This paper analyses the potential, supply and demand of wood ES, which are elements of the ES cascade model (Potschin and Haines-Young 2011, Potschin-Young et al. 2018, see Fig. 1) at two spatial scales: continental and national. In addition, due to the availability of more accurate data, supply and demand are also examined at a third, regional scale. The statistics at continental scale provide a summary overview on the studied characteristics and are, therefore, expressed in absolute values (the total volume, m³, of wood available for supply or supplied or consumed per year on a continental scale). In the national and regional analysis, the area of the administrative region or country is taken into account (expressed in m³/ha/year). The study area covered European countries for which comparable data on potential, supply and demand were available between 2008 and 2018. A total of 24 European Union (EU) countries, as well as Switzerland, were included in the study (Suppl. material 1, Figure S1). At the continental scale, we analysed the whole of Europe as a single region, while at the national scale, we used country borders (Eurostat 2016). At the regional scale, we assessed supply and demand using nomenclature of territorial units for statistics (NUTS 3; n =1061) and local administrative units (LAU; n = 957) from year 2016 (Suppl. material 1, Table S1,Eurostat (2016)). As the geographical size of NUTS 3 regions varies greatly (min. 14.0 km² and max. 105,907.5 km²) across Europe, we combined these two administrative regions to create study units with as similar areal coverage as possible to improve their comparability (min. 0.4 km² and max. 21,762.4 km² in combined data; Suppl. material 1, Table S1 & Figure S1). The areas we used corresponded well with the statistical data collected in previous ES studies at a European scale (e.g. Ala-Hulkko et al. (2019)).

Figure 1.

The cascade model describes the flow from nature to humans through the division of ES into five steps. The original first two steps, biophysical structure and process and function, have been replaced by properties and conditions and potentials to harmonise these terms with the mapping aspects of ES (see Bastian et al. (2013), Spangenberg et al. (2014)). The relevant mapping aspects of each step in the cascade model are described below the cascade model. The wood ES indicators and their definitions used in this study are quantitative proxies for the mapped ES. The spatial and temporal scales of these indicators are shown at the bottom of the figure.

Forests cover almost 35% of Europe’s land area (Köhl et al. 2020), but are unevenly distributed across the continent. Finland and Sweden are the countries with the largest forest cover, whereas in the Benelux Regions, the forest area is the lowest in comparison to other European countries (Köhl et al. 2020, Suppl. material 1, Figure S2). Following Köhl et al. (2020), the majority of forests in Europe are dominated by coniferous species (46%), followed by broadleaved (37%) and mixed (17%).

This study uses the ES cascade conceptual framework as the basis for choosing representative data indicators for mapping the potential, supply and demand aspects of wood ES (Haines-Young and Potschin 2010, Potschin-Young et al. 2018, Fig. 1). The idea of a provisioning ES begins with the valuation of a biophysical structure in which humans can potentially use natural resources to achieve or maintain their well-being. However, in the case of wood ES, this potential does not represent the amount of the resources available for supply. The availability of a resource for supply can vary locally depending on environmental, economic or social restrictions (Levers et al. 2014, Alberdi et al. 2016). These restrictions can be related to, for example, the number of areas under protection, environmental conditions for tree growth and issues related to topography or accessibility (Alberdi et al. 2016). Nevertheless, the wood ES potential provides a background information on the total European forest resources and can be used to study the relationship between the available European forest resources and their supply and demand, as well as their changes.

Supply and demand are independent variables of the ES cascade model and their spatial representation can vary, depending on the type of ES mapped. Here, we have distinguished supply and demand from the potential, representing wood in total, as it is not common for the entire forest cover to be used annually. Consequently, total stocks or biomass cannot be an estimate of the annual wood resources used. Hence, we define supply (frequently referred in literature as ES flow) as the amount of the mobilised service within the ecosystem capable to provide a service at a given location in a certain time (Burkhard et al. 2012, Villamagna et al. 2013, Dworczyk and Burkhard 2021). As mentioned above, the supply in this study does not represent the hypothetical maximum yield of the forest; instead, actual yearly wood production (m³) describes supply more realistically. Demand is defined as the need for the ecosystem-based service by the end users (Haines-Young and Potschin 2010, Burkhard et al. 2012, Bastian et al. 2013) and serves as socio-economic focused pressure on the ecological structure. Here, finalised demand was estimated by consumption rates as the sum of services consumed by the population in a given time and space (Burkhard et al. 2012, Wolff et al. 2015, Dworczyk and Burkhard 2021). For the provisioning ES such as wood, demand is mainly driven by population size and growth, income level, technological changes, urbanisation, policy-making and the availability of alternative materials (Brack 2018). In this paper, we choose to estimate the demand for wood, from the perspective of an end user, as it is not possible to take into account the supply chain of wood, its processing and the industry effect. The demand is strictly interconnected to service supply in many ES examples. However, it is not equal to supply and serves as an independent element of ES consumption structures. Fig. 1 illustrates the the conceptual background applied in this study together with the mapped elements and use indicators.

In this study, we examine the distribution, relationships and temporal trends of wood ES potential (in continental and national scales), supply and demand (in continental, national and regional scales). In addition to the statistical comparison, most of the analyses are carried out using the space-time cube tool from Esri’s Arc GIS Pro 3.0.3. software. We decided to use the space-time cube as it provides useful tools for evaluating variations in data over space and time simultaneously. It combines statistical analysis with map visualisation tools, allowing users to explore dynamic patterns, trends and changes over time within specific geographical areas. This holistic approach supports not only informed decision-making, but also spatio-temporal understanding. It has been used, for example, in examining forest loss or the temporal effects of drought (e.g. Harris et al. (2017), Ahmadi et al. (2022)). The space-time cube, as its name suggests, can be used to analyse temporal changes in space and we performed the analysis at both national and regional levels. At the continental scale, the temporal variation was examined statistically, as we did not find it useful to perform the space-time cube analysis on a single uniform continental polygonal dataset. Therefore, space-time cubes were created from defined potential, supply and demand polygon locations (Esri 2023) corresponding to the administrative areas for which the data were gathered (national (n = 25) and regional (n = 2018)) for annual time intervals between 2008 and 2018. Summary descriptive statistics were calculated for the national and regional levels based on the potential, supply and demand values. These descriptive statistics, including mean, standard deviation and coefficient of variation, calculated as part of space-time cube local outlier analysis, were used to outline the main information on the spatial distribution of European wood ES potential, supply and demand, within the period analysed. Based on these statistics, the geographical locations with the highest and lowest potential, supply and demand levels within the period studied can be identified for different spatial scales.

In addition, with visualisation of the space-time cube, we were able to detect the temporal trends in annual levels of wood ES potential, supply and demand for individual polygons (at national and regional scales). The space-time cube tool also performs a Mann-Kendall trend test, which is a series rank correlation analysis (Mann 1945, Kendall and Gibbons 1990). This analysis was used to identify countries and regions where annual potential, supply and demand values have decreased or increased. This test compares and sums the values over time. To identify statistically significant trends, the sums are compared with the expected sum (zero, no trend). The analysis is carried out independently for each polygon location studied (countries at national scale and administrative regions at regional scale), resulting in detection of an individual trend observed within the analysed period. Potentially detected trends are then indexed with a z-score and p-value, in each analysed polygon. The statistical significance of the trend is indicated by small p-value (p < 0.01 = 99% confidence; p < 0.05 = 95% confidence; p < 0.1 = 90% confidence). Trend direction is detected, based on the z-score. A positive z-score indicates an increase in the values within the period analysed and negative z-score vice versa (Esri 2023).

Additionally, we also calculate the ratios between the potential, supply and demand at the national level, as well as the ratio between total and local supply and demand in regional level. The ratios allow us to explore the relationships between studied variables.

The mapping aspects of the ES cascade (potential, supply and demand) considered in this study are dynamic. At continental level, European wood ES potential has been gradually increasing through the period studied (Fig. 2a). The overall values of the total wood ES supply in Europe have increased over time (Fig. 2b). The greatest increase was observed from the year 2015 onwards. The overall demand values have also shown an increasing pattern and the trend of the change is stronger than overall supply (Fig. 2b). Demand has shown a slight upward trend each year since 2009. In Europe, overall supply was higher than demand and supply also grew faster than demand, with a difference of around 16% between the first and last years of the study period, while the difference in demand was around 10%.

a

b

c

Figure 2.

Amount (sum, m³) of wood ES potential. The Mann-Kendall test was used to calculate the significance of each trend. Note that the y-axes in Figures A to C are scaled differently.

a: supply 
b: and demand 
c: across European continent between 2008 and 2018 

At the national level, our results showed spatial differences in the distribution of wood ES potential, supply and demand across Europe. The availability of wood resources depends on both natural factors (climate and growing conditions) and human activities (land use and forestry). This can be seen from our results, where we found that central and northern Europe have the greatest wood ES potential (Fig. 3). These include Austria, Switzerland, Germany, Finland, Sweden and the Baltic States. The variability of the wood ES potential between years is low in all the countries studied (Suppl. material 1, Figure S5 A1). Due to lack of regional wood ES potential data, it is not possible to assess these aspects within individual countries. However, with the high capacity for supply, the above-mentioned countries are also the largest producers of the studied ES, based on supply mean values at a country level. This means that the countries mentioned above have the largest wood supply compared to potential. However, Poland, Hungary, Ireland and especially Portugal, are also particularly over-supplied compared to potential (Fig. 4a). Regionally, the high supply can be seen in the regions of southern and south-eastern Finland as well as in southern and south-eastern Sweden (Fig. 5a). Relatively high supply values were also detected in southern regions of Germany and across Austria, Slovenia, the French Region of Aquitaine and parts of northern Portugal. The lowest levels of wood supply were characteristic of southern and western Europe and the regions of Lapland in Sweden and Finland. The regions with high wood ES supply values are also characterised by high ratio between total European supply and local regional supply (Suppl. material 1, Figure S7 A). The coefficient of variation for national level supply is generally low (Suppl. material 1, Figure S5 B1). At the regional level, the variability of supply values varies, being highest in regions such as central Finland near the eastern border, south-eastern France, north-western Italy and Sicily (Suppl. material 1, Figure S6 A1). Nevertheless, the distribution of the coefficient of variation is low in most of the studied regions (Suppl. material 1, Figure S6 A2).

a

b

c

Figure 3.

The distribution of mean values of wood ES.

a: potential 
b: supply and 
c: demand at national levels between 2008 and 2018. 

Figure 4.

The comparison between (a) mean potential and supply and (b) mean potential and demand between 2008 and 2018 at national level. This ratio (in %) expresses how much of supply (or demand) is contained in comparison to the potential.

a

b

Figure 5.

The distribution of the mean values of wood ES.

a: supply and 
b: demand at regional levels between 2008 and 2018. 

National level wood ES consumption is also concentrated in central and northern Europe, although it is also high in countries such as Belgium and Portugal (Fig. 3c). This can be seen both from the mean ES demand values and the ratio between national potential and the demand (Fig. 4b). A closer look at the wood ES demand at regional level shows that it is concentrated in large population centres (Fig. 5b) and in areas where the wood industry plays a significant economic role which matches the conceptual approach of this study. Commonly, the densely populated metropolises have accumulated high needs for consumption of analysed wood ES. In general, the areas with the highest levels of demand were the southern regions of northern (Finland and Sweden) and central (e.g. Germany, Poland and Czech Republic) Europe, as well as major European cities and capitals such as Paris, Lisbon or Riga. This phenomenon is even more clearly visible when looking at the ratio between the total European demand and the demand of single regions (Suppl. material 1, Figure S7 B). The coefficient of variation for demand at national level is generally low (Suppl. material 1, Figure S5 C1). Regionally, the variation in demand values is higher in the single regions, with overall low variability across most of the regions (Suppl. material 1, Figure S6 B).

Temporal changes in supply and demand, as well as the potential of wood ES available for supply at national level, have been substantial during the period studied. While the wood ES potential has grown throughout most of the European countries (Fig. 6a), the changes in wood ES supply and demand were more country-specific (Fig. 6b, c). Greatest up-trends (at 99% confidence level) for supply were characteristic for countries such as Poland, Sweden, Spain and Denmark. The upward trend in demand was observed in countries such as Latvia, Estonia, Portugal and Hungary.

a

b

c

Figure 6.

National-scale temporal trends in wood ES. Confidence levels of the trends were used to indicate in which countries the trends were the strongest.

a: potential 
b: supply and 
c: demand between 2008 and 2018 using space-time-cube analysis. 

At a more specific regional level, the result of the space-time cube analysis also showed an increasing trend in supply and demand across many European regions (Fig. 7). The increasing supply patterns between 2008 and 2018 were dominant in a large number of the analysed regions. These includes various regions, in particular in central and southern Sweden, Poland, Ireland, in certain regions in France and in the north and west of Spain. Decreases in supply were detected in single regions, including southern Spain, northern Italy, southern Sweden and a few regions in France, Germany and Greece. Interestingly, some of these regions were also important supply hubs (Fig. 5). This includes, for example, a few regions in southern Sweden and French Aquitaine.

a

b

Figure 7.

Regional scale temporal trends in supply, and demand of wood ES between 2008 and 2018 using space-time-cube analysis. Confidence levels of the trends were used to indicate in which regions the trends were the strongest.

a: supply 
b: demand of wood ES between 2008 and 2018 using space-time-cube analysis 

Trends in demand have generally increased, but differed from supply patterns. As shown in Fig. 7b, increasing demand trends were detected in countries such as Estonia, Latvia, Poland, Romania, Bulgaria, Ireland and Portugal. Rising demands were also noticeable in northern Europe and southwest Germany. On the contrary, consumption in France decreased within the period studied.

There are various drivers of supply and demand of wood ES, both socio-economic and environmental. Levers et al. (2014) have defined those main drivers for wood harvest, referred to in this study as wood ES supply, and include: share of plantation species, terrain ruggedness, the potential of forest available for supply, forest cover and accessibility (environmental drivers), as well as country specific policy, environmental law and bio-economy factors (socio-economic drivers). An example socio-economic driver might be the forest transition, due to which the wood ES potential grows and so is the supply (Fig. 2a). Locally, the supply can be intensified in case of extreme weather events or species diseases, for example, expansion of bark beetles (Palahí et al. 2021). The demand is mostly driven, in addition to population, by other socio-economic factors such as urbanisation, changes in income and policy, development in technology and availability of alternative resources (Brack 2018). For example, the promotion of renewable energy sources in Europe during the green transition is expected to increase the demand for wood ES as a fuel (Jonsson 2013, Bull 2018). Socio-economic shock events (e.g. economic crises, wars, border closing) may cause changes in the demand for wood ES, as it affects the economic ability of the population to afford the services. An example of such an event is visible in our demand data (Fig. 2c) where we can still see the effect of low demand for wood ES after the economic crisis of 2008. Listed drivers likely have an impact on spatio-temporal variety in supply and demand values and the upward trends of ES cascade elements analysed in this study.

Our results, possibly driven by the above-mentioned factors, can directly or indirectly affect forest management, biodiversity or trade-offs between other forests ES. Forest management issues are strongly associated with economy, as the provision and consumption of studied wood ES is highly dependent on economic factors (Jenkins and Schaap 2018, Muys et al. 2022). In addition, the rising position of wood in the context of bioeconomy is making it more important than ever before (Jonsson 2013, Bull 2018). The upward trend observed in our wood ES supply, demand and potential data can possibly be associated with the growing bioeconomic importance of wood ES across Europe in the period studied. Some studies suggest that this trend in wood ES exploitation has continued for a long time and has placed wood ES above the need for the supply of other types of forest ES which can potentially affect their supply (Jenkins and Schaap 2018, Pohjanmies et al. 2021, Dasgupta 2021). Even though some studies suggest that, due to climate change, forests are predicted to grow faster and harvesting may have positive effect on the carbon balance (D’Orangeville et al. 2018), the growing wood ES supply can have a long-term impact on, for example, the habitat quality, connectivity of European forest ecosystems and can reduce the capacity to supply many other ES, such as cultural forest ES or provision of non-wood products (Mitchell et al. 2014, Pohjanmies et al. 2021). As forests are one of the most effective carbon sinks that play an important role in mitigating climate change (Palmer 2021), which is currently one of the greatest challenges facing humanity, the steady increase in wood ES potential could be a positive signal for climate change mitigation efforts. However, the growth of wood ES supply and demand levels at the same time may have an impact on the goals of increasing the amount of carbon sinks. In this study, we have only focused on one forest ES, but to obtain a more complete picture of the state of ecosystems and their ability to provide multiple services, more attention should be paid to analysing the trade-offs between different ES over time (Chaudhary et al. 2017, Pohjanmies et al. 2021).

Provisioning ES, such as wood, are typically used on a cyclical time scale. Forests, depending on species, can grow for several decades before being harvested. For this reason, the analysed example of ES is also not static and depends on temporal variation (Bastian et al. 2012, Rau et al. 2018). As forest harvesting usually takes place after decades, depending on geographical conditions, the supply of ES rarely peaks. As forests grow, the potential of ES increases continuously, while the supply temporarily remains more or less at zero until the final harvest. Therefore, forest landscapes are characterised by regular growth and harvest phases, which need to be taken into account by using temporal data when studying provisioning ES (Burkhard et al. 2014). Our study shows that there have been changes in forest use over the studied decade in many European regions. We can discuss the general impacts of these changes on forest quality or provision of other services. However, in this study, we did not consider factors such as differences in tree species, length of the growth, monocultures and country-specific forest management techniques (Palmero-Iniesta et al. 2021), that can have an impact on regional dynamics of wood ES potential, as well as supply and demand. In addition, supply data are based in official harvest statistics, which rely on the information compiled in studied countries, where the methods for collecting the data might be different. Our data consist of only the information about officially reported harvest. For example, illegal logging was not acknowledged in our data and that may have had an impact on wood ES supply levels. When it comes to demand data, we acknowledge that there might be different ways of understanding the demand for wood, for example, by including the forest industry chain into consideration. However, here, we focus on the ES concepts and analyse the demand from the perspective of the end consumer.

Further studies at different scales could be performed to explore more precise patterns of potential, supply and demand of wood ES. In addition, we only mapped the spatial characteristics of potential, supply and demand at the sites, without considering the spatial flow from the producing to benefitting areas (but see, for example, Ala-Hulkko et al. (2019)) or the effect of global wood market (e.g. trade processes related to transportation and storage) and global consumption patterns. Additionally, even though we considered national trends in wood ES potential and wood ES supply and demand, we did not measure the sustainability level of these indicators. It is often an issue in ES studies to measure what level of service use is sustainable. In this study, we discovered upward trends for all mapped elements of ES cascade. This might influence sustainable use of the resources, but more specific investigations are beyond this work. It should also be kept in mind that there is a wide variety of ways to measure and map ES indicators and although efforts have been made to develop harmonised measurement methods, it may still be difficult to compare the results of different studies (Saarikoski et al. 2015, Boerema et al. 2016).

Despite all the limitations mentioned, the results of this study present a summary of how wood ES potential, supply and demand have been allocated across Europe and how they have changed over time. Such an assessment provides an overview of current changes and the situation of wood ES in Europe and responds to the need to map a wider selection of different mapping aspects of the ES cascade model. Our study adds to the previous EU level assessments of wood ES potential, supply and demand. For example, Maes et al. (2020), although using different definitions of wood supply and demand, similarly found increasing trend in these ES characteristics, supporting our conclusions. Moreover, the methods we used proved to be an effective way to illustrate the temporal patterns of wood ES use across Europe. By analysing changes in potential, supply and demand of ES over time, identified trends lead to a more accurate assessment compared to a single year's analysis (see Burkhard et al. (2014), Maes et al. (2020)). In addition, understanding how spatial scale affects ES characteristics is crucial, as found patterns may vary with changing spatial resolution and extent (e.g. Qiu et al. (2018), Shi et al. (2023)). To acknowledge this phenomenon, we mapped the trends in three different spatial scales across Europe. In the case of ES analysed in this study, the increasing trends of production and consumption are not only visible in those countries and regions that play an important role in wood ES provision, but are also noticeable across the vast number of analysed areas. Even though we have only analysed a period of a decade, we can see a growing trend in the wood ES potential as well as supply and demand patterns. These trends may indicate changing patterns of forest use, that can further affect other ES provided by forests, their biodiversity and habitat quality.

Identifying changes in different scales is important, but regional level information is particularly valuable as it provides insights into how forests have been used over the long-term and helps to target interventions more specifically to areas of intensive forest use. The establishment of conservation plans for the areas where supply and demand are not in long-term balance, together with an assessment of biodiversity, could be a step towards achieving the goal of the EU’s Biodiversity Strategy 2030 (European Commission 2020) and help maintain the capacity of forest ecosystems to provide multiple services. Furthermore, different regions may require different policy strategies to conserve forest resources (Blattert et al. 2023). Nevertheless, more studies on other ES and their trade-offs are needed to develop a complete picture of the state of total ES production and consumption. In addition, more detailed and not too broad scale data on both the supply and demand, as well as the potential of forest ES, are needed to create a comparable system from which we can see the pros and cons of their increasing and decreasing trends.