GEOGRAPHICAL INFORMATION SYSTEM STUDIES AT THE STATE INSTITUTE OF STATISTICS OF TURKEY


1. Introduction

Geographical information system (GIS) and remote sensing (RS) had been one of the key subprojects envisaged in the National Information System project of SIS [1,2,3]. GIS is a location specific description of the world around us which can be seen as a very complex system envisaging physico-ecological and socio-economic sub-systems that always interact. The attempts of a digital description of that world create a computerized GIS which is usually a partial description of the world in relation with some feature tasks.

There are many factors that have impact on the success of the implementation of a GIS. Among those factors, the most important ones are system objectives, procedures, data, organization of system management, and people. This means that there are both t echnical and non-technical factors to be considered by those who want to plan and run such systems efficiently and effectively [4].

2. Recent GIS Developments in Turkey

During the recent years, GIS and RS have gained a momentum in Turkey. Universities, research organizations, private sector and public sector organizations have initiated programs in GIS and RS. Therefore technical knowledge and knowhow about GIS is ste adily improving. This is a promising situation and it can rapidly be transformed into a significant national asset if appropriate investment is made into hardware,software and especially manpower. Success will also require a government initiative in data management and in the integration of data flows within and between agencies. Many agencies use remote sensing data to provide various levels of input to the GIS database. It is clear that coordination of these activities would be beneficial to all sectors of government and the adaptation of national standards is needed [5].

Within the Turkish Universities there are some GIS and RS programs that are associated with the faculties, colleges and departments of engineering and a few that are associated with some specific application fields.

The private sector also has been getting organized to meet an expected demand for GIS/RS equipment and services. There has been considerable investment in hardware , in personnel, and this requires a growth in the market if the capacity is to be produc tive. Vendors for integrated systems with GIS software is already established in Turkey. The success of their commercial activity will depend on the level of governmental concern with geographical data collection and analysis.

GIS and RS are valuable tools in data management and data collection, respectively. Since both are inherently geographical, they have major roles in the regional and national planning studies. This is so far understood by major government agencies, and the leading universities in Turkey [5].

3. SIS Activities

3.1. General Information on SIS

A GIS provides a powerful means for mapping and analyzing many types of information which is obtained from conventional sources, such as census data, socio-economic information, questionnaire surveys and government records. State Institute of Statistic s has exclusive responsibility for producing and disseminating official statistics in Turkey. This information can be grouped as follows:

Given the wide spectrum of data to be collected and analyzed, GIS become one of the most crucial tool to be used in an official statistics bureau. However, many critical factors need to be considered for successful implementation of the technology in o fficial statistics activity. Actions must be taken to improve to profit and to minimize the negative aspects, notably by reducing costs and duplication of efforts [6].

3.2. GIS Strategy of the SIS

As mentioned above, Turkey has already entered the family of nations who started the GIS implementation at several organizational levels. The State Institute of Statistics (SIS) of Turkey assumes the coordination task of data and information flow at th e national level. Thus, a GIS implementation and coordination task has bearings on SIS. Initiated from a current agricultural crop yield estimation project, SIS has established a geographical information system within its own structure [10].

GIS development for large regional area can be realized through projects with small and manageable regional coverage. This is due to the fact that countrywide statistical GIS applications can be achieved best by gradual improvements through regional co verage increased in time.

For the purpose of creating a geographically referenced address reference that covers the addresses of all types of data sources, a systematic approach is required. The objectives of such a system are derived from the users that make use of the result s. The data type, the way data are presented, and the time frame of available data are important to users. In a multi-user environment, the diversity of user needs gives rise to multi-objectives that should be set as aims of the short,- immediate-, and lo ng-term plans for the system at operation or production mode. Obviously, the user needs are not stable. SIS's objectives about the system are as follows [6]:

  1. Long-term Objectives
  2. Intermediate-term Objectives
  3. Short-term Objectives

Attempts of combining the created inventory by the programs that collect the data mentioned above give rise to some important issues to be considered:

3.3. Ongoing Projects

Geographical Information Systems Division was established at the State Institute of Statistics in 1991. In addition to conducting studies to achieve the above mentioned objectives, GIS division is giving support to remote sensing, water statistics and natural accounts activities. The ongoing projects are summarized below.

3.3.1. Water Budget

In 1994, a new division was activated on water statistics and natural resource accounts in order to accelerate studies on establishment of environmental statistics database. The aim of the SIS in starting with natural resource accounting was to provide information to decision makers about depletion and degradation of natural resources both in quantity and quality as a result of various economic activities.

Water, biologic diversity and soil erosion are the major environmental problems in Turkey as far as the natural resources are taken into consideration. Turkey's priorities concerning water is ostensible among the others because of data availability, ge o-strategic importance, scarcity and heterogeneous distribution of the water resources in the Mid-East region.

In inland water accounting the smallest unit will be chosen as the river basin (Figure 1). If we consider a basin as a system; ''water in'' is the precipitation and the inflow of surface and ground water flow from other catchment areas, and ''water out '' is the evaporation, transpiration and outflow of surface and ground water to other catchment areas. There is an equilibrium between them known as the hydrologic cycle. Hydrologic cycle is determined by physical structure (land cover, slope, geological structure, land use, etc.) and climate (precipitation, temperature, moisture, wind, etc.) in each catchment area. GIS will be used in determining water potential with high accuracy and formation of effective water management.

In a pilot project quantity and quality of water resources will be examined considering economic, ecological and sociological functions of water such as the use and discharge of various activities for surface, ground and thermal water resources, etc.

The following data will be collected on:

  1. Water resources,
  2. The quality of surface waters,
  3. Water abstraction and consumption by major activities,
  4. Discharge and load of waste water by major activities,
  5. Available treatment plants capacities and efficiencies waste water treated and the resulting effluent both in quantity and quality [7].

3.3.2. Crop Estimation Project Using RS

As country where the crops play major role in the agricultural sector, reliable and rapid estimation of the crop yield is one of the major issues for Turkey. Timely and reliable forecasts about crop yield will enable the decision makers to take precaut ionary actions for food import and export policies of the country.

For the purpose of determination of the agricultural structure of the country, SIS conducts agriculture census every ten year. In addition, yield estimates are determined by the Ministry of Agriculture on a yearly basis. General Census of Agriculture i s based on the declarations of the respondents and therefore the results are controversial. Furthermore, the costs of obtaining those results are extremely high. On the other hand, the yearly estimates reported by the Ministry is based on subjective eval uations and always suspected for their reliability. Having all these shortcomings, it is inevitable for Turkey to establish an Agricultural Crop Monitoring System which will provide reliably and timely information with a reasonable cost. In 1991, SIS was appointed with the project named ''Crop Yield Estimation via Remote Sensing'' by State Planning Organization of Turkey.

The main objectives of the project is to establish a rapid, reliable and economic system which will

With a "statistical modeling" approach depending on area sampling, sample areas representing the agricultural structure and variations of its relevant features are chosen and information obtained from these areas are used for statistical forecasts abou t the cultivated areas and thus for the crop yield for whole country. The SIS initiated such an agricultural statistics system within the context of Geographical Information System and Remote Sensing activities. In order to draw random sample of fields, l and cover stratification of Turkey (Figure 2) was established with satellite images and ARC/INFO was used to construct the spatial database, and their relationships with provincial boundaries [8,12].

Within the context of the project extensive ground surveys have to be constructed for collecting reference data in order to obtain better classification of the images. GIS is also used in the management of ground data and as a result of integration wit h the image processing software used (ERDAS Imagine), it facilitates the classification of the images [12].

3.3.3. Address Register and Management System

The conventional setup of data collection practices of official statistics system comprises of geographic support for censuses and surveys. Hence, development of GIS in such institutions can be conveniently started in connection with census geography w here frames for census and surveys can be found as addresses of several entities such as households, business enterprises, parcels, land sites and the like [6].

Data inventory of a statistical system is based on some address frames. Therefore, GIS development in official statistics start with introducing geographical referencing to data. This costly and tedious job creates statistical information with spatial context. Construction of required database is costly, needs long time and technical knowledge.

SIS is bound to increase the accuracy in data collection, production and dissemination. Therefore, a pilot project is embarked on construction of geographically referenced address database system. The major goals of the project is to construct an area frame for statistical surveys and to publish data in block base.

In order to achieve the above mentioned objective, five sub-city districts have been digitized and address registers of dwellings have been constructed. The data file of addresses and polygon coverage of buildings are linked with each other and addre sses were geographically referenced with ARC/INFO software. KVR-1000 satellite image was used to determine the area changes between the time interval that base map was produced and satellite image was taken (Figure 3) [9] .

On the same subject, in an other pilot study, street networks were constructed and address ranges were defined for streets. Then, address registers were matched on both sides of streets according to their house number. This pilot project has proved us that the most critical problem to cope with is finding an up-to-date base map. This problem can be solved by acquiring high resolution satellite images which will be commercially available in the near future.

3.3.4. Thematic Map Production

Thematic maps are used in illustrating the class characterizations of a particular special variable. In statistical organizations GIS is used extensively for the purpose of thematic map production. Increasing number of maps in SIS publications are now produced by GIS division (Figure 4). With the ability of integrating the GIS software with statistical packages it will be easier to produce better looking statistical maps [11].

Thematic maps are also available on the Internet WWW server of the SIS.

4. Conclusion

In this paper the long, medium and short term strategies of SIS relevant to GIS and RS have been given and the studies that have been constructed in the last few years have been summarized.

Switch from manually handled map and standard statistical tabulations to databases incorporated with digital maps is a long and painstaking process. Some efforts of informatics people should be spent on demonstrating the benefits of GIS to data colle cting and data disseminating units of a statistical institution.

5. References

  1. SIS, "The Feasibility and Cost Benefit Study Report for SIS National Information System and Statistical Infrastructure Development and Policy Project", State Institute of Statistics (SIS), Ankara, October 1993.

  2. Güvenen , O., Z. Aktaş and H.M. Aral, "Restructuring the Informatics Infrastructure of the State Inst›tute of Statistics of Turkey", ISIS'94, Bratislava, Slovak Republic, 16-20 May 1994, LES/SEM 33/R 34-

  3. Güvenen, O. and Z. Aktaş, "National Information System (NIS) Project of State Institute of Statistics (SIS) of Turkey", Proceedings of ISI 50th Session, Beijing, 21-29 August 1995, pp. 441-442.

  4. Gebizlioğlu, Ö. L., "Perspectives of the Turkish Geographical Information System", Proceedings of the Third Independent Conference of IAOS / ISI, pp:2/2, 1-8, 1992.

  5. Falconer, A., "Development of a Program for Training in Geographic Information Systems and Remote Sensing Technology for Turkey", p:11-24, 1994.

  6. Gebizlioğlu, Ö. L., H. M. Aral and N. Teksoy, "Impact of Remote Sensing on Official Statistics", The Meeting of Selected Statistical Experts Preparatory to the Third Meeting of the Heads of the National Statistical Organizations of the OIC Member Coun tries, Cairo, Arab Republic of Egypt, 30 October-1 November 1995.

  7. Egemen, N., "Summary Report on Present and Planning Studies on Natural Resource Accounting of Turkey", Workshop on National Resource Accounting for Islamic Countries, Ankara, Turkey, 12-14 June 1995.

  8. Gebizlioğlu, Ö. L., H. M. Aral and N. Teksoy, " A Statistical Approach to Crop Yield Estimation via Remote Sensing", Turkish Journal of Physics, Vol:19, no:8, 1995.

  9. Çalış, N., H. Aztopal and H. S. Öğün, "Address Geocoding in GIS and its Application", Research Symposium'93, SIS, Ankara, Turkey, 22-24 October 1993.

  10. SIS GIS Division, "Geographic Information Systems Strategies in Information System and Official Statistics", Proceedings of 2nd Remote Sensing and Its Applications Seminar, Bursa, Turkey, 15-17 May 1994.

  11. Çalış, N., H. Aztopal and H. S. Öğün, "Statistical Data Dissemination Using GIS" Research Symposium'94, SIS, Ankara,Turkey, 21-23 October 1994.

  12. Demirbüken, H., N. Çalış, H. Aztopal, H. S. Öğün, H.A. Gemalmaz and A. Ay, "Estimation of Crop Areas by Remote Sensing and Geographic Information Systems", 10th ARC/INFO European User Conference, Prague, Czech Republic, 2-4 October 1995.

İlk Sayfaya Dönüş       Bir önceki sayfa       webmaster@die.gov.tr