1. Introduction
The evolution of geoinformation systems and technologies and the gradual digitization of previously non-public or otherwise difficult-to-reach maps [
1], together with the overall advancement in cultural landscape research, have yielded valuable information facilitating detailed monitoring and evaluation regarding historical landscape development. The latter two steps are among essential tools that allow us to solve topical problems of landscape ecology, mainly because through their informative value, they open new possibilities for historical ecology as well as the application of multiple current and potentially novel methodical approaches to enable relevant analyses [
2]. Historical land cover and land use, for example, are mostly reconstructed by means of historical topographic maps combined with aerial images [
3,
4,
5,
6]. Considering a long-term perspective, namely one of about 200 years, a combination of various maps with different legends, scales, and mapping modes facilitates presenting the results of land-use development within several basic categories [
7,
8]. Maps generated within shorter time horizons allow us to investigate the progress in functional land use by exploiting detailed categorization of individual areas [
9,
10]. Another approach to the historical development of land-use/land cover (LULC) used in the Central European concept of landscape ecology is based on the summary statistics for administrative units divided into eight basic landscape element categories, with the smallest mapping unit being the cadastral area; this technique, although suitable for the LULC changes and driving forces’ analysis of landscape macrostructures, cannot be used for detailed analysis at the local level [
11,
12] if appliedwithout relevant aerial images.
Landscape structures are usually mapped together with the advancement of transport networks [
8], a major aspect of analyzing long-term land-use changes [
13]. This factor is examined in view of its impact on urban sprawl [
14,
15,
16], the distribution of fresh industrial and retail activities [
17,
18,
19], and local leisure and tourism potentials [
20,
21]. In applicable evaluations, development is classified as a key technological force behind landscape transformations [
22,
23,
24]. More broadly, transportation and land use are part of a retroactive feedback system [
25], meaning that the localization of suitable transport infrastructure and good transport accessibility define the potential of localities to sustain regional development on one hand, and the nthropogenic transformation of original landscape components into elements of the corresponding socio-economic structure on the other [
26,
27,
28,
29].
Generally, transport corridors and urban sprawl embody the most significant elements that contribute to landscape fragmentation, reducing the relevant qualitative indicator values [
30,
31]. The expansion and structuring of an urban area interact with other components of the transportation system, including, but not limited to, public transport accessibility [
32,
33,
34] and the behavior of individual transit participants in choosing the means of transport [
35,
36,
37] or negotiating the safety risks [
38,
39,
40]. The health risks and quality of life reductions associated with human exposure to significant adverse transport-generated effects have been broadly discussed in the literature and at policy-oriented forums [
41,
42,
43,
44,
45,
46]. In this context, for instance, noise and air pollution manifest themselves especially in cities, where the population densities and pollution rates are high; paradoxically, such negative environmental conditions may most seriously affect population groups that actively seek and practice sustainable transportation [
47,
48].
Increasing the modal share between walking, cycling, and public transport services embodies the central objective of modern transport policy, as adopted by many countries and cities, aiming to lessen the volumes of individual road transportation. Encouraging these modes of transfer closely relates to the trend towards concentrating urban areas into structurally compact cities [
49,
50].
The criteria defining acceptable walking distance by commuters are crucial for accessibility studies. Zielstra and Hochmair [
51] assumed the walking distances to urban bus and major rail stations to be 400 and 800 m, respectively, the authors point out that the existence of pedestrian-only segments of a network may increase the service area of the stations. Yang et al. [
52] set up the threshold values of access to urban rail stations in Bejing to 472 m as the optimum and 862 m as the maximum acceptable walking distances. The walking acceptability rate will double in pedestrian-friendly, attractive environments [
53], including parks, natural leisure sectors, and traffic-calmed residential streets. Geographical information systems (GIS) are widely used by researchers to analyze the accessibility, with the two basic options being linear buffering and network analysis [
54,
55].
This paper aims to identify the impact of major types of transport infrastructure on the urban structure in the historical context and to prove the connection between changes in the functional use and structure of an urbanized area on the one hand and its transportation development on the other, in the context of post-socialist countries of Central Europe. The basic hypotheses and assumptions include:
Constructing and developing transport (road, railroad) infrastructure may emerge as the driving forces behind the expansion of all functional types of built-up areas.
The intensity of landtake by transport infrastructure changed significantly in the post-socialist period as a result of general social changes.
The growth (or “sprawl”) of built-up areas having a residential or service character intensifies road transport volumes.
The accessibility of passenger railroad transport may generate increased demand for housing within a walking distance from railroad stations.
2. Materials and Methods
Landscape is a dynamic system that is constantly changing over time. The present landscape structure, consisting of land use and land cover, is the result of gradual variation in the original natural landscape caused by humans [
56]. The landscape structure takes into account biotic and abiotic processes and phenomena in the landscape and expresses the spatial arrangement of landscape properties. The landscape structure elements were classified by utilizing the methodology of landscape elements mappable in Slovakia as proposed by Petrovič, Bugár and Hreško [
57], which comprises six basic groups/categories: trees and shrubs; grass and herbaceous vegetation; agricultural crops; exposed bedrock and subsoil; water bodies and wetlands; residential and built-up areas. Our study employs the third hierarchy level, where 64 of the landscape elements specified in
Table A1 (the
Appendix A) are defined. The categories most important in this respect are:
Individual housing (code 611)—permanent residential built-up areas made of individual houses with private gardens or private housing estates
Collective housing (code 612)—various types of block of flats and apartment and multi-story housing objects. A detailed description with relevant typology is available from [
58].
Industrial and technical building (code 651)—areas of industrial activities such as manufacturing, logistics, and public services (e.g., water supply), including associated lands and infrastructures.
All transport-related objects and areas—road and freeway (code 661), parking (code 662), railroad track (code 663), railroad station (code 664), and airport (code 665). When generalized into traditional means of transportation, road transport is represented by codes 661 and 662, while the corresponding railroad transport codes are 663 and 664.
The landscape structure in the targeted region was examined and evaluated in five periods, represented by the years 1949, 1970, 1993, 2003, and 2018. Software QGIS 2.16.2 to 3.4 was used for spatial data acquisition, processing, and cartographic interpretation. The manual digitization (vectorization) of the maps was carried out at the scale of 1:2000 where possible, starting from the most recent period and proceeding further into the past. Following sources were used:
Black/white aerial images, 1949 (Institute of Military Topography, Banská Bystrica);
Black/white aerial images, 1970 (Institute of Military Topography, Banská Bystrica);
Basic topographic maps 1:10,000, 1993 (Geodetic and Cartographic Institute Bratislava);
Color orthophotomaps from the year 2003 (Eurosense + Geodis Brno);
Color orthophotomaps from the year 2016 (website mapy.cz).
The landscape structure elements identified in 2016 were verified via field survey and updated to match the conditions of 2018; simultaneously, we clarified or eliminated the discrepancies that had occurred in the course of the mapping and photographed the region.
Differences between the individual periods were determined in the paired years 1949/1970, 1970/1993, 1993/2003, 2003/2018, and 1949/2018. Using appropriate papers by Feranec et al. [
59,
60], Kopecká [
61], and Cebecauerová [
62], 14 types of changes were established, guided by processes that indicate the formation or demise of certain landscape element.
Within
Table 1, the types of changes most important for the transportation impact analysis are:
Urbanization, showing the expansion of residential areas and their subcomponents, such as housing units or sports and leisure grounds, in the city and the country. The relevant events and procedures take place at the expense of agricultural, forest, and marshlands, and they significantly affect the various natural or artificial water systems;
Industrialization and technologization, meaning a growth in the transportation, industrial, and technical or technological infrastructures occurring at the detriment of agricultural, forest, marsh, and water-retaining lands.
One of the partial processes forming industrialization and technologization is the landtake by transport infrastructure. This process represents the change of a landscape element into one from the category of transport-related objects and areas (from 661 to 666) between two mapping years, with the deduction of the original transportation areas that changed into other landscape elements. The derived indicator—the intensity of landtake by transport infrastructure—is defined as the ratio of landtake by transport infrastructure and the number of years in the respective period, so the unevenly long periods between the individual mapping years will be transformed into comparable numbers. Partial processes, such as the landtake by roads and freeways (change to landscape element code 661), can be also identified if necessary.
The sources that yielded the details on land use and landscape transformation further allowed us to outline the main transport networks. It was true, especially for the old topographic maps, since the aerial images cannot show the exact categorization of the identified elements to expressways, highways, or first-, second-, and, third-class roads, where first-class roads are intended mainly for long-distance and international transport, second-class road’s major use is for traffic between regions, and third-class roads are used to interconnect municipalities or to connect them to other roads. The basic transport network infrastructure was then completed with the road traffic intensity data collected from the nationwide road traffic census held in the Slovak Republic—formerly a part of Czechoslovakia—from 1949 to 1950, every five years after 1980. The census covered all expressways, highways, first- and second-class roads, and selected portions of third-class roads. Currently, following the census of 2015, a novel methodology has been applied [
63], including partial changes of interpretation, by which, for example, a tractor unit towing one or more semitrailers now counts as a single vehicle. The traffic intensity is presented in the form of the traffic-engineering characteristic called Annual Average Daily Traffic (AADT) [
64], where the quantity is expressed as the number of vehicles per 24 hours on a given section of the road. The characteristic represents the average value for all days of the year, comprising also weekends or less dense traffic periods (such as wintertime). The individual survey years, as related to those of the landscape structure mapping cycles, are indicated in
Table 2 below. If the future traffic intensities are to be estimated, the most common approach is to use growth factors. These factors are officially published by the authorities [
65] and regularly updated after each national traffic census.
For the purposes of the research, we cannot assume the total traffic volume within the study area, as the extent/length of surveyed roads differed between the years. Each of the censuses, however, comprised the first- and second-class roads. Their paths inside the study area varied through time due to modernization-induced rerouting and, most recently, the construction of a part of the D1 freeway (completed in 2015). However, the primary traffic flow directions were preserved.
The accessibility rates were computed via least-cost modeling with the use of the r.cost tool [
71]. Least-cost modeling is based upon a GIS raster called a cost-surface. The values within a cost-surface are employed to represent the per unit distance cost associated with traversing different parts of a landscape [
72]. The cost surface was the whole road network at the spatial resolution of 5 m. Each cell in the input cost-surface map will contain the category value five, which represents the cost of traversing that cell. The output map layer represents the accumulated traversal cost, calculated as the product of the cost and the distance (in our case, Chebyshev distance [
73]) traversed. As a result, an insight into the development of the networks related to the accessibility distances was gained. Subsequently, we compared these distances with the development of three selected landscape elements (individual housing (611), collective housing (612), and industrial and technical buildings (651)), emphasizing changes in distances up to 500 m, the limit widely considered the best, and from 501 to 1000 m, the range assumed to be the acceptable walking distance. Regarding the more remote zones, the effect of the station(s) may rest predominantly in the necessity or possibility of using other means of transport (including, for example, municipal transport or bikes).
For each time period, analyses in ArcGIS 10.6 were performed to investigate the relationship between each polygon’s accessibility to the railroad station. While around six to eight thousand landscape element polygons entered the analyses in the oldest periods, more than 14 thousand did so in the most recent one. Using the Zonal Statistics Tool, the statistical parameters for all types of landscape elements were obtained. Only those landscape elements that have the highest explanatory value for the relationship between the transport accessibility and the development of the urban fabric are presented in the results.
3. Study Area
The study area is located in the Martin county, within the Žilina administrative region, which occupies a substantial part of northern Slovakia (
Figure 1). The area itself comprises two major self-governing municipal units, namely, the city of Martin and the town of Vrútky. Between 1949 and 1990, the settlements formed a single city together. Currently, despite being independent of each other, Martin and Vrútky maintain close social and economic bonds within an urban agglomeration. Their total area amounts to 8604 ha (Martin: 6770 ha; Vrútky 1834 ha), and the population equaled 61,357 persons as of 31 December 2018 (Martin: 53,768; Vrútky: 7589).
The study area is a part of the Turčianská Kotlina Basin, whose western border is formed by the Malá Fatra mountain range, which constitutes a certain transportation barrier (
Figure 2). Two rivers, the Váh and the Turiec flow through the basin and both were previously regulated. The weather is characterized by frequent atmospheric inversions leading to, among other issues, air pollution.
The area was colonized in the Eneolithic period. During the second half of the 19th century, it grew into a hub of Slovak national politics and culture, thanks to the introduction of and connection to a railroad track (see below). Consequently, the population increased, and the two municipalities expanded rapidly. A further period of growth caused by industrial development followed the end of WWII.
3.1. Transportation Networks
The Y-shaped backbone transportation network within the area pursues orographically preconditioned lines that are simultaneously also defined by the Trans-European Transport (TEN-T) core network. The Váh valley conducts, in the west-to-east direction, the northern branch of the multimodal Rhine–Danube Corridor (the road and railroad components being the D1 freeway and the track Žilina–Košice–EU to Ukraine border, respectively). Concerning D1, construction work has been completed only on an isolated section between Dubná Skala and Turany (commissioned in 2015), including the access route forming the initial sector of the R3 expressway.
While the importance of the first-class road I/18, or more concretely, the portion that passes through Vrútky, decreased after the opening of the D1 section routed in parallel, the road transportation in Martin still relies on the individual branches of I/65. As the backbone network remains incomplete, most first-class roads in the Martin subregion are in an unsatisfactory condition due to being heavily overladen, carrying the burden of not only local but also transit traffic. Moreover, there are locations where the entire load has been routed via third-class or urban roads.
The planning and construction of railroads too were adjusted to respect the landscape patterns. In this context, 1871 and 1872 embodied the threshold years, marking the commissioning of the Žilina–Poprad line, a major subsection of the Košice–Bohumín railroad [
74]. The track ran along the Váh valley to be joined—from the south, through the Turčianská Kotlina Basin—by the Hungarian northern railroad proceeding from Zvolen [
75]. The lines met at the Vrútky station; and the route originating in Žilina to pass Vrútky and run towards Košice and Ukraine is significant even presently, having considerable importance for passenger transport and potential for long-distance freight hauling. The area comprises several sidings that interconnect the railroad network and various industrial plants or facilities, and it is served by the Priekopa stop situated between the Vrútky and Martin stations.
The air transport relies on the domestic, leisure aviation-oriented civil airport at Martin–Tomčany.
3.2. Landscape Structure
Although the entire study area of 8604 ha has experienced major landscape structure changes (see
Table 3) since approximately 1950, the seven decades have been characterized by a predominance of landscape elements classified as trees and shrubs. Over time, the category has come to dominate, albeit by a relatively narrow margin, and it currently occupies more than half of the area. The second-largest share has been assigned to agricultural crops whose areal extent gradually shrank. The area of grass and herbaceous vegetation exceeded residential and built-up land; however, after 1993, the share has dropped markedly down to one-third of its original extent. Residential and built-up land, expectably, has expanded progressively to more than three times above the former proportion values. The two remaining classes have been represented only marginally, with none of them taking over 1% of the area.
The landscape structure maps for the individual periods are presented in
Appendix B,
Figure A1. More details about the landscape structure development within the study area during the post-socialist decades can be found in research by Nozdrovická et al. [
76].
5. Conclusions
The paper was designed to identify and characterize the interaction of landscape structure changes and transportation systems that occurred or were adopted in the city and town of Martin and Vrútky, respectively, after 1949. The two urban units, located close together, are administratively independent of each other but share infrastructures and maintain strong social and economic bonds. Previously, the settlements were joined to form the Martin–Vrútky conurbation, and, later, Vrútky became a city district of Martin. The area is generally diversified, offering a wide variety of natural elements and landscape uses. To a great extent, the landscape structure evolution reflects significant historical events and driving forces that exerted an impact on landscape utilization in Slovakia following the mid-20th century. The political system installed in 1948 markedly influenced the development of the agricultural land in that the state had become the exclusive landowner, with private land ownership abolished [
84]; analogically, ideology-induced transformations widely affected also the urbanization and industrialization of Slovakia. The results of the research confirmed transportation and the relevant infrastructure as a major factor of importance for long-term landscape changes in the monitored zones.
The highest traffic intensity rates determined through the series of censuses denote a gradual traffic increase; interestingly, the indicators grew ten- to twentyfold throughout the overall period, especially on the backbone routes substituting for the presently incomplete European multimodal corridor. However, the actual traffic intensity growth cannot be directly related to the expansion of the residential and utility built-up land within the area. To facilitate such a purpose, it would be necessary to know the details of both the traffic flow as regards its internal, external, and transit components and the variation in the behavior of road users. The data are obtainable exclusively via travel behavior surveys [
85], direction and target questionnaires, and exact traffic models [
86]. These specialized surveys and instruments, unlike common nationwide censuses to determine traffic intensities, are not conducted or applied regularly, being financially challenging and technically demanding if employed over large areas. Consequently, appropriate information on the past periods is unavailable as well.
By evaluating the accessibility of the railroad stations, we exposed the formerly pivotal role of railroad transportation in localizing individual housing and large industrial premises into the city and town centers; this scenario characterizes the years 1949 and (partially) 1970. The median availability of the railroad stations for collective housing increased significantly in 1970, with the growth completed in 1993 and remaining stable since then. The median availability of the railroad stations to the industrial and technical buildings rose markedly between 1949 and 1970; since the latter year, however, the newly built enterprises were harder to access from the railroad stations. In the more recent decades, the expanding motorization process, which formed and gained momentum in the 1960s, allowed residential, industrial, and utility grounds to spread farther from the train stations, as was also necessary due to the more densely built-up central sectors of the settlements.
The results of the relationships and links between transport accessibility, the development of transport infrastructure, and landscape elements can be important in urban spatial planning and strategic planning of city development.