GIS for SUSTAINABLE DEVELOPMENT - PART 3B doc

Cross-Border Region of Hungary and Austria Analyzed with Historical Cadastral and Land Register Data Susanne Steiner CONTENTS 20.1 Introduction ..... The objective of the study, carrie

Learning from Practice: GIS as a Tool in Planning Sustainable Development

Natural and Cultural Heritage

Cross-Border Region

of Hungary and Austria Analyzed with Historical Cadastral and Land

Register Data

Susanne Steiner

CONTENTS

20.1 Introduction 350

20.2 Project Area 350

20.2.1 A Brief Historical Overview 351

20.2.2 Land Management System in Austria 353

20.2.3 Land Management System in Hungary 354

20.3 Methodology 354

20.3.1 Previous Investigations and Results 355

20.4 Data Acquisition 355

20.4.1 Nonspatial Data 355

20.4.2 Conceptual Database Design 355

20.4.3 Spatial Data 357

20.4.4 Modeling the Cadastral Reality 357

20.4.5 Parcel-Related and Ownership Changes 358

20.4.6 Database Processing 359

20.5 Results 359

20.5.1 Land-Use Changes 360

20.5.2 Changes in Parcel Geometry 360

20.5.3 Ownership Changes 361

20.6 Problems Encountered 362

20.7 Outlook 363

20.8 Conclusions 363

Acknowledgments 364

References 364

Many land-use planning decisions ultimately are tied to land ownership and parcels.Therefore, the analyses of past land use dynamics cannot be considered only ashistoric research, but in fact, the development of land and property is an interestingstarting point for decisions in recent land management To serve the demand, forexample, of land planners to base their work on research and historic data, a studywas conducted to show that past developments in land administration can illustratefuture trends For the study, a border region between Austria and Hungary waschosen, where changing agrarian policy conditions and their effects over the inter-vening years can be comprehended and compared.

Border regions provide interesting historical details regarding the structure ofagricultural land, the enlargement of settlement area, and economic and socialsituations An interesting example is the mentioned border region between Hungaryand Austria, where changing nationalities of the area due to historical facts influencedthe economic development The last century, with changing political conditions,shaped the region on both sides of the borderline in different ways Agricultural land

as well as building land was affected by political upheavals, and those changes havebeen registered in land administration documents

The objective of the study, carried out by a bilateral project team consisting ofHungarian as well as Austrian researchers, was to reconstruct and analyze cadastre-based land use (the term “land use” is referring to cadastral determination, whichneed not necessarily correspond with real land cover) and ownership changes in thecross-border region of Austria and Hungary and to quantify the influences of externaleffects like socioeconomic and/or political changes on parcel-based land use andproperty rights The investigations in both regions were based on cadastral and landregister data that were available in a time series of one hundred years Based onderived results, some general statements on the rural development trends for theseparticular border regions could be drawn up

A geographic information system was used for analyzing and visualizingchanges regarding land use and parcel geometry based on a database and on spatialcadastral data

In this chapter the methodology of acquisition, analyses, and visualization of spatialand nonspatial land administration data is described It will be shown how the integration

of two different land administration data as well as historic and recent data sets intoone cross-national database was solved In addition, a model will be presented which

is able to reconstruct a certain point of time in “cadastral history” by imitating theprocess of land registration The last section will provide some results concerning land-use, parcel, and ownership changes in the investigation area, and finally the outlookoffers some future perspectives on the region based on the analysis results

20.2 PROJECT AREA

The cross-border region between Austria and Hungary had been an area withouthinterland for over 40 years The so-called iron curtain created an insuperable barrierwhere social and economic exchange between both regions was no longer possible

[1] Due to political reasons these two border areas developed in very different waysregarding their rural conditions Differences are still visible in the landscape; forinstance the average size of agricultural patterns in Hungary is much higher thanthose in Austrian arable land This characteristic agricultural structure is still recog-nizable (e.g., on satellite or aerial images; Figure 20.1).

The area of two small adjacent municipalities was chosen for comparing andadequate for the research due to their location near the borderline and due tocomparable socioeconomic and demographic conditions The criteria are, on the onehand, comparable number of inhabitants, similar ownership structures as well asdistribution of property and parcel patterns (at least originally in pre-Communistictimes) On the other hand, both regions are considered as rural areas from the regionalplanning point of view, with agricultural land dominating land use, based on similarclimatic and soil conditions

Additional criteria for selecting the project areas Girm in Austria and Harka in

Hungary included the availability of historical and current cadastral maps in terms

of complete time series for one century The investigation area covered building land

as well as agricultural-dominated parts

20.2.1 A B RIEF H ISTORICAL O VERVIEW

Land administration is defined as “the process of determining, recording and seminating information about tenure, value and use of land when implementing land

dis-FIGURE 20.1 Recent satellite image (Landsat TM) showing the agricultural land at the

border area of Hungary and Austria The white line highlights the boundary between scaled patterns of the Austrian territory and the obviously larger fields in Hungary.

small-analyzing ownership and land-use changes (Figure 20.2) They were considered as

FIGURE 20.2 Location of project areas.

management policies” [2]; in other words, land administration is “the process of istering the complex rights, restrictions and responsibilities pertaining to land” [3].Land administration generally consists of spatial data, respectively, the cadastreand nonspatial data known as land register While in the cadastre the parcel infor-mation itself concerning location, area, and land use is stored, the land registerprovides ancillary information (e.g., property rights or debt) related to the parcelsand the ownership.

admin-During the era of the Austro-Hungarian Monarchy (K&K Monarchy) from 1867

to 1918, the land registration was administered by one common cadastral mappingauthority, which basically aimed at creating tax maps of land property Over time,different political and economic developments (First and Second World War, Com-munistic era in Hungary from 1949 to 1989) resulted in different agrarian and landadministration systems and, combined with that, in different land management systems

As a consequence, the different systems in land registration led to diversities inregistration of spatial parcel-based data and ancillary ownership data An interestingaspect of this case study was to find out if and in which way political systems andrelated land management influenced and still influences land use and ownership

20.2.2 L AND M ANAGEMENT S YSTEM IN A USTRIA

The cadastre is kept and maintained by the Federal Office of Metrology and

Sur-veying (BEV) Cadastre and Land Register together form the basis for land

man-agement in Austria These two “spatial related data pools” have always been keptseparately by two different authorities but with high effort on synchronization ofdata The roots of Austrian cadastre mapping trace back to the early nineteenthcentury, when land surveying had been introduced as taxation basis for the wholemonarchy These ancient cadastre maps are maintained in archives and are still usefulfor historical investigations of any kind (Figure 20.3)

FIGURE 20.3 Historical cadastral map of the Austrian test site from 1857.

system [4] A unique parcel identifier for each estate realized the link between spatialdata and land register data that is still kept in a separated database Recently, thecadastral data is already available as GIS vector format including topology (previ-

ously it was simple line data in dwg drawing file format), but there has not yet been

developed a common geographic database for cadastre and ownership informationtogether The land register that represents the legal status of all real property ismaintained and kept by the local courts of law Since 1987, the land register hasalso been available in digital format, but only authorized users such as lawyers haveaccess to these data, in order to protect the data privacy [5]

20.2.3 L AND M ANAGEMENT S YSTEM IN H UNGARY

Land and property registration in Hungary has been operational for nearly one and

a half centuries, having its origins, as mentioned before, in times of the K&K Monarchy.

After all-embracing collectivization processes during the Communistic era, vate ownership of agricultural land was substituted by cooperative and state property,leading to huge agrarian production centers and collective farms These are farms

pri-in which a group of farmers pool their land, domestic animals, and agriculturalimplements, retaining as private property only enough for the members’ own sus-tenance The profits of the farm are divided among its members, in contrast tocooperative farming, where farmers retain private ownership of the land

Today’s land administration consists of the Department of Lands and Mapping(DLM), Regional Development (MARD), and the Land Office Network (LON) Theregistration of ownership and immovable properties has been fully computerized

since 1991 (with the support of the European Commission under the PHARE

pro-gram) The system is open for the public and is backed by state guarantee Additionalservices are the updating of land-use information, classification/evaluation of land,land consolidation piloting, and maintenance of land-related statistics [6]

During the recent economic transition, a major priority of successive governmentshas been to redistribute land from state ownership and from cooperatives to individuals.This process has been managed for agricultural areas partly by the Land Offices andhas placed great demands on the offices, which had to provide information on the pastand present ownership status, carry out definitive surveys, subdivide large plots intomany small ones, and register more than 2.4 million new owners “Land privatizationaffects more than half of the territory of Hungary (5.6 out of 9.3 million hectares) Thenew parcels created during land privatization are scattered all over the country and thismakes it impossible to keep the old cadastral maps up-to-date” [7]

20.3 METHODOLOGY

In this project phase cadastre and land register data from the present land tration systems was acquired and collected The basis for a comparability of differentdata sets is the harmonization, which was solved by the integration of data into onedatabase for both countries

adminis-While spatial data (cadastre) was stored within a geographic database, nonspatialdata (all additional information mainly about ownership) was registered within aseparate relational database Data flow between these two separate databases wasrealized through the use of a unique parcel identifier Based on the resulting com-prehensive data records, all land-use and ownership changes that ever occurred for

a period of one century could be reconstructed, analyzed, and visualized

Spatial analyses and the presentation of visual results were performed by ageographic information system These analyses were topologically based on vector-ized (digitized) cadastral maps, where each parcel represents a spatial object having

a unique identifier

20.3.1 P REVIOUS I NVESTIGATIONS AND R ESULTS

Preliminary studies had been carried out that were considered as prototypes for thestudy The core aims of this pilot project were basically the same (analyzing landuse and parcel changes as well as ownership changes), but focusing on a muchsmaller investigation area and covering only the center and parts of agricultural land

of both municipalities The results and experiences of these previous studies turnedout as basis for further investigations Due to limited temporal and financialresources, the pre-study mentioned was restricted to just fifty years as a researchperiod

20.4 DATA ACQUISITION

Data collection of land register data was carried out at the responsible local andregional public authorities in Hungary and Austria Historical data could be trackedout in several archives and was transferred into digital format (database entries ofownership data, digitizing of cadastre data) while the current data sets were alreadydigitally available Data acquisition represented the most time-consuming phase ofthe project For this reason, a well-planned database for fast and semiautomated dataentry was an important precondition

20.4.2 C ONCEPTUAL D ATABASE D ESIGN

The project database consists of the main entities “ownership,” “parcel,” and “landbetween ownership and parcel in Austria and Hungary is a one-to-many relationship:use,” defined as a one-to-many relationship (Figure 20.4) Basically, the relation

one owner holds one or more parcels The particular regulation in the Austriancadastre and land register though knows so-called “storage numbers,” which areidentification numbers for one or more owners These owners can be married people,brothers and sisters, or agrarian communities Therefore, the exact relationshipshould be described as a many-to-many relationship Since this fact makes modelingmuch more complicated, the project consortium agreed on the simplified version of

a one-to-many relation: one ownership holds one or more parcels This system wasvalid for the Hungarian system until about 1950 as well After the assumption ofthe Communistic system the conditions turned to a “modified” one-to-many rela-tionship: one collective held a lot of parcels Recently the registration system inHungary was again changed into a “real” one-to-many relationship: one owner (notownership) holds one or more parcels Since for one parcel often exists more thanone land use type (e.g., building and adjacent garden on one parcel), also the relationbetween land use and parcel was organized as a one-to-many relationship Furthertables act as look-up tables in the database Figure 20.4 shows the entity relationshipmodel

For the conceptual database design these considerations were important basicissues The most challenging part of the project was the harmonization of twonational land administration systems and the modeling of the “cadastral and landregister reality.”

Detailed work had been invested in the conceptual design of the relationaldatabase, especially for the data acquisition The database had to guarantee an easy,fast, and semiautomated entry of thousands of records Moreover, redundancies andincorrect entries should be avoided The database had to deal with historical as well

as with recent data by simultaneously compensating different format, language, andmeasuring unit systems of diverse systems (recent metric system and historic

“Klafter (Öl)”system).

A relational database organizes data in tables Each table is identified by a uniquetable name and is organized in rows and columns Since data are often stored inseveral tables, those tables can be joined or referenced to each other by commoncolumns (relational fields) These columns often contain identification numbers thatact at the same time as primary key [8] The primary key here is a combination ofparcel-ID (including sub-ID resulting from parcels that had been split up previously)and the identification number of each cadastral district, which represents a uniqueidentifier for each test area The composition of two or more columns guaranteesthe uniqueness of the primary key Based on this key, the records are joined to thespatial objects, in particular, parcels in the GIS

FIGURE 20.4 Entity relationship model of project database.

The spatial data for the project were derived from historical cadastral maps bydigitizing After the preprocessing steps of scanning and geo-referencing the ana-logue maps, each year of interest was digitized, including updates and geometrychanges of parcels Digitizing basis was dependent on the availability of cadastralcommonly used desktop GIS Current cadastre maps are already digitally available.

A new approach in storing geographic data related to nonspatial data is thegeodatabase model from ESRI® (Environmental Systems Research Institute, Inc.,Redlands, CA) Especially for purposes like the present project, it would be feasiblebecause it facilitates digitizing and editing by implementing topology rules, and it

is able to validate features by previously defined integrity rules The geodatabasemodel supports an object-oriented vector-data model where entities are represented

as objects An object is a collection of data elements and operations that togetherare considered as a single entity, which can adopt properties, behavior, and relation-ships [9] It is a data model where geo-data (and nonspatial data) from several datasources can be unified into one general system Editing rules, especially splitting ormerging policies, are interesting functionalities for the purposes of digitizing parcels.Due to the limited project time, the advantages of this editing and data storageapproach could not be tested in its entirety, but it can be suggested as a convenientmethod for further research in the field of cadastral investigations

20.4.4 M ODELING THE C ADASTRAL R EALITY

In general terms, a model is a representation of reality In this project it was intended

to represent and imitate the reality of land registration in a simplified model in order

to reproduce a status quo of property and land for a requested analyses period.Combined with the conceptual database design, the most challenging part of thedata acquisition phase was therefore the modeling of the real-world land registrytransactions into simplified functions These “functions” represent the process ofregistration regarding changes of ownership, land use, as well as parcel size Thefirst registration of a parcel or owner, the reunion or cancellation of a parcel, theadjustment of area, and many more are examples of such “functions.”

By anatomizing each course of registration into its basic elements, the modeling

of the real world was simplified, and basic modules were established The nation of various atomic components resulted in realistic functions, which were part

combi-of the database model Together with the date combi-of change, the feasible function foreach action in the land register was stored as an entry in a special table of thedatabase If more steps (respectively, modules) for one “land registration” weremaps (see Figure 20.3) Cadastral data was generated in both countries with a

necessary, more entries were stored, whereupon the exact time flow had to beconsidered to maintain the referential integrity rules of the database.

Figure 20.5 shows the scheme of modular assembly for the collectivization process

in Hungary; in this example the action consists of four basic modules (four steps).The function “collectivization of parcels,” for example, which was applied forthe Hungarian test area, consists of four basic modules: “cancellation of old (previ-ous) parcel,” “cancellation of land register ID,” “new entry of new land register ID,”and “new entry of new parcel” (in that order) The modules, according to thisexample, exactly represent the procedure of the real-world land registration Eachfunction is predefined by an identification number (type of change) and furtherparameters

20.4.5 P ARCEL -R ELATED AND O WNERSHIP C HANGES

The starting point for modeling the “cadastre reality” was the knowledge of basicchanges and processes that can occur on land use, parcel geometry, and ownership

FIGURE 20.5 Scheme of modular assembly for the collectivization process in Hungary.

Step 3: Entry of new

ID of land register Step 4: Entry of new parcel ID

The following list of possible changes regarding cadastre and land register wererelevant for the project areas:

• Change of parcel geometry (e.g., merging or splitting of parcels)

• Land consolidation (basically similar to splitting or merging parcels)

• Collectivization in Hungary (comparable to merging parcels)

• Privatization process in Hungary (comparable to division of parcels)

• Adjustment of parcel boundaries based on recent surveying

The schematic presentation of possible geometry changes is shown in Figure 20.6

An example of a typical land use change is the transformation from arable land

to residential area Ownership changes can be distinguished between the change ofproperty rights due to assignment, inheritance, purchase, expropriation, etc and thechanging number of owners

Finally, table joins in the GIS organize the linkage between the nonspatial datafrom the database tables and the spatial objects (parcels) These joins formed thebasis for analyses, queries, and visualization in the GIS

Land consolidation

Adjustment of parcel boundaries

according to the database model (Figure 20.4), represent the status quo of land use

geometry The last two issues were mainly analyzed using a GIS, since in this casethe geometric basis was significant For ownership changes it was sufficient toexecute queries directly within the database The visualization of the results wasperformed within the GIS The following examples give a brief overview on theresults of the investigation.

20.5.1 L AND -U SE C HANGES

Land-use change means a transition from one type of land use to another for variousreasons Within one hundred years, half of the area (55% of the analyzed parcels)

remained unchanged in the Austrian test site Girm while a considerable percentage

of area turned into forests but also into vineyards (Figure 20.7)

The conversion of agricultural land into forested land is a phenomenon that can

be observed very often in rural and border regions in Austria: an area of lowproductive conditions is given up or afforested, and a favored area with goodconditions gets intensified or is released for the increasing residential area [10]

Land-use changes in the Hungarian test area Harka can be mainly traced back

to collectivization processes between 1960 and 1990, where the agricultural landwas cropped according to centralized plans The parcel structure changed from smallfields into large agricultural production areas In the phase of restitution andreprivatization from 1995 up to now, the parcel size decreased again, and related tothat, the variation in land use types increased

20.5.2 C HANGES IN P ARCEL G EOMETRY

Based on extensive land consolidations, the geometry of the larger part of theagricultural land in the Austrian test area was reshaped between 1981 and 1986

As Figure 20.8 shows, land consolidation was necessary because of the narrow,long fields in the center of the village making difficult an efficient cultivation ofland In the same period the large-scale landed property of a big land owner in thenorth of the territory was portioned and bequeathed Apart from those main encroach-ments, only marginal parcel changes were registered within the one hundred years

of the research period

FIGURE 20.7 Land use changes in the Austrian test site; most of the affected arable land

changed into forested land (19%).

(Figure 20.8)

Even more affected by geometrical changes was the Hungarian test area, wherethe collectivization process from 1960 to 1990 caused reorganizations of nearly allanalyzed parcels and again from 1995 to 2000, where the whole area was “shuffled”due to reprivatization of the previous collective land.

20.5.3 O WNERSHIP C HANGES

Chart 20.1 and Chart 20.2 illustrate the changes of ownership in the tested ipalities of Austria and Hungary summarized and visualized as five-year steps

munic-In the case of Girm there is one characteristic low point (1945) in the chart

reflecting the years of World War II where the real estate market was nearly brokendown The number of ownership changes is increasing from then to a peak in theyears around 1970, a reason for the high percentage could not be statistically fixedthough it is supposed that the figures can be traced back to the general boomingtendency in this period A high number of property changes are noticeable also forthe years around 1985 where the related effects of land consolidation (exchange ofparcels, replacement, etc.) were documented in land registration (Chart 20.1)

The municipality of Harka reflects perfectly the political conditions in Hungary

(Chart 20.2) There are just a few ownership changes until 1940, also with a low point

in 1945 The highest number of property transfers can be noticed in the years around

1960, when the collectivization was carried out and parcels of private ownership

became one collective agricultural land, the “Nyugatmajori Állami Gazdaság”

FIGURE 20.8 Parcel changes in the Austrian test site between 1981 and 1986: only the gray

colored parcels remained untouched (right).

(“Nyugat major state collective”) Between 1960 and 1991 nearly no ownershipchanges are registered; there seems to be a hold up in the real estate market due tothe political conditions From 1995 to 2000 the beginning of the reprivatizationprocess and the restitutions of the former collectivized land are illustrated by con-stantly increasing figures.

20.6 PROBLEMS ENCOUNTERED

Originally the idea of the present project was to conduct analyses back to the middle

of the nineteenth century, where land register was founded in the K&K Monarchy

and the earliest cadastral maps were established Unfortunately several “missinglinks” like lost historical cadastral maps or land registration documents due tochanging political conditions and social turbulence made it impossible to follow thered thread

CHART 20.1 Ownership changes in the Austrian test area (in percentage).

CHART 20.2 Ownership changes in the Hungarian test area (in absolute figures).

Changes in the numbering system of parcels without registering the old onesand changing or lost ownership identification numbers in Hungarian land adminis-tration systems represented further obstacles in reproducing historical facts Theconsequence was to restrict the investigation period to the last century.

20.7 OUTLOOK

Based on the results of this study, possible future scenarios of land use, agriculture,rural development and property rights in the border region can be estimated consid-ering Hungarian’s access to the European Union in May 2004

With the termination of the Structure Funds programming period of the EuropeanUnion at the end of 2006, the “objective 1” funding stops for the Austrian borderregion, and instead Hungary will benefit from this funding Effects on land use (e.g.,decrease of arable land due to abandonment, afforestation, or transformation intobuilding land) and distribution of property (increase of dynamism of tenure) can beexpected In Hungary, based on the increased traffic of tenure and the reprivatizationprocess the average size of arable land per private or cooperative farm is presumed

to decrease After the process of compensation and restitution, since 1992 tural cooperative societies of a new type have been founded (corporations, limitedcompanies) [11] These cooperative farms have economic and production advantagescompared to the small-scale agriculture in the Austrian border area This could lead

agricul-to an intensification of cultivation, since soil conditions in West Hungary are optimal.The land market of Hungary has not yet been fully liberalized with the accession

to the European Union, but nevertheless the trans-boundary traffic of land in theHungarian border area to Austria is very high Farms run by Austrians on Hungarianterritory are estimated around 500 and 700 (± 20%) for the year 1999 [11] In the

meantime the number is supposed to have increased

With the constantly increasing road traffic in middle Europe, the extension andcompletion of the transnational road networks can be expected This assumptionwould definitely evoke direct and indirect consequences and changes on the borderregion of Austria/Hungary, combined with impacts on land use, ownership structure,and traffic of real estate

in terms of land cover (e.g derived from classification of satellite images), due todifferent definition for “land use” in land administration