Koncepcja badania pokrycia terenu w Polsce i określenie prognozy jego zmian

Barbara Prus


Land cover carries the broadest range of information on space, its structure, as well as its development and investment status. It is defined as a geo-bio-physical cover seen from the ground or via remote registration. This notion is closely related to the definition of land use, which expresses the function or the purpose of the use of space from a socioeconomic point of view, therefore, land use follows from the juxtaposition of land cover with its utilization. Forms of land use affect the image of land cover, which is the result of continued environmental and anthropogenic changes. At the same time, land cover affects many elements of the natural environment, such as the local climate, the formation of physiographic conditions, and visible forms of the natural and cultural landscape. This is the reason why recording land cover and monitoring the changes within it is so important. Land cover can be analysed at various levels of detail. The aggregated format, adopting certain simplifications and generalizations, distinguishes five classes of land cover: anthropogenic areas, arable land, woodland and semi-natural ecosystems, wetlands, and water bodies. The natural cover of the area comprises vegetation, including woodland, grasslands, arable land, as well as open areas (sands and solid rock) and water bodies. Land cover and land use in the local and global aspect, as well as the study of the way they are changing, are of interest to specialists in various fields. Since the last decade of the twentieth century, the CORINE Land Cover program (Coordination of Information on the Environment), supervised by the European Environment Agency, has been collecting information on the condition of the natural environment and its resources, including land cover changes in Europe, based on aerial photographs taken by the Landsat satellite. Funded from the PHARE fund, the CORINE Land Cover (CLC) program was extended to include the countries of Central and Eastern Europe. In 1993, Poland joined the program alongside the Czech Republic, Slovakia, Hungary, Romania and Bulgaria. The changes in land cover turned out to be highly significant, and therefore it was decided, in the year 2000, that the data should be updated in six-year intervals. Thus, the CLC for Poland encompasses the following editions: 1990, 2000, 2006, and 2012. The geometric accuracy of satellite images makes it possible to record minimum sections with an area of 25 ha, and in the case of their changes, of 5 ha. Ground cover classes distinguished in the CLC program are organized according to three levels of detail. The first level includes five main types of land cover forms, the second level distinguishes 15 forms, and the third, 44 forms. In Poland, 31 or 32 land cover forms can be distinguished at level three, depending on the analysed year. The area of Poland comprises about 312 thousand km2, and its terrain forms follow a layout of latitudinal stripes. The lowlands constitute 91.3% of the country’s area; the highlands make up 5.6%, and the mountains, 3.3%, of which 0.2% are high mountains. The area of Poland covers 344 mesoregions that differ in terms of physiographic conditions. The new division of Poland into physico-geographical regions is the result of the work by 26 experts from 14 universities and scientific institutions. It is a modification of the division proposed by J. Kondracki and A. Richling in 1994. The hierarchical division of regions includes megaregions, provinces, sub-provinces, macroregions, and mesoregions. The mesoregions, distinguished within the space, are geometrically homogeneous areas, characterized by morphological and morphogenetic similarity of formations, recurrence of particular forms of the terrain, and specific land cover. The area structure of the mesoregions is quite diverse. The minimum area is just over 6 km2, and the maximum area amounts to approximately 3778 km2. The purpose of this work was to develop a concept for studying land cover in time and space, and to determine the trajectory of land cover changes in Poland, as subdivided into mesoregions. Additionally, based on the observation of changes in land cover in the years 1990–2012, two versions of the forecast of changes for the years 2012–2024 and 2012–2034 were developed. Qualitative and quantitative analyses were carried out for the period 1990–2012, which was dictated by the availability of uniform CLC data from four measurement series: 1990, 2000, 2006, and 2012. Five basic forms of land cover in Poland were indicated. Using spatial queries, statistical summaries and analyses were performed, which made it possible to describe land cover changes in three time intervals: 1990–2000, 2000–2006, 2006–2012, and to develop an assessment of these changes in reference to the economic and political situation. The analysis of the largest transformations of land cover forms (in terms of area size) was carried out, thus determining trajectories of the changes. A detailed description of mesoregions was made in terms of land cover as well as according to area structure. 8 types of mesoregions were determined, which are homogeneous in terms of land cover forms, based on the taxonomic method, the agglomeration method, and Ward’s combinatorial clustering method. The highest number of mesoregions (105) represents a high level of urban sprawl (low-density urban development), as well as an average level of arable land and forests. Having said that, the percentage share of the latter in the total area is relatively small (7%). In terms of surface share, two types of mesoregions dominate, having a high share of arable land and, at the same time, a low share of forests. These types comprise 75 mesoregions, which corresponds to 35.4% of the total area. The work details the changes in land cover in three research periods, and analyses area structure of changes over the 22-year period. The changes taking place were juxtaposed against the distinguished types of mesoregions that were homogeneous in terms of land cover for the third level of detail according to CLC. The prediction of changes was based on the method of extrapolating the mean empirical increment. The mean errors for each forecast were determined ex ante. Based on the Markov model, probability matrices of transition for individual land coverage classes, at CLC level of detail 3, were calculated and summarized. Based on the results of the forecast, the changes in land cover were juxtaposed against selected forms of nature conservation (namely, those having the most restrictive legal protection), indicating the locations of potential spatial conflicts. The analyses that we have carried out demonstrated that arable land dominates among the land cover forms in Poland, as well as forests and semi-natural ecosystems, occupying a total of about 93% of the total area. On the level of detail 3, land cover changes were recorded on the area of 7481.5 km2, which is 2.4% of the total area of the country. The largest number and the largest surface area of changes were recorded in the last of the analysed periods, that is 2006–2012. The changes were noted in the area of 3110 km2, which corresponds to 41.6% of the total changes throughout the research period, and indicates the growing dynamics of those changes. Transformations within the category of forests and natural ecosystems accounted for around 60% of all changes. In this category, the largest number of directions of land cover changes has been recorded as well. Also arable land was subject to significant changes (about 30% of all changes observed). The size of anthropogenic areas, forest areas as well as semi-natural ecosystems was dynamically growing, while the area of arable land was decreasing at a rapid pace. Trajectories of land cover changes, including sequences of transformations in subsequent points in time, indicate irreversible transformations towards anthropogenic directions. The persistence of land cover in mesoregions, calculated for three research intervals, indicates that the greatest fluctuations appeared in the first research period, that is, in 1990–2000. The largest land cover changes, reaching 21.6% of the surface, occurred in south-western and western Poland (within the area of three mesoregions). In the first research period, after excluding the outliers from the analysis, the percentage of mesoregions was dominated by those with a percentage share of changes not exceeding the value of 7.7%. The second period (2000–2006) presented a lower level of changes in individual mesoregions, not exceeding 5%. The third period of the analysis covered changes of up to 11.7%, especially in western Poland. The changes that took place covered increasingly larger areas in total, but the size of individual changes decreased. In the first period, large changescovered only a small number of mesoregions, while in the last examined interval it was changes of a lesser intensity that covered a larger number of mesoregions. The results we obtained allow us to conclude that the largest number of mesoregions were subject to transformations within the range of 0–3.5% (269 mesoregions), covering as much as 81.8% of the area in total. The highest stability of land cover during the 22-year research period was found in mesoregions located in the southern and central part of Poland, as well as in Podlasie and Roztocze regions. The first version of the forecast developed for mesoregions (2012–2024) indicates an increasing probability of changes (the forecasted maximum value of changes is 20.7% of the mesoregion’s area). In this version of the forecast, approximately 25.9% of the mesoregions are either not subjected to changes, or these changes do not exceed 0.3% of the area, with the mean error of forecasting calculated ex ante at the level of 0.04%. The second version of the forecast (2012–2034) envisages changes that cover a larger area than in the first version, however with smaller fluctuations – up to 17.9%, with a forecasting error of 0.05%. In this version of the forecast, only 9.6% of mesoregions (that is, 33 of those) are not subjected to changes in land cover. The probabilities of transitions between individual classes of land cover, determined and arranged in the form of a matrix, indicate the growing probability of changes in the anthropogenic direction. Subsequent reported changes result from the implemented motorway construction projects as well as from the principles of tree stand renewal program run by the “State Forests” National Forest Holding. The most stable land cover classes in Poland include water bodies and wetlands as well as selected categories from the group of anthropogenic areas. Juxtaposing the results of the forecasts with the locations of forms of nature conservation found in Poland reveals the possible spatial conflicts, especially in the case of protected landscape areas, both in the instance of the first and the second version of the forecast. The analyses that we conducted indicate the feasibility of adapting the physiographic division of Poland for the purpose of the research into changes in land cover, and for predicting changes in subsequent research periods. They can provide the basis for determining problem areas that should be monitored due to spatial conflicts.
Other language title versionsAn approach to studying land in Poland and forecasting its changes
Book typeMonograph
Author Barbara Prus (FoEEaLS / DoSPaLA)
Barbara Prus,,
- Department of Spatial Planning and Landscape Architecture
PublisherUniwersytet Rolniczy im. Hugona Kołłątaja w Krakowie, MNiSW [80]
Publishing place (Publisher address)Kraków
Issue year2019
Book series /Journal (in case of Journal special issue)Zeszyty Naukowe Uniwersytetu Rolniczego im. Hugona Kołłątaja w Krakowie. Rozprawy, ISSN 1899-3486, (0 pkt)
Publication size in sheets14.5
Internal identifierWIŚIG/2019/65
Languagepl polski
Koncepcja badania pokrycia terenu w Polsce i określenie prognozy jego zmian 51,61 MB
Koncepcja badania pokrycia terenu w Polsce i określenie prognozy jego zmian 464,31 KB
Score (nominal)80
Score sourcepublisherList
Citation count*
Share Share

Get link to the record

* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Are you sure?