture or private water supply purposes. However, due to the small volumes involved this has had minimal effect on the estimated licensed and actual abstraction totals Estimates for irrigation are solely for spray irrigation. Return requirements, which set out the actual abstraction information requested from abstractors, were standardised across England & Wales from 1 April 2008. Returns are now requested on a financial year basis. Prior to April 2008 returns were either by calendar year or financial year. To account for this, data collection for 2008 was the subject of two requests. At the end of the period January 2008 to March 2008 and at the end of the period April 2008 to March 2009. This may have had the effect of underestimating actual abstraction.

OECD Total: ross freshwater abstractions per capita, as percentage of total renewable resources and as percentage of internal resources are OECD Secretariat estimates. Data exclude Chile and the United Kingdom refers to England and Wales only

Colombia: The data presented for the different years are not comparable, given the different methodology applied and the database used. Production of electricity refers to the water demand for cooling by the thermoelectric sector, which does not include water demand for the generation of hydroelectricity. Amounts of water abstracted from surface and groundwater sources are estimated. Groundwater demand is estimated at 15% of total demand. For more information: http://www.ideam.gov.co/ , http://documentacion.ideam.gov.co/openbiblio/bvirtual/023080/023080.html

Costa Rica: Data includes only legally authorised abstraction. There is a significant amount of water that is illegally abstracted and is not included. It is possible that some concessions that were withdrawn are not included. Concessions granted in previous periods are also not included. Public supply includes (authorised) water abstracted by private households. There might be manual errors in the compilation of the source database on water abstraction concessions. Data should therefore be interpreted with great caution. The volumes of water concession for hydroelectric generation represents between 80% and 90% of water granted. For more information see National Institute of Statistics and Census: http://www.inec.go.cr/, Directorate of Water: http://www.da.go.cr/, and National Meteorological Institute: http://www.imn.ac.cr/.

Latvia: Data are estimates

Lithuania: The large decrease in total abstracted water in 2010 is due to the decrease in the production of power stations: a power plant, using large amounts of water to cool its atomic reactors, has been closed. Public water supplied to small enterprises is reported as "total water used by industry" (NACE 10-45), and it cannot be broken down by sector

Brazil: The category "industry" refers to ISIC 10-33 categories