EUROSTAT
Directorate ESectoral and Regional Statistics
Unit E-4Regional statistics and geographical information
Luxembourg,05.03.2015
WP DOC NR D/GIS/120
WORKING PARTY MEETINGGeographical Information Systems for Statistics – Joint meeting with National Mapping Agencies and National Statistical Offices
to be held in Luxembourg,
building BECH – Room AMPERE
on2 March 2015
Chaired by: Mr Gunter SCHÄFER
Draft Minutes
Working document for Item 2 on the Agenda
Participants:
The list of participants is contained in Annex 1
Contact:
Atanas Trifonov, telephone: +352 4301-36650,
Item 1Adoption of agenda, D/GIS/100, All
Agenda was adopted without comments.
Item 2Minutes of the previous meeting of the working group meeting “Geographic Information Systems for Statistics”, Minutes of previous meeting held on 18 - 19 March 2014, D/GIS/101, All
Minutes adopted without comments.
Item 3A revised mandate of the GISCO working group, The mandate will be adopted by the DIMESA meeting in June 2015, D/GIS/102, D/GIS/103, All
The current and the revised versions have been made available as Room Documents in CIRCABC. The new mandate represents an update of the objectives oft he GISCO working groupandrefocuses itswork on the integration of statistical and geospatial information.Alignmentwith the activities of UN GGIM and a better consideration ofthe policy relevance of the GISCO working group activities are other reasons to change the mandate.
Comments:
The change of the mandate could be an opportunity to include reporting on the activities of the Eurostat Task Force on the integration of statistical and geospatial information and focusing on itsactivities.
New Mandate has been approved but Eurostat will give NSIs an occasion to make further comments in a written consultation.
Session 1: The integration of statistics and geospatial information - European and global initiatives
Item 4,Overview of the initiatives for integration of statistics and geospatial information, D/GIS/104Ekkehard Petri, DG EUROSTAT
ESS Vision 2020 was adopted by the ESSC in May 2014, to prepare the ESS for the challenges of the current economic and social crisis and to support the Europe 2020 strategy.
The program has a strategic orientation towards evidence-based decision making and performance-management of public authorities; new information needs from globalisation and crises management (economic, political, environmental, societal); strive for relevance, timeliness and punctuality; etc.The Implementation of the ESS Vision 2020 ESS.VIP Programme will contain a ESS.VIP portfolio with strategic projects.
ESS Vision 2020 is important for GISCO working group as it s the guiding frame for the development of the ESS, facilitates the communication with statisticians andis aligned with the European Statistical Programme. ESS.VIP.ADMIN on administrative data sources addresses geocoding of population and sample frames (persons, dwellings), providing a link to the geocoding of the 2021 and beyond Censuses and geocoding of surveys. There is a potential stronglink to GEOSTAT 2 project.The integration of geospatial and statistical information could contribute to the implementation of the ESS 2020 Vision e.g. by fully geocoding the Census 2021
Comments:
No
Item 5Report on the activities of the Eurostat Task Force on the integration of statistical and geospatial information, D/GIS/105, D/GIS/106, Ekkehard Petri, DG EUROSTAT
Eurostat Task Force on the integration of statistical and geospatial information has held three one-day meetings until now on Nov 2013, on March 2014 and on Dec 2014. The key topics identified during the discussions are covered by the presentation.
There is the need of agreeing on definitions for "spatial statistics" and "Integration of statistical and geospatial information". Such terms have been used by different communitieswith different meaning. The task force has proposed definitions in line with the ESS scope of activities.
Further discussions and cooperation should be based on reaching a common understanding for the scope and direction of development in the areas of "Data sources", "Spatial statistics products", "Confidentiality" and "Cooperation between NSIs and NMCAs".
The task force has recommended the implementation of two projects 1) Implementation of INSPIRE, possibly based on ELF and 2) Census 2021, based on GEOSTAT 2 as show cases for the integration of statistical and geospatial information.
The task force has elaborated several groups of priorities, among which are the following principles proposed to underlie the future activities of GISCO working group:
- Definition of spatial statistics (see above).
- Classification of data sources resulting in a 3-tier spatial reference framework.
- All countries need a single, geocoded address, building and dwelling register for statistics and administrative data, based on the GEOSTAT 2 recommendations.
- The next census 2021 should be fully geocoded at unit record level.
Comments:
The task force proposes to call statistics produced at NUTS3 administrative level and below spatial statistics as this is the level of territorial disaggregation where the spatial component starts to become dominant. The separation might be rather artificial, but one that could be worked with.
The general principles proposed in the presentation should be agreed to serve as basis for future actions. Data access (Point 3) is one of the items that should facilitate the work and is a keysolution to agree upon.
We need definitions, common framework and a project.
EUROSTAT can launch projects, but they often reach only the statistical offices. It is imperative toreach a wider number of participants.
The next Census is a very important and an interesting case, as it may be used to introduce substantial change. There is unfortunately littletime as the legal aspects need prolonged preparation. It is required to act now. The European Regulation talks only about the outputs and one of the proposed outputs is the geo-grids. The priority as set out by the task force should be to geo-code micro-data of the census, in line with the recommendations to be produced by the GEOSTAT 2 project.Some guidancefrom this group are needed, toformulate a corresponding proposal for the DIMESA meeting in June 2015.
The report has to be condensed and presented to the DEMESA, who will have the mandate to re-discuss the priorities and approve them.
Decision:
The principles have been generally agreed. Eurostat will organise an official written consultation on the report and the main messages to be presented to the DIMESA in March/ April.
Session 2: Merging statistics and geospatial information – putting the integration into practice
Item 6Overview of the 2012-2013 Grant agreements for merging statistics and geospatial information in Member States, Call for 2015, D/GIS/107Hannes REUTER, DG EUROSTAT
Every year since 2012, Eurostat has been launching calls for proposals supporting activities related to the merging statistics and geospatial information in Member States. Projects under two calls 2012 and 2013 have already been completed or are close to completion. The calls target activities carried out in cooperation with NSIs and respond to the ESS action "Merging Statistics and Geospatial Information".
Until this day projects delivering solution on Table Join Service, Resource Descriptor Identifier, Data cycle for Web Mapping, Address Databases, Population grids, Commuting analysis have been completed. These have already been presented to our Working Group.
Six long term projects from the 2012 and 2013 calls remain to be finished. They cover actions promoting 1) precise depiction of distribution of resident population data for the 2001 census, 2) creation of tools to enable Malta's National Statistics office to spatialize, standardize and disseminate, 3) creation of GEOSTAT 1km grid, 4) integration of health care statistical data with administrative / population/ spatial information, 5) development, harmonisation and propagation of an administrative register of small scale statistics, 6) exploration of temporal accuracy of geo-referenced data from Mobile phone operators.
Comments:
Eurostat expresses thanksthe participants highlighting their contributing to the GRID datasets.It has become clear that many activities targeted topics identified by the Task Force. For the 2013 the address matching has been the most important topic, and has become clear that it should be developed further.
Item 7For the first time grids in Croatia, D/GIS/108, Žaklina ČIZMOVIĆ, Branko CRKVENČIĆ, Croatian Bureau of Statistics
The project develops a national population grid datasets with a grid cell size of 1km2, created in line with the methodology developed by GEOSTAT and the EFGS. The activities are based on the 2011 Census data collected from 1 to 28 April 2011. The presentation covered the background, the methodology, the challenges and the major problems solved during the work on the population grid. The approach used to address the data protection problems was explained during the presentation.
Comment:
The shape files containing the house numbers has been produced by the Croatian NMA. This data has been updated as part of the project, using the explained methodologies and an updated dataset has been returned to the NMA. The project has so benefitted the work of the NMA.
Item 8Spatial analysis on web, Integration of statistics and geospatial information, D/GIS/109,Marja TAMMILEHTO-LUODE, Jani KYLMÄAHO, Statistics Finland
"Spatial statistics on web (SSW)" is a project sponsored by Eurostat 2014-2015. The goal of the project is to improve the shared use of statistical and map data by facilitating theanalysingof statistical grid data. The project has developedtoolsfor spatial analysis of statistical grid data and geographic information, for presenting the analysis results and for transferring the data to user’s own data processing systems. NLS is developing the map functionalities needed using OSKARI platform. The beta version of the analysis application can be used in the Map window of the National Geoportal of Finland Paikkatietoikkuna. The project provides an inspiring example of the cooperation between the NMS and NSI of Finland, with well defined roles, formal agreements and committed cooperation. So far the project has delivered an impressive set of spatial analysis tools available to the general public through eh OSCARI platform. The project activities are not yet completed, but well on track to achieving the objectives in time.
Comments:
From within the ELF project there has been an effort to implement a table joining services in the OSCARI platform. Such services have been developed by the Slovenian colleagues and it is the intention to host them under the OSCARI Elf platform.
Item 9Towards an official and collaborative Catalogue of Postal Addresses in Spain, D/GIS/110, Ignacio Duque RODRIGUEZ, National Statistical Institute (Spain)
Three Spanish institutions Map Agency, Cadastral Agency, National Statistical Office, are working to develop a Catalogue of Postal Addresses in Spain. The objectives are to create 1) An official catalogue of georeferenced addresses and 2) An official catalogue of georeferenced streets and roads.
Task Force for technical design has been established and started work in April 2012. It includes members from central government, regional governments, and local administration and INSPIRE experts. The objectives of the Task Force were to develop standardised, INSPIRE compliant technical parameters for the collected data and establish the data model characteristics. The database was launched in November 2011 (Census date) with Census data. The second step of the project was launched in January 2014.
Current work is focused on developing of non structured and geographic queries, updating procedures and on improving the coverage of georeferenced addresses. Future steps will be the addition of INSPIRE services and data model and data interchange of thoroughfare (streets and roads) entities.
Comments:
Addresses are the backbones and statistics. The project is fully open to the administration; the available data are open to the public and other users. The work on the development of this dataset could add experience to the GEOSTAT project.
Item 10Multiyear migratory flows – experience from Italy, D/GIS/111, Alessandro CIMBELLI,ISTAT
The project had three general goals: 1) Identify, map and analyse migratory flows by means of normalization and geocoding of birthplace names registered in the residence permits database; 2) test and use software for data normalization (openRefine) and spatial open dataset (GeoNames, GADM); 3) understand how to enhance the quality of information of place of birth
Two releases of the residence permits database of the Ministry of Interiors (2012, 2013) were used to process, map and analyse the migratory flows for 5 studied countries (China, Egypt, India, Philippines, Ukraine). Normalisation, geocoding and mapping built up the core of the methodology.
Very good results for most of the countries (4/5) and the results are confirmed by demographic analysis. On the negative side very bad results were achieved for Ukraine as in 50% of “place of birth” is given as "Ukraine" and often the place name is referred to "adm" level and in other case to a "ppl".
The work in the following year will be focused on the publishing of the results as OGC services, analyses of the 2014 release (and 2015, if delivered in time) and a study reconciliation process foreseen in OpenRefine.
Comments:
It is very impressive that the study could get the administrative boundaries. Administrative data is openly available, but inconsistent. In some cases it is available down to the third administrative level.
Item 11Calculating commuting distance – a straight forwardprocess?D/GIS/112, Ingrid KAMINGER,Statistics Austria
The primary goal for the project "Census 2011 - Enriching Commuter Statistics" was to model the variables road distance and commuting time using georeferenced register data and a road network.
The motivation behind the project was the change to register based Census in 2011 and the fact that most commuting variables published in previous Censuses are not included in administrative data sources (commuting time, commuting distance, commuting frequency, means of transport used for commuting) and for the Census 2011 publications very coarse estimates for some of these variables were derived from a testing file of a LAU2-to-LAU2 matrix
The project developed calculation methodologies and delivered the results for the 2011 Census and 2012 data. The algorithms used for 2011 may be reused. Potential benefits for the ESS are 1) STAT will document the work, 2) STAT will provide the model and routing script and 3) countries having a similar data basis could organise the routing system for their country and could be able to derive comparable results. The national benefits are the routing network and technical infrastructure, data for „local“ commuters (within municipality or Vienna district) which were not available before and much more precise distance for short distances (<30km).
Comments:
Labour market areas data are a topic of great interest and this type of studies should be carried further.
Item 12Merging statistical and geospatial information – call for proposals 2015, D/GIS/113, Hannes Reuter, DG EUROSTAT
We are pleased to announce the new, 2015 Call for proposals for merging statistics and geospatial information. Its legal basis is the Regulation (EU) No 99/2013 of the European Parliament and of the Council of 15 January 2013 on the European statistical programme 2013-17.
The Call for proposals covers topics aiming to: improve the integration of geospatial and statistical information, target processes required for a (semi) automated regular update, promote stable and unique national identifier systems, link geo- and statistical information providing additional value, create new information (open data initiative, Big Data streams), develop web applications to show the spatial distribution of statistics.
Total available ESTAT funding for the Call for proposals is 450 000€. Call constraints are: co-financing 70%, mono-beneficiary, duration up to 24 months with shorter time periods preferred.
Comments:
The 70% rule for co-funding is a problem for NGOs and SMEs, but the Eurostat grants are subjected to the Financial Regulation of the EU. In general the grant is not aiming to fianceNGOs. If there is an interesting topic the NGOs have to team up with Member States.
We will like to have the input by early July and be ready with the evaluation by end of September.
Session 3: Evidence-based policy making using geospatial information
Item 13Analysis of the EU-funded investments in airport infrastructures, D/GIS/114, Luc T’JOEN,EU Court of Auditors
The study[1]has been undertaken to answer the question: Were Cohesion policy investments in airports needed, well-planned, well-managed, and cost-effective? A total of 20 airports have been studied in Spain, Italy, Greece, Estonia and Poland representing a total investment of 1,233 Mln Euro.
The study used road data (2009 TeleAtlas road network), population data based on the 2006 population grid and tourism data from 2006, derived from nights spent in tourist accommodations and number of beds to produce airport catchments data. It analysed further the number of residents living in the overlapping areas of catchment areas by applying a uniform criterion of 120 minutes’ driving distance from the audited airport.
The results of the study showed few qualitative benefits, poor sustainability (Cumulated loss of almost 1 billion € (2000-2013) for all 20 airports, only 4/20 regularly made profits, 7/20 do not have prospects of breaking even in the future). Little impact has been achieved only 4/20 airports demonstrated even limited job creation linked to EU investment and 3/20 airports had generic studies on positive effects from the location of an airport.
The study has been very well received and the successful approach will be replicated in a study of the EU investment in maritime ports.
Comments:
The upfront analysis is very important and much more useful than ex-post analysis.The use of catchment method for planning of airports has been recommended for usage in infrastructure planning.Norway uses AIS data for identifying used and unused sea ports.
Such type of analysis will play an increasing role in evidence-based decision making. The quality of the input data for such analysis is vital. Only with quality statistical data from Member States such as the GEOSTAT population grid the results accurate results can be produced. It is therefore the genuine interest of Member States to provide the best available data to avoid distortions from poor quality information, e.g. from modelling and disaggregation.