Earth observation

for decision-making Earth observation data is a unique source of commensurable information. It can be combined with administrative, social and economic data at multiple scales for in-depth policy analysis.

The OECD is currently working with earth observation data providers and key partners to develop its geospatial data capacity.

CONTACT

Environmental Performance and Information team

Head of Division Nathalie Girouard

Senior Economist Ivan Haščič Ivan.Hascic@oecd.org

Statisticians Alexander Mackie and

Miguel Cárdenas Rodríguez

Communications Clara Tomasini

Clara.Tomasini@oecd.org

Image credits: USGS/ESA; Copernicus data/ESA; ESA/ATG Medialab. This document and any map included herein are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.

http://oe.cd/env-data

San Francisco Bay Area, Landsat-8, ESA.

March 2017

Targeted and efficient environmental policies need a strong evidence base that accounts for the geographical distribution of environmental phenomena and economic activity. The existing evidence base of environmental policies has traditionally suffered from data gaps or incoherent time series.

Earth observation from satellites, aircrafts and drones, in-situ measurements or ground-based monitoring stations, can provide a unique and timely source of data that is commensurable across countries, regions and cities.

It can help harmonise international reporting on natural resources, ecosystems and environmental sinks.

It can be combined with other geo-referenced socio-demographic, economic and public administration data to make indicators and analysis more relevant and targeted.

Earth observation is not new, but it is only recently that investments in satellite capabilities, open and free access to data and tools, and advances in algorithms and data processing have started to enable the widespread use of this information at scale, and beyond the specialised scientific community. These developments offer opportunities for improving the range and robustness of environmental data and indicators.

3

Tianjin, China. Combination of 3 radar scans, Copernicus 2015.

A unique source of information

54

Earth observation data can contribute to more detailed and more harmonised indicators, without requiring any additional reporting by countries.

This map is produced using the Joint Research Center’s Human Settlement Layer: a global,

Less than 20%

20-40%

40-60%

60-80%

Greater than 80%

Increase in built-up area 1990-2014

Source: JRC HSL (2016); FAO GAUL (2014).

Earth observation can provide insights into phenomena that are otherwise impossible to measure.

high-resolution (38m), multitemporal dataset which measures the extent of built-up areas, and the time period in which they were built. This newly available dataset spans a time series from 1975 and will be updated yearly from 2017.

Here, the information on built-up areas is summarised by administrative regions. This data can be used to assess the impacts of land use policies, and better understand local, regional and global patterns of urbanisation.

Population exposure to PM2.5, μg/m3

(excluding estimated contributionof mineral dust and salt spray)2015

Less than 66-1212-2424-48Greater than 48

76

Global insights the example of air pollution

The measurement of population exposure to fine particulate matter (PM2.5) can be harmonised across all countries.

Recent work used pollutant concentrations derived from satellite observations, ground monitoring stations and chemical transport models, and calculated exposure by weighting pollutant concentrations with populations in each grid cell of the underlying datasets. This method represents major progress compared to traditional data collection methods.

With this information, pollution abatement efforts can be focused on areas where exposure is highest.

Information on population exposure to air pollution draws on merging satellite data, in-situ measurements and demographic data.

Estimates for chronic (annual average) outdoor exposure to PM2.5 expressed in micrograms per cubic metre. Top: total exposure to PM2.5. This map takes into account all sources of PM2.5, which is most relevant for assessing the health consequences of exposure. Bottom: Exposure to PM2.5 excluding the mineral dust (for example, sand in the deserts) and sea salt component. This map highlights more directly anthropogenic sources. However, human activity can also play a significant role in dust emissions (e.g. through agricultural practices).Source: OECD Green Growth Indicators 2017 (forthcoming). FAO GAUL (2014).

Less than 10

10-15

15-25

25-35

Greater than 35

Population exposure to PM2.5, 2015

8 9

Sources. Donkelaar et al (2016); OECD FUA boundaries; ESRI, USGS, NOAA and FAO GAUL.

Sub-national focus Time series

Mexico cityToluca

Querétaro

Puebla

Xalapa

Celaya

Tehuacán

Orizaba

Pachuca de Soto

CuautlaCuernavaca

Guanajuato

MoreliaTexcoco

Poza Rica de Hidalgo

San Juan del Río

Uriangato

Iguala de la Independencia

Chilpancingo de los Bravo

Santiago

Rancagua

Curicó

Linares

San Fernando

Melipilla

Valparaíso

San Antonio

Chillán

Milano

Ulm

München

Torino

Basel

Innsbruck

Augsburg

Zürich

Bern

StuttgartStrasbourg

Ingolstadt

Trento

Bolzano

Verona

Asti

BresciaNovara

Freiburg im Breisgau

Padova

Piacenza

Rosenheim

Reutlingen

Cremona

Mulhouse

Sindelfingen

Kempten (Allgäu)

Ferrara

Tokyo

Osaka

Nagoya

Niigata

Toyama

Sendai

Kofu

Iwaki

Nagano

FukuiMatsumoto

Shizuoka

Kanazawa Maebashi

HamamatsuToyota

Koriyama

Tsu

Fukushima

MitoUtsunomiya

Toyohashi

Yamagata

Yoshiwara

Fujieda

Numazu

Takaoka

Yokkaichi

Mexico Chile

Eastern Alps, Europe Japan

Population Exposure to PM2.5 2015 (µg/m3 )Less than 10 10-15 15-25 25-35 Greater than 35

Percentage change in population exposure to PM2.5, 1998-2015Decrease

60 45 30 15 0

Increase

The same measurements are conducted at regular intervals, permitting consistent time series and analysis.

Concentration estimates can be overlaid with population density data to produce indicators at finer scales, such as at the level of urban areas.

Traditional environmental data has long suffered from data breaks, due to changes in reporting methods, and from gaps (missing information). Earth observation provides consistent time series to compare different periods of time, and to derive trends.

Milano

Ulm

München

Torino

Basel

Innsbruck

Augsburg

Zürich

Bern

StuttgartStrasbourg

Ingolstadt

Trento

Bolzano

VeronaBresciaNovara

Freiburg im Breisgau

BiellaPadova

Piacenza

Rosenheim

Reutlingen

Pavia Cremona

Mulhouse

Sindelfingen

Tübingen

Vicenza

Kempten (Allgäu)

LuganoLecco

WinterthurSt. Gallen

Luzern

Como

Vigevano

Bergamo

Percentage change in population exposure to PM2.5, 1998-2015Decrease

16 12 8 4 0

Increase

1110

Combining earth observation data with other geospatial information

Earth observation data is collected via satellites and aircraft (1, 4, 5 and 6), and in-situ measurements such as air-pollution monitoring stations.

Geospatial data is a broader term, typically used to include any data that is explicitly associated to a specific location, such as cadastral data or administrative boundaries (2 and 3).

There is a wealth of information to gain by combining high- and medium-resolution datasets from earth observation, with georeferenced administrative or census data.

These six images show the same area: a border of the old-growth rainforest in the Taman National Park, near the city of Palangka Raya, in Indonesia.

For example, the datasets shown can be used to quantify and identify patterns of agricultural land use (4), urbanisation (5) or forest change (6), within and around the boundary of the protected area (3). They can better characterise the effects of protected area designation or other land-use policies on ecosystems and local economies.

Earth observation data can be combined with geo-referenced public administration information, to improve the policy relevance of indicators.

6 -Forest cover

5 - Built-up area change

4 - Cropland

3 - Protected area

2 - OpenStreetMap

Sources: Hansen et al Global Forest Change (2013); JRC HSL (2016); ESA-CCI Land Cover 2015; UNEP; WCMC World Database of Protected Areas (2016) Esri, DigitalGlobe, GeoEye, Earthstar Geographics, CNES/Airbus DS, USDA, USGS, AeroGRID, IGN, and the GIS User Community OSM & Contributors 2017

1- Photograph

1312

Geospatial data in OECD work

COUNTRy REvIEWS

CROSS-CUTTING WORk

POLICy ANALySIS

Earth observation data is now routinely used in country reviews and horizontal work. It also supports policy analysis.

0-10 μg/m3

10-15 μg/m3

25-35 μg/m3

ASSESSMENT AND RECOMMENDATIONS

OECD ECONOMIC SURVEYS: TURKEY © OECD 201648

Turkey also faces important water management issues. Water competition across

sectors is growing and is expected to become more challenging with increased

urbanisation, expansion of irrigation areas (Turkey is the only OECD country planning to do

so) and climate change. Pressures from agriculture are particularly strong, as it represents

85% of freshwater withdrawals – the second-highest proportion in the OECD (OECD, 2013b).

In the face of these challenges, Turkey’s efforts to upgrade water management along the

lines of the EU Water Framework Directive are welcome. However, progress with

transparency is needed to support monitoring and contribute to the implementation of

reforms. For example, groundwater use is growing disproportionately and calls for

improved surveillance. Without adequate information systems it will not be possible to

monitor and manage depletion rates and external effects (OECD, 2015f).

Bibliography

Acar, S., L. Kitson and R. Bridle (2015), Türkiye’de Kömür ve Yenilenebilir Enerji Teşvikleri (Coal andrenewable energy subsidies in Turkey), International Institute of Sustainable Development,Winnipeg.

AFAD (2013), “Türkiyedeki Suriyeli sığınmacılar: Saha araştırması sonuçları (Syrian refugees in Turkey:Findings from field research)”, AFAD, Ankara.

Aghion, P. et al. (2015), “Innovation and top income inequality”, NBER Working Papers, No. 21247.

Ahrend, R., A. Goujard and C. Schwellnus (2012), “International capital mobility: Which structuralpolicies reduce financial fragility?”, OECD Economic Policy Paper Series , No. 2, OECD Publishing, Paris.

Alper, Y., Ç. Değer and S. Sayan (2012), “2050’ya Doğru Nüfusbilim ve Yönetim: Sosyal GüvenlikSistemine Bakış (Social Security system in the light of demographic trends to 2050)”, TÜSİAD,Istanbul.

Andrews, D. and F. Cingano (2014), “Public policy and resource allocation: Evidence from firms in OECDcountries”, Economic Policy, April.

Atun, R. et al. (2013), “Universal health coverage in Turkey: Enhancement of equity”, The Lancet,Vol. 382, No. 9886.

Figure 24. Air pollutionPopulation exposure to PM2.5, 2013

Source: OECD calculations based on data from Brauer et al. 2016.1 2 http://dx.doi.org/10.1787/888933389016

WHO safety norm

0

10

20

30

40

50

60

CHN

IND

KOR

ITA

CHL

TUR

POL

JPN

DEU

OEC

D

MEX

ESP

USA

SWE

AU

S

µ g/m3

A. International comparisons

0

5

10

15

20

25

30

Ista

nbul

Gaz

iant

ep

Burs

a

Ank

ara

Kony

a

Koca

eli

Kays

eri

Agr

i

Izm

ir

Ant

alya

Zong

ulda

k

µ g/m3

B. Turkey’s regions

20131990WHO guidelines

“Industrial regions display particularly high pollution levels and a particularly low improvement between 1990 and 2013. Similar to GHG emissions, a firmer mitigation effort is needed for fine particles. This would necessitate progress in the transparency of emission sources and formalisation would help.”

OECD Economic Surveys: Turkey 2016

“Poor air quality remains a major public concern across Chile, particularly in large metropolitan areas, in the surroundings of large industrial and mining sites and in cities where wood burning prevails.”

The 2016 Environmental Performance Review of Chile recommends implementing pollution prevention and decontamination plans in all areas that exceed air quality standards and improving air quality networks and access to air pollution information.

“The quality of our local living environment has a direct impact on our health and well-being.” For this reason the OECD Better Life Index takes into account the share of the population exposed to concentrations above the World Health Organization limits of 10 micrograms of PM2.5 per cubic meter of air.

The forthcoming Green Growth Indicators 2017 use earth observation data analysis for indicators on population exposure to air pollution, the extent of natural vegetated land, protected areas, land use change, forest change, etc.

What is the impact of urban structure on the environment and on people’s well-being? Spatial Planning INstruments and the Environment (or SPINE) is an OECD project to assess the environmental and economic effectiveness of spatial and land use planning instruments, as well as the potential gains from policy reforms.

1514

Collaboration Opportunities

The OECD is working closely with earth observation data providers and key partners to develop, combine and analyse geospatial data.

The OECD Space Forum was established to assist governments, space-related agencies and the private sector to investigate space infrastructure’s economic significance, its role in innovation and potential impacts for the larger economy.

OECD analysts increasingly use earth observation data to support their policy recommendations. In 2015, the OECD joined the Group on Earth Observations, a partnership of more than 100 national governments and 100 participating organisations, as an observer.

The OECD is working closely with earth observation data providers and partners: the United States NASA, the European Space Agency, the Global Observation of Forest and Land Cover Dynamics (GOFC-GOLD), and the academic community in an effort to integrate results of frontier research into better policy guidance and provide feedback to the data providers to guide future investments.

The OECD Working Party on Environmental Information works with environment ministries, environment agencies and national statistical offices to develop internationally harmonised methodologies for new and improved indicators. Reliance on earth observation data has gained momentum. Population exposure to PM

2.5, based on geospatial data, has become a green growth headline indicator.

There are key opportunities for the improvement of environmental information in two areas of particular importance:

z Monitoring natural resources such as land, soils and oceans and environmental sinks, including air and water pollution. Earth observation data can offer a wealth of information on water quality, soil carbon and soil moisture, and the oceans. The analysis currently applied to PM2.5 pollution could be extended to many other pollutants.

z Combining geospatial data with other datasets to better assess environmental risks and the associated costs of exposure of humans, ecosystems, built property and economic activity to pollution, natural hazards and industrial risks. Earth observation data could be further combined with administrative and socio-economic data. For example, OECD teams have combined data on air pollution with household income.

Making best use of these opportunities will require voluntary contributions from countries to support specialised staff.

ESA’s satellite Sentinel 3