The walkability index quantifies how pedestrian-friendly and accessible a neighborhood is for walking. Given the available data for Bulgaria, a walkability index has been developed based on the following three criteria:

• Population density: This layer was operationalized using the cadastral data for the buildings with housing function and the number of floors, which was further enriched with the number of people per address location based on the average housing area per person (1) for every city in relation to 2021 census data (NSI, 2023 (2)). Higher densities are assumed to feature higher walkability.

• Land use: This variable was derived by intersecting the cadastral (GCCA, 2016 (3)) and the European Urban Atlas data for the year 2018 (EU, 2023 (4)). We considered the following land use categories: mixed, residential, public and commercial, green and recreational, industrial, transportation and communication, agriculture, forest and natural, water, and other. Some of the land use / land cover types were grouped or divided (5). The land use diversity was quantified based on Shannon’s diversity index. The higher the Shannon index scores, the higher the land use mix. Pronounced land use diversity is assumed to support walking.

• Intersection density: 4-way crossings served as a proxy variable for urban form - street and neighborhood design. The dataset was obtained from OpenStreetMap (6) for the year 2023 and classified by comparing the classifications in OpenTransportMap, the agglomeration register (7), and classifications of the national ordinance (8). The crossings along the primary road and street network (highways, urban highways, arterial and main roads 1st – 4th class) were manually revised to exclude physical barriers such as intertwined junctions and dead-end streets touching transit sections. Then, all the primary and secondary street crossings and pedestrian alleys in green areas were aggregated based on 20 m buffers to reduce the more complex patterns of pedestrian infrastructure in public spaces – parks and gardens, squares, and intermodal hubs. It is assumed that more 4-way crossings reflect a better-connected street network supporting better walkability. Due to different units, we z-scored the three input variables (i.e., population density, land use diversity, and 4-way intersection density) and summed them within a buffer. We computed the walkability index for a regular grid, allowing us to link the data to other surveys. The grid cell size of 300 m was pragmatically chosen. Cells without a population were excluded.

References and Footnotes

1. Based on the calculation of the Gross Floor Area (GFA) for the buildings with housing function in the relevant city divided by the number of residents from the Census 2021 with the following estimates – Sofia (55 m²/p), Plovdiv (52 m²/p), Varna (60 m²/p), Burgas (48 m²/p), Ruse (51 m²/p)

2. National Statistical Institute (2022) Population by Statistical Regions, Districts, Municipalities, Settlements, Sex and Age: https://infostat.nsi.bg/infostat/pages/reports/query.jsf?x_2=1962

3. GCCA (2016). Archive cadastral data of 2016 with classification based on Ministry of Regional Development and Public Works (2016). Ordinance No. RD-02-20-5 of December 15, 2016 on the content, creation and maintenance of the cadastral map and cadastral registers - promulgation DV. issue 4 of January 13, 2017 Classifier for the landuse permanent purposes of the lots

4. EU (2020). Urban Atlas 2018 Revised LU/LC layers in Geopackage format, Country Bulgaria - cities Sofia (BG001L2_SOFIA_UA2018_v013), Plovdiv (BG002L2_PLOVDIV_UA2018_v013), Varna (BG003L2_VARNA_UA2018_v013), Burgas (BG004L2_BURGAS_UA2018_v013), Ruse (BG006L2_RUSE_UA2018_v013).

5. 01_Mixed: EU, 2020 code 11100 with GCCA, 2016 groups 2 (public buildings and complexes (codes 1100-1210)) and 7 (objects of cultural and historical heritage (codes 2000-2040)) >= 50%; 02_Residential: EU, 2020 code 11100 with GCCA, 2016 groups 2 and 7 < 50%, EU, 2020 codes 11210, 11220, 11230 and 11240; 03_Public&Commercial: EU, 2020 code 12100 with GCCA, 2016 groups 2 and 7 > 50%; 04_Green&Recreational: EU, 2020 codes 14100 and 14200; 05_Industrial: EU, 2020 code 12100 with GCCA, 2016 groups 2 and 7 < 50%; 06_Transport&Communication: EU, 2020 codes 12210, 12220, 12230, 12300 and 12400; 07_Agriculture: EU, 2020 codes 21000, 22000, 23000, 24000 and 25000; 08_Forestry&Natural: EU, 2020 codes 31000, 32000 and 33000; 09_Water: EU, 2020 codes 40000 and 50000; 10_Other: EU, 2020 codes 11300, 12100, 13100, 13300 and 13400

6. OpenStreetMap contributors (2023). Data obtained by QGIS QuickOSM plugin by Overpass API with Key ‘highway’ (Vales – ‘motorway’, ‘trunk’, ‘primary’, ‘secondary’, ‘tertiary’, ‘unclassified’, ‘residential’, ‘… _link’, ‘footway’, ‘path’, ‘track’, ‘pedestrian’, described at https://wiki.openstreetmap.org/wiki/Map_features#Highway)

7. Ministry of Regional Development and Public Works (2015) Agglomerations register under Protection From Environmental Noise Act: https://www.regag.eu/?l=2

8. Ministry of Regional Development and Public Works (2022). Ordinance No. RD-02-20-2 of December 20, 2017 on planning and design of the communication and transport system of the urbanized territories

Epidemiology Example

• Dzhambov AM, Dimitrova D, Burov A, Helbich M, Markevych I, Nieuwenhuijsen MJ. Physical urban environment and cardiometabolic diseases in the five largest Bulgarian cities. Int J Hyg Environ Health. 2025;264:114512. doi: 10.1016/j.ijheh.2024.114512.

Downloads

• walkability_300m_grid.zip