The ICP Forest Level I 2020 forest ecosystem network (16 x 16 km) was downsized to a 3 x 3 km cell and used as a base to distribute the experimental plots within the city and over its peri-urban territory. Each experimental plot covers an area of 0.1 ha and is affiliated with a specific functional zone, described in detail in the following Geospatial Analysis section.

The soil sampling was realised in 2011 on the territory of Sofia region covering the city and its surroundings. The soil samples were collected by coring in five points for each experimental plot in three soil depths: 0-10; 10-30; and 30-50 cm, but, in the present study, only the superficial soil layer is discussed. Although heavy metal content in soils could be related to soil parent material composition, the contribution from anthropogenic sources affects superficial soil layers; therefore, the determination of heavy metals content was focused on the superficial soil (0-10 cm depth) as the indicator for anthropogenic pressure effect on soil condition.

The soil samples were analysed for textural composition (pipette method titration with hydrochloric acid - HCl), pH (potentiometrically), carbon content (Corg.) according to the Turin method, for total nitrogen (N) according to Kjeldahl and cation-exchange capacity (CEC) according to Chuljian (1978). The content of heavy metals in the soil was determined after treatment with nitric acid: hydrochloric acid - HNO₃:HCI, 1:3 (ISO 1146). The metal concentrations in the extract were determined by atomic absorption spectrophotometry (AAS).

The grid with the experimental plots and the results from the samples provide information about the condition of urban soils in different functional zones (Fig. 1).

To distribute the data from the soil sampling and relate it to data for the ecosystem service “soil quality”, provided by the functional zones, the possibilities of the geographic information systems (GIS) for spatial reference and interpolation were used. The combination of data different in their genesis and characteristics provide opportunities for solving different types of research problems (Todorov and Kirilov 2022). GIS is particularly useful for mapping and assessing the value of ecosystem services in different contexts. GIS have been used to map the ecosystem services, provided by soils, across space using indicators, the ecological production function approach, in which detailed information is used to map different ecosystem services and underlying processes affecting those services in an area and the benefits-transfer approach, in which ecosystem service values from one area are transferred to the same ecosystems in other areas (Nemec and Raudsepp-Hearne 2012). Using that approach, the indicators of soil quality are mapped with respect to the Methodology for assessment and mapping of urban ecosystems condition and their services in Bulgaria (Zhiyanski et al. 2017) which is based on the corresponding EUNIS classification on level 2 (European Environment Agency 2021) combined with the National Concept for Spatial Planning on level 3 (National concept for spatial development 2013) and are proposed for detailed typology as level 3 to outline 10 urban sub-types. The sub-types were generalised into three zones according to the main functions of the urban territory and considering different land use, land cover, soil sealing, pollution and anthropogenic influence. The residential zone includes residential areas with different densities, green areas include all park territories and industrial areas as shown in Fig. 2. To include the peri-urban territory in the soil parameters analysis, an additional zone that is not covered in the EUNIS classification - urban agriculture, was included. To define this area, the Urban Atlas, Copernicus Land Monitoring Service (2018) was used, selecting only the arable land on the outskirts of Sofia. Combining these datasets gave an insight into the interdependencies that exist (or not) between the soil quality and the functions of the territory. To outline the relationship between functional zones and the ecosystem services concept applied at national level, the sub-types of urban ecosystems defined by the Methodology for assessment and mapping of urban ecosystems condition and their services in Bulgaria (Zhiyanski et al. 2017) are generalised. (Fig. 2):

The territory of Sofia region is, thus, differentiated in these four functional zones and each of them is attributed to the scores obtained for the soil quality. Interpolation was applied for matching the range of the terrain data with the range of the functional areas. The Inverse Distance Weighted (IDW) technique was used to compute an average value for unsampled locations, based on values from nearby weighted locations. The weights are proportional to the proximity of the sampled points to the unsampled location and can be specified by the IDW power coefficient (Gimond 2022). Tuncay et al. (2016) estimate IDW interpolation on spatial variability of selected soil properties and show that the interpolation results are consistent with the soil analysis results, thus enabling the extension of the obtained values to any similar non-sampled region. Robinson and Metternicht (2006) tested the performance of spatial interpolation techniques for mapping soil properties. In all uses of IDW, the power of one was the best choice, which may be due to the low skewness of the soil properties interpolated. That is why the IDW interpolation was done with the power of one for all the parameters, setting the environment's extent to the range of the functional zones. As an output, a separate raster layer that contains the distribution of the values of each parameter is generated. This raster is related to the polygon file of the functional zones. The values of the parameters are then calculated by zones. The spatial patterns of the accumulation of heavy metals related to the functions of the territory are obtained. Maps that visualise the accumulation of heavy metals in the soils in Sofia distributed by functional zones are generated.

Specific features of chemical soil properties in urban areas are contamination with heavy metals and alkalisation (Craul 1999, Yang et al. 2014). As indicators for the level of anthropogenic pressure in functional zones, the following indicators are used: the concentrations of heavy metals Cu, Zn, Pb and Fe, which are the main pollutants in the urban environment of Sofia, while soil quality is assessed by pH, total carbon (C) content, total nitrogen (N) and the cation-exchange capacity (CEC). The indicators are assessed in five categories associated with different scores from 1 to 5 (Table 1):

The assessment of each indicator was therefore summed and the mean value is used in populating the matrix for the needs and supply of ecosystem services, suggested by Burkhard (2018) and supplementing it with separate evaluation for the soil quality (reflecting the mean score on indicators pH, CEC, N and C), while the evaluation of the level of anthropogenic pressure reflects the evaluations of the accumulation of heavy metals. The matrix is presented in Table 2 and has the following structure (Table 2):

The indicators are evaluated in a scale from 1 to 5, where 1 shows the worst assessment that the indicator receives. The values of the indicators are classified in five equal intervals. These values are also juxtaposed to the Safety concentrations and Maximum permissible concentrations as regulated in Regulation No. 3 (Ministry of Agriculture 2008) on the standards for permissible content of harmful substances in soils. By using the spatial analysis Zonal Statistics, a cross-reference between the evaluation of the soil potential and the functional zones is provided.

To determine the level of anthropogenic pressure on soils in the studied functional zones, the results for heavy metals concentrations were differentiated and interpolated by using IDW interpolation and dissemination of the outcome by functional zones that include the green zone, industrial zone, residential zone and urban agriculture. The scores of indicators for anthropogenic pressure are visualised as consolidated results for each of the heavy metals, including data for the main soil pollutants as determined in the Environmental analysis in the Programme for Sofia (Sofiaplan, Sofia Municipality (2021)) and data for the transport network and power plants in Sofia, which are the main sources of air pollutants (Table 3).

The intervals correspond to the maximum permissible concentration of the heavy metals acceding to Regulation No. 3 оn the standards for permissible content of harmful substances in soils (Ministry of Agriculture 2008). All concentrations above these values are assessed as “1-Very high”.

Copper (Cu) could influence the fertility of the soil. Cu is not particularly mobile in the soil and only a small part of this element is absorbed by plants. The natural behaviour of copper in soils is to be concentrated in the surface due to its tendency to complex with organic matter (Chuljian 1978). The values of the samples are below the safety concentrations of 60 mg/kg. The lowest value of 9.34 mg/kg is detected in the industrial zone and highest accumulation in the residential zones. In the green zones, the values are in the interval 20-25 mg/kg which indicated low and very low level of pollution and anthropogenic pressure, respectively. From a spatial point of view, the concentrations are higher in north-western Sofia, which may be in relation to the underlying geology considering the determined high concentration in the deeper layers (Ballabio et al. 2018), (Fig. 3).

The concentration of iron (Fe) varied from 10270 mg/kg to 29650 mg/kg. The highest concentration is detected in the residential territories and the lowest – in the industrial and green zones, but the range of the values is small. The values of this indicator vary from high to low in all the functional zones and no clear pattern of distribution could be determined. The range of concentration values indicates a relatively low level of pollution with iron in the studied region. It could be summarised that the iron content in the superficial urban soil layer is within the limits and without exceedance of safety concentrations (Fig. 4).

The values for accumulation of Lead (Pb) range from 35 mg/kg to 130 mg/kg, which exceeds the safety concentrations of 45 mg/kg and is on the verge of the Maximum permissible concentrations. Only limited territories for urban agriculture show lower accumulation of Pb. Therefore, Pb appears to be defined as one of the main pollutants in urban soils in Sofia and needs special attention (Fig. 5).

The accumulation of zinc (Zn) varies from 52 to 130 mg/kg, keeping it below the safety concentrations, but still with predominantly high values. All the functional zones include both minimum and maximum values. No specific pattern could be determined for the distribution of that indicator. The peri-urban territories that are designated for urban agriculture show lower values of Zn accumulation, except for two locations, which are situated directly next to the industrial zones (Fig. 6).

To complete the research task, the relationship between the functional zone and the indicators for soil quality was also investigated.

The evaluation of the indicators of soil quality is based on scoring their values, including C, N and pH and CEC, following the matrix presented in Table 4 and disseminating the results into equal intervals that are evaluated with grades from 1 (very bad) to 5 (very good) (Table 4).

The intervals were used to assess the dissemination of the indicators in the territory of Sofia region.

Acidity (pH) of studied urban soils varied from 6.37 to 8.41, which refer to the neutral-alkaline soils. pH plays an important role in the stabilisation of organic matter. Furthermore, total heavy metal concentrations in the soil correlate positively with pH. This correlation is highly specific for every metal compound and highly influenced by the existence of oxygen, organic matter and soil textural fractions (especially clay minerals) (Sintorini et al. 2021). The pH is higher in the south-eastern part of Sofia, where the soils are more alkaline (Fig. 7).

The carbon content in studied urban soils is low to very low compared to the surrounding natural soils. The lowest concentration of C is observed in the eastern part of Sofia. The peri-urban territories are distinguished by the highest concentrations. In the urban territory, the green zones and the residential zones share similar values for this parameter (Fig. 8).

CEC is a useful indicator for soil fertility and, in experimental plots, it varies from 27.71 to 56.4 meq/100g. Soils, having a high CEC, change pH much more slowly. CEC is linked to the organic carbon content, clay content and type of land-use and management. Again, there is a tendency for lower values in the eastern part of Sofia and values over 30 meq/100g in the peri-urban territory are confirmed (Fig. 9).

The pattern of distribution of the total nitrogen is very like the pattern of distribution of carbon, nitrogen, CEC and soils with higher pH. For all parameters, characterising the soil quality, the same geographical pattern of poor characteristics to the east and good in the peri-urban territory is observed. The distribution within the four functional zones has no clearly defined spatial pattern (Fig. 10).

The simple matrix was completed to assess the potential of urban functional zones to sustain “soil quality”. The evaluation of each parameter by zone was summed and the mean value of the soil quality was aggregated. The results are presented in Table 5.

The soil quality parameters in the urban zone show rather ‘bad’ soil quality in the residential and industrial zone, medium potential in green zones and good potential in zones for urban agriculture. As visible from the maps, the eastern regions show lower quality (Table 6).

The anthropogenic pressure is assessed as medium in all functional zones, which demonstrates the effect of pollution on the overall urbanised region of Sofia including both urban and peri-urban territories. The scores for concentrations of Pb are poor and reduce the general score leading to worse overall evaluation.

A parallel between the soil quality and the condition of functional zones in terms of the anthropogenic pressure effect could be drawn. Green zones and urban agriculture zones have better overall evaluation than the residential and industrial zones and still have medium to good potential to sustain soil quality. It could be underlined that the demarcation is minimal and the scores have close values. The assessment shows that the soil quality is well provided by green zones and urban agricultural zones. Considering the overall anthropogenic pressure and the observed geographical patterns in the studied region of Sofia, the green infrastructure and peri-urban zones have the potential to contribute to a better condition of the soils through improvement of the main soil parameters.

Considering the assessment of the functional zones in Sofia, better potential in maintaining soil quality is determined for green zones and zones for urban agriculture versus residential and industrial zones, which are characterised by low potential.