Aesthetic value can be defined as the pleasure that people receive from scenic beauty provided by natural areas and landscapes . Evaluation of aesthetic value is interesting for public, political and scientific purposes . It is essential for people’s well-being, for their physical and psychological health . Identification of aesthetic valuable landscapes is helpful in defining areas to be placed under protection or for landscape planning[4, 5]. The visual beauty of a landscape is also the key factor affecting recreational potential of the landscape. The aesthetic value of a landscape is based on properties of the landform, structures, climate, vegetation and so on and the way they work together . It is what the landscape offers us but it is sensed and appreciated subjectively.
Studies on landscape aesthetics have been carried out since the 1960s . There are two ways to evaluate landscape beauty: subjective and objective approach . In the first case, the assessment is done by observer(s) through perceptual surveys, public questionnaires or interviews on personal preferences [8, 9, 10, 11, 12]. The evaluation depends on personal factors like age, gender or education . In the second case, the objective approach is based on expert analysis of the landscape. The landscape aesthetic value can be evaluated by measuring and scoring objective factors as diversity of landscape structures, vertical and horizontal articulation , contrast, colourfulness , or with use of landscape metrics . Standardized approaches for the assessment and monitoring of landscape aesthetics are still missing. According to Sevenant and Antrop , research on aesthetic landscape perception needs to focus on a range of landscape types, since the predictors of the aesthetic perception of the public differ in relation to the landscape types in which they are located. That is the reason why we decided to make study, in which we are evaluating all types of landscape occurring in Slovakia.
An interesting data source that reflects appreciation of landscape by people are the millions of geolocated photos uploaded to social media [15, 16, 17]. Platforms like Flickr or Google Panoramio collect geolocated photos of landscape and provide the metadata (exact location, time, photograph identification number) for the analyses. Such data have been used for example for evaluating of aesthetic value as cultural ecosystem service in Cornwall, UK , in Nebraska , or in coastal zones of Lithuania . Locations of photos in Flickr were used as proxy to estimate the visitation rates and travelers’ origins at recreational sites  or national parks . Metadata from geolocated photos have been used to identify and analyze the most popular visual attractions of European metropolises  or for extracting scenic routes in the state of California . Density of Panoramio photos was used as a proxy for evaluating the value and meaning of landscape types at European scale .
Landscape typization is one of the tasks resulting from the European Landscape Convention. National typologies [25, 26, 27] as well as European typologies [24, 28] are based on combination and synthesis of various layers of geographical and biophysical data . In Slovakia, the Landscape types, or Representative Landscape Ecosystems of Slovakia, were designed as homogeneous units considering landscape character, functions and current land-use . They were delineated as combination of 13 land cover units and 18 abiotic condition units grouped to lowland, basin and mountain landscape.
Knowledge and identification of landscape types is an unavoidable condition in the strategic planning process and effective protection of regionally unique landscapes . People’s landscape preferences and visions are important inputs for effective landscape planning [4, 32]. Linking the ecological measures with human aesthetic preferences is a promising approach for finding broader acceptance for ecological measures . In this study, we present a simple and fast way to evaluate the appreciation of landscape aesthetic value on national scale. Our aim is to use location data from social media photos as an indicator for evaluating appreciation of aesthetic value of landscape types in Slovakia. In particular we: (1) analyze the distribution of social media photos in Slovakia; (2) compare the density of social media photos in Slovak landscape types.
2 Data and Methods
The landscape types were delineated by overlaying of abiotic landscape structure (type of relief, quaternary deposits, climatic regions, and soil types) and land-cover map (CORINE Land Cover 2006 - CLC). This process yielded to a patched map of homogeneous areas, which were further interpreted, generalized, and regionalized to a final map of Slovak Landscape types . Overall 126 basic landscape types, which represent unique combinations of land cover in different abiotic conditions, were defined. For the analysis of appreciation of landscape aesthetic value we used final map of landscape types (Figure 1, Table 1) taken from Atlas of Representative Landscape Types of Slovakia .
From available social media, we chose Google Panoramio, which is directly focused to presentation of geolocated photos of landscape sites. Photo’s metadata (containing geographic coordinates, capture time and photographer identification number) were downloaded using the Panoramio REST API . Altogether 3 330 962 photos, taken in years 2005 - 2014 were analysed (Table 2).
Firstly we calculated the number of photos per square kilometre. To avoid the bias caused by high density of photos from the place where photographer lives, we counted only one photo per user per square kilometre. This also reduced the bias caused by photographing more photos from one place by one user. To prevent the boundary effect (lowered photo density along the boundary of the study area), we included the photos located within the analyzed distance behind the boundary (totally 10 092 photos). To analyze the appreciation of landscape aesthetic value for specific landscape types, we calculated the photos density (average number of photos per unique user per square kilometre) in each landscape type separately (Table 3). Then, we summarized the photo densities for each group of landscape types.
3.1 Density of Panoramio photos in Slovakia
Figure 2 presents the density of Panoramio photos from unique user per square km in Slovakia. The most photos are taken in urban landscapes, where people live and take photos more often. Therefore we interpret our results for urban and non-urban landscape separately.
Hotspots in urban areas are generally situated in city centres, especially close to tourist attractions like historical monuments, museums, parks, restaurants and city squares. The place with the highest photo density in urban areas is Old Town Bratislava (121 photos/km2) where the Bratislava Castle, St. Martin’s Cathedral, Michael’s Gate and riverbank promenade are most often photographed. The second highest density of photos (114 photos/km2) is found along the river bank in Šturovo where people enjoy the views on Esztergom Basilica. The third highest density of photos (104 photos/km2) is found in Starý Smokovec, a popular place for skiing and other winter activities. The fourth highest density of photos (94 photos/km2) is found in Banská Štiavnica, a town in central Slovakia with historical value and proclaimed to be a World Heritage Site by UNESCO. The fifth highest density of photos (84 photos/km2) is found in Bojnice, a small city with a famous spa, zoo and a romantic medieval castle.
Hotspots in non-urban areas are more scattered. Most of them are located near natural monuments, natural interests, castles, lakes, hills, mountains or place where people can have beautiful view on landscape. The density of photos is also influenced by the accessibility and distance from touristic paths. The highest density (142 photos/km2) of photos is found in Skalnaté pleso, a picturesque mountain lake close to favorite ski, tourist and health resort in High Tatras. The second highest density (139 photos/km2) of photos is found in Spiš Castle, the largest castle complex in Central Europe. The area was included in the UNESCO list of World Heritage. The third highest density (137 photos/km2) of photos is also found to be close to a castle – the Orava castle situated in northern Slovakia. The fourth highest density (111 photos/km2) of photos is found in hill Beskyd located in The Western Tatras on Slovak-Polish border. The fifth highest density (101 photos/km2) of photos is found in a camp near to a small village Červený Kláštor in Pieniny National Park. This camping area is famous for nature and rafting on the small river Dunajec. which makes a border with Poland.
3.2 Density of Panoramio photos in the Landscape Types of Slovakia
Most of the photos were taken in agricultural landscapes (81 646 photos), which cover the largest part of Slovakia (19 261 km2). Within the abiotic landscape types, the most photos were taken in core areas of uplands (61 180 photos) and core areas of highlands (61 351 photos). The basin landscape types were found to be most attractive (6.65 photos/km2), followed by mountain landscape (2.42 photos/km2) and lowland landscape (1.87 photos/km2).
Regarding the landcover typology, the highest density of photos was found in urban landscape types (Table 3) where people live and therefore take photos more often. Outside the urban landscapes less intensive type of landscapes were found to be visually more attractive. The lowest densities of photos were found in the agro-forest landscape with deciduous forest, deciduous forest landscape and scattered settlements landscape. More preferred landcover types are grassland landscape and forest landscapes (except the deciduous forest landscape type) and the most preferred is subalpine and alpine landscape. Subalpine and alpine landscapes can be found at Vysoké Tatry mts., Nízke Tatry mts. and in very small scale at Veľká Fatra mts.
With use of geolocated social media photos we identified places and landscape types where the aesthetic value of the landscape is most appreciated by people. Working from the premise that higher density of photos means higher landscape appreciation, alpine and subalpine landcover type is found to be the most valuable. These landscape types cover High and Low Tatra Mountains and partly Velká Fatra mts. The high density of photos could be related to scenic beauty of the mountainous landscapes as well as the easy accessibility for tourists Alpine and subalpine landscapes were not affected by industrialization and intensification of agriculture, that took place in Slovakia in second half of 20th century and the untouched landscape is still visually attractive to tourists/photographers. Rocks, wooded grasslands and shrubs are generally more photographed in comparison to the actual land cover .
Despite the negative relationship between anthropogenic elements and visual quality , urban landscape has very high number of uploaded photos. The main reason for the high density of photos in urban regions is the higher number of people living and taking photos in urban areas compared to the number of people living in other landscape types In addition, visual appearance is positively affected by long history, dramatic historical events, rich culture, which evidence we can see in many churches, castles, castle ruins, ancient city walls, statues or fountains. Manmade elements such as churches, traditional rural buildings, and castles are perceived as beautiful . Wood et al.  proves that the cultural attractions have higher visitation rates than natural ones. On the other hand arable land and urban features tend to be photographed less frequently in comparison to they proportion in a landscape .
Water elements, vegetation and traditional farm buildings can increase the appreciation of landscape . Our research shows that cultural heritage sites with long history (e.g. Orava and Spiš Castles, Červený Kláštor) are the most preferred by photographers. Sklenicka and Molnarova  did a study based on questionnaire survey where the highest preference among vegetation types was for managed coniferous and wild deciduous forest. According to our results (Table 2), deciduous forests are not so highly appreciated. It may be because the deciduous forest is located mostly on low mountains that are less attractive for tourism and with less number of viewpoints suitable for taking scenic photos. In Filova et al.  forest land was the least rated. Results of the same survey shows that arable land is particularly highly rated . In our study most photos were taken in agricultural landscape. Agricultural landscape covers the largest part of the Slovakia therefore total density of photos per square kilometre (number of photos divided by area of arable land) was low. The low appreciation of lowland landscape could be explained by intensive agricultural activities that makes this landscape unattractive.
Use of geolocated social media photos for analyzing landscape aesthetic values and its appreciation by people has some limitations. Results are strongly biased by the accessibility to scenic locations. Most photos are taken in the places where people live or go for vacations. Therefore, we cannot be conclusive about solely objective landscape aesthetic values (based on properties of the landform, land structures, climate, vegetation and manmade structures) without any bias from its accessibility to people. In other words some landscapes could be visually more attractive, but because of their inaccessibility this attractiveness is not sensed and appreciated by people. This is similar to the concept of ecosystem services - every ecosystem has its functions independent from human presence and when people receive their products, only then we can talk about ecosystem services Second limitation of our analysis is that the study is based on geolocated photos in social media, taken and uploaded by a small subset of people. This subset consists of technically skilled people with access to computers or smart phones that may not represent the entire population and their visual preferences could be different from the entire population. A common problem for all the studies that use geolocated photos is the distance of the photographed object from the place where the camera records the position coordinates. In our case this offsite effect occurs on the landscape type boundaries, where the area that belongs to one landscape type could be photographed from the place beyond the boundary of other landscape type. On one hand we can consider the off-site effect as a source of uncertainty. This uncertain area, is relative small in comparison to area of landscape types and therefore we think that the influence of off-site effect on our result is minimal. On the other hand referring to definition of landscape by European Landscape Convention as “area, as perceived by people”, we can consider the scenic view on the surrounding landscape as a quality of the landscape (or a place) where the viewpoint is.
Analysis of geolocated photos is a simple and fast way to evaluate the aesthetic value of the landscape and its appreciation by people. We found the most attractive hotspots to be Alpine mountains and around castles. Two most attractive hotspots in urban areas are located in the city centre close to river. Beside that we proved that less intensive landscape types are more aesthetically valuable and appreciated by people. One interesting exception is landscape types with deciduous forest, where we found low densities of the photos.
Our analyses could be used as arguments for protection of visually attractive places or landscapes. It can also help to identify places that may be objectively visually attractive, but are not visited because of low accessibility or some other constraints. We can also find the places, for example nature protection areas, that are visited by people even when entrance is not allowed (for example 48.75 % of photos from touristic paths in High Tatras are taken at the period of year when the access is restricted).
The contribution was prepared within the grant project of the Ministry of Education of the Slovak Republic and the Slovak Academy of Sciences No. 2/0171/16 “Changes in Slovak Landscape Driven by European Union Agricultural Policy”
Kumar, P., The economics of ecosystems and biodiversity: ecological and economic foundations. Earthscan, London, 2010 Google Scholar
Frank, S., Fürst, Ch., Koschke, L., Witt, A., Makeschin, F., Assessment of landscape aesthetics-Validation of a landscape metrics-based assessment by visual estimation of the scenic beauty. Ecol. Indic., 2013, 32, 222-231 CrossrefWeb of ScienceGoogle Scholar
Brook, I., Aesthetic appreciation of landscape. In: Howard, P., Thompson, I., Waterton, E. (Eds.), The Routledge Companion to Landscape Studies. Routledge, London, 2013, 108-118 Google Scholar
Filova, L., Vojar, J., Svobodova, K., Sklenicka, P., The effect of landscape elements on public visual preferences: ways to use knowledge in the context of landscape planning. J. Environ. Plan. Manage., 2015, 58, 2037-2055 CrossrefGoogle Scholar
Stefunkova, D., The possibilities of implementation of landscape visual quality evaluation to the landscape ecological planning. Ekol., 2000, 19, 199-206 Google Scholar
Purcell, A. T., Peron, E. M., Berton, R., Why do Preferences Differ Between Scene Types? Environ. Behav., 2001, 33, 93-106 Google Scholar
. Barčáková, I., Approaches to evaluation of the aesthetic (visual) quality of landscape. Geogr. Cas., 2001, 53, 343-356 Google Scholar
Baránková, Z., Dobrovodská, M., Štefunková, D., Babicová, D., Moyzeová, M., Petrovič, F., Participation of local people on identifying the landscape values and future development in historical agricultural landscapes. Ekol., 2011, 30, 216-228 CrossrefGoogle Scholar
Daniel, T. C., Muhar, A., Amberger, A., Aznar, O., Boyd, J. W., Chan, K. M. A., Costanza, R., Elmqvist, T., Flint, C. G., Gobster, P. H., Grêt-Regamey, A., Lave, R., Muhar, S., Penker, M., Ribe, R. G., Schauppenlehner, T., Sikor, T., Soloviy, I., Spierenburg, M., Taczanowska, K., Tam, J., Von Der Dunk, A., Contributions of cultural services to the ecosystem services agenda. Proc. Natl. Acad. Sci. U.S.A., 2012, 109, 8812-8819 CrossrefGoogle Scholar
Crossman, N. D., Burkhard, B., Nedkov, S., Willemen, L., Petz, K., Palomo, I., Drakou, E. G., Martín-Lopez, B., Mcphearson, T., Boyanova, K., Alkemade, R., Egoh, B., Dunbar, M. B., Maes, J., A blueprint for mapping and modeling ecosystem services. Ecosyst. Serv., 2013, 4, 4-14 CrossrefGoogle Scholar
Casado-Arzuaga, I., Onaindia, M., Madariaga, I., Verburg, P. H., Mapping recreation and aesthetic value of ecosystems in Bilbao Metropolitan Greenbelt (northern Spain) to support landscape planning. Landsc. Ecol., 2013, 29, 1393-1405 Web of ScienceGoogle Scholar
Hochmair, H. H., Spatial association of geo-tagged photos with scenic locations. In: Car, A., Griesebner, G., Strobl, J. (Eds.), Proceedings of the Geoinformatics Forum Salzburg. Wichmann, Heidelberg, 2010, 91-100 Google Scholar
Depellegrin, D., Blažauskas, N., Vigl, L.E., Aesthetic value characterization of landscapes in coastal zones. In: Baltic International Symposium (BALTIC), IEEE/OES, 2012, 1-6 Google Scholar
Gschwend, Ch., Purves, R. S., Exploring Geomorphometry through User Generated Content: Comparing an Unsupervised Geomorphometric Classification with Terms Attached to Georeferenced Images in Great Britain. Trans. GIS, 2012,16, 499-522 Web of ScienceCrossrefGoogle Scholar
Figueroa Alfaro, R.W., Evaluation of Cultural Ecosystem Aesthetic Value of the State of Nebraska by Mapping Geo-Tagged Photographs from Social Media Data of Panoramio and Flickr. Community and Regional Planning Program: Student Projects and Theses, University of Nebraska, USA, 2015 Google Scholar
Levin, N., Lechner, A. M., & Brown, G., An evaluation of crowd-sourced information for assessing the visitation and perceived importance of protected areas. Appl. Geogr., 2017, 79, 115-126 CrossrefGoogle Scholar
García-Palomares, J. C., Gutiérrez, J., Mínguez, C., Identification of tourist hot spots based on social networks: a comparative analysis of European metropolises using photo-sharing services and GIS. Appl. Geogr., 2015, 63, 408-417 Web of ScienceCrossrefGoogle Scholar
Alivand, M., Hochmair, H., Extracting scenic routes from VGI data sources.In Workshop on Crowdsourced and Volunteered Geographic Information ACM, NewYork, 2013, 23-30 Google Scholar
Tieskens, K.F., Schulp, C.J.E., Levers, Ch., Lieskovský, J., Kuemmerle, T., Plieninger, T., Verburg, P.H., Characterization of European Cultural Landscapes: accounting for structure, land use intensity and value of rural landscapes. Land Use Policy. 2017, 62, 29-39 CrossrefGoogle Scholar
Lipský, Z., Romportl, D., Typologie krajiny v Česku a zahraničí – stav problematiky, metodiky a teoretická východiska. Geografie – Sborník ČGS, 2007, 112, 61-83 Google Scholar
Romportl, D., Chuman, T., Lipský, Z., Typologie současné krajiny Česka. Geografie, 2013, 118, 16-39 Google Scholar
Mücher, C. A., Garcia-Feced, C., Perez-Soba, M., Wascher, D.M., Landscapes of Europe. In: Perez-Soba, M., Wascher, D. M. (Eds.), Landscape character areas – places for building sustainable Europe. Policy Brochure as deliverable from Eu’s Accompanying measure project European Landscape Character Assessment Initiative (ELCAS), funded under the 5th Framework Programme of Energy, Environment and Sustainable Development. Information Press, Oxford, 2005, 6-8 Google Scholar
Šedivá, A., Izakovičová, Z., Assessment of representative landscape types of Skalica district. Ekológia, 2015, 34, 329-338 Google Scholar
Lieskovský, J., Bezák, P., IzakoviČová, Z., Protection of representative landscape ecosystem of Slovakia - new landscape ecological approach. In: 10th International Multidisciplinary Scientific GeoConference - SGEM2010, www.sgem.org, SGEM2010 Conference Proceedings/ ISBN 10: 954-91818-1-2, June 20-26, 2010, Vol. 2, 717-724 pp
Izakovičová, Z., Evaluation of the changes of landscape types of Slovakia. GEOREVIEW: Scientific Annals of Stefan cel Mare University of Suceava. Geography Series, 2013, 22, 1-10 Google Scholar
Dunkel, A., Visualizing the perceived environment using crowdsourced photo geodata. Landscape and Urban Planning, 142, 2015, 173–186, https://doi.org/10.1016/j.landurbplan.2015.02.022 Web of ScienceCrossref
Lieskovský, J., Kočický, D., GIS Use in Analyze and Description of Representative Landscape Ecosystems of Slovakia. In: Geography and sustainable development: proceedings. Skopje: Macedonian Geographical Society, 2009, 469-473 Google Scholar
Bezák, P., Izakovičová, Z., Miklós, L., Moyzeová, M., Špulerová, J., Mojses, M., Kočický, D., Petrovič, F., Boltižiar, M., Hreško, J., Hrnčiarová, T., Šatalová, B., Lieskovský, J., Lehotský, M., Štefunková, D., Dobrovodská, M, Baránková, Z., Gajdoš, P., David, S., Halada, L., Oszlányi, J., Reprezentatívne typy krajiny Slovenska. Ústav krajinnej ekológie SAV, Bratislava, 2010 Google Scholar
Oteros-Rozas, E., Martín-López, B., Fagerholm, N., Bieling, C., Plieninger, T., Using social media photos to explore the relation between cultural ecosystem services and landscape features across five European sites. Ecological Indicators (in press), https://doi.org/10.1016/j.ecolind.2017.02.009
Panoramio. Panoramio API, 2017 https://www.Panoramio.com/api/data/api.html
Svobodova, K., Sklenicka, P., Vojar, J., How does the representation rate of features in a landscape affect visual preferences? A case study from a post-mining landscape. Int. J. Min. Reclam. Env., 2015, 29, 266-276 Web of ScienceCrossrefGoogle Scholar
Sevenant, M., Antrop, M., The use of latent classes to identify individual differences in the importance of landscape dimensions for aesthetic preference. Land use policy, 2010, 27, 3, 827-842 Web of ScienceCrossrefGoogle Scholar
About the article
Published Online: 2017-11-22
Citation Information: Open Geosciences, Volume 9, Issue 1, Pages 593–599, ISSN (Online) 2391-5447, DOI: https://doi.org/10.1515/geo-2017-0044.
© 2017 J. Lieskovský et al.. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. BY-NC-ND 4.0