Land consumption and urbanization in rural areas: understanding the dynamics of land take and demographic changes in rural municipalities in Catalonia, Spain

Land consumption and urbanization in rural areas. Understanding the dynamics of land take and demographic changes in rural municipalities in Catalonia, Spain Rodrigo d’Avila a , Melisa Pesoa a,* , Gonzalo García a , Javier Rocamonde b a Universitat Polit´ ecnica de Catalunya, Department of Urbanism, Territory and Landscape, Av. Diagonal, 649, 08028, Barcelona, Spain b Universidade da Coru˜ na, Urban Planning Department, Campus da Zapateira, 15008, A Coru˜ na, Spain ARTICLE INFO Keywords: Land consumption Land take Rural towns Rural depopulation Catalonia Spain ABSTRACT Several international organizations agree that land consumption is in a critical situation, yet demand for urbanized spaces persists, even in declining populations. This phenomenon has been primarily applied to large urban agglomerations, so the dynamics of land take in non-urban areas, particularly those situated far from large metropolises, should not be overlooked. The central question of this article is whether increased land consumption in rural areas corresponds to population dynamics. By examining different remote sensing datasets regarding small municipalities in Catalonia (Spain), together with the population census, it seeks to ascertain whether there is a correlation between soil imperviousness, as an indicator of urbanization processes, and demographic changes in rural context. The study provides a comprehensive examination of land consumption for 2009–2018, following the real estate crisis, in Catalonia’s rural municipalities, elucidating the complexities of achieving sustainable land use in the context of population decline. The results of the study indicate that even after the 2008 crisis, rural towns continued to grow in size, generating more land consumption, even without population growth. This stands in contrast to the post-crisis dynamics observed in Spain’s major cities. The findings underscore the importance of analyzing regional dynamics in functional areas rather than by individual municipalities, to achieve a comprehensive understanding of the process. These findings are vital for the development of future regional planning policies, ensuring the optimal allocation of resources and the provision of necessary support to municipalities for the sustainable management of their growth. 1. Introduction In recent decades, urbanized areas have grown exponentially in different patterns (Chakraborty et al., 2022; Güneralp et al., 2020; Seto et al., 2011). This growth has been at the expense of soil, which is one of the most complex ecosystems in nature and contains a quarter of the planet’s biodiversity. Soil, as a non-renewable resource, is under severe pressure not only from urbanization, but also from the expansion of agriculture, livestock, and industrial forestry. In the report entitled “Future Brief: No Net Land Take by 2050?” (European Commission, 2016), the European Commission highlights the challenges that countries around the world are facing concerning land consumption, with a particular focus on European countries (EEA, 2018). Diverse studies highlight the environmental impacts of land consumption on biodiversity (Hasan et al., 2020; Seto et al., 2012) as urban development can cause fragmentation, isolation, and degradation of natural habitats; simplification and homogenization of species composition; disruption of hydrological systems; and modification of energy flow and nutrient cycling (Alberti et al., 2003). The European Environmental Agency, UN-Habitat, and other organizations concur that land consumption is in a critical situation (European Commission, 2011; EEA, 2018), yet the demand for urbanized spaces persists, even in territories where the population is declining. This is driven by new lifestyles that require more space per capita, as well as by the competition between municipalities to attract new developments due to supposed economic income, the development * Corresponding author. Departament d’Urbanisme, Territori i Paisatge, Escola T` ecnica Superior d’Arquitectura de Barcelona ETSAB, Av. Diagonal, 649 - Edifici A - 4a planta, 08028, Barcelona, Spain. E-mail addresses: [email protected] (R. d’Avila), [email protected] (M. Pesoa), [email protected] (G. García), javier.rocamonde@udc. es (J. Rocamonde). Contents lists available at ScienceDirect Habitat International journal homepage: www.elsevier.com/locate/habitatint https://doi.org/10.1016/j.habitatint.2025.103443 Received 24 November 2024; Received in revised form 8 April 2025; Accepted 14 May 2025 Habitat International 162 (2025) 103443 Available online 23 May 2025 0197-3975/© 2025 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/bync-nd/4.0/ ). of productive activities, energy production, or the construction of new infrastructure (Colsaet et al., 2018). The concept of “land take” is open to question due to a certain ambiguity, as it includes both urbanization and the artificialization of soils (Decoville & Feltgen, 2023; Marquard et al., 2020). This has led to uncertainty regarding the monitoring and limitation of these processes, as not all artificialized areas are sealed or impervious, which could result in misinterpretation of the actual density of urban developments (Decoville, 2018; Decoville & Schneider, 2015). It should be noted that although the European Commission announced in 2011 the objective of ending net land take at the EU level by 2050 (European Commission, 2011), the implementation methods were not detailed at that time. This was formalized a decade later, in 2021, with the publication of the Soil Protection Strategy for 2030 (European Commission, 2021). It is therefore up to countries to set concrete targets for 2030. Furthermore, in 2015, the UN set a target in their SDGs to measure this: the ratio of land consumption rate to population growth rate (11.3.1 LCRPGR). 1 This links the expansion of urbanized areas to actual population growth, to measure its efficiency. Estoque et al. (2021), call attention to the lowest efficiency of land use in Europe and North America, having the highest LCRPGR for the period 1975–2000 and 2000–2015, while the Eastern and South-Eastern Asia SDG region is catching up. This means that urbanized areas are expanding at a higher rate than the population growth, something already confirmed by studies such as Schiavina et al. (2022), among others. Several studies had analyzed the drivers of these land consumption patterns in different contexts. Colsaet et al., 2018 confirm the importance of well-known drivers such as the increase of population, GDP and transport facilities, but also highlight the importance of policy factors, while other factors such as common policy instruments remain unclear. Csomos et al. (2024) underline that industrialization and population policies are propelling land take and land use change, while local climate strategies are ineffective tools to halt land take and land use change. Some research has investigated the public policies needed to achieve the 2050 target (Evers, 2024; Gradinaru et al., 2023; Lacoere & Leinfelder, 2023) and highlighted the practical difficulties involved in measuring land consumption and defining reference thresholds (Eichhorn et al., 2024; Romano et al., 2023). Moreover, data is not always available or regularly updated. However, remote sensing tools have made more data and products available, allowing better monitoring (Melchiorri et al., 2018). Despite the controversies, the concept of land take and the land use efficiency analysis has been primarily applied to large urban agglomerations (EEA, 2021, 2023; Hu et al., 2021; Masini et al., 2019; Salvati et al., 2018; Siedentop & Fina, 2012; Zoppi & Lai, 2014). The European Environment Agency reports that the number of artificial surfaces in the EU increased by 7.1 % between 2000 and 2018, primarily around large cities (EEA, 2019a) (Fig. 1). In this regard, it has been used to discuss how low-density urban sprawl has contributed to the loss of biodiversity and agricultural land (EEA, 2021; T´ oth, 2012). Nevertheless, the dynamics of land take in non-urban areas, particularly those situated far from large metropolises, should not be overlooked. Mougin et al. (2024) notes that in France, for instance, rural areas with a high degree of dispersion have experienced the most significant growth over the past five decades. This could be defined as micro-urbanization (Chai & Seto, 2019), which draws attention to the significance of small and medium-sized settlements within the wider context of global land expansion. This emphasizes the necessity for a comprehensive analysis of built-up land dynamics across the full spectrum of human settlements, ranging from rural areas to major metropolitan centers (Li et al., 2022). The drivers of hinterland occupation include mobility infrastructure, housing, public facilities, industry and logistics centers, recreational enclaves, raw material extraction sites, waste disposal, and energy production. The factors driving these occupations are diverse, ranging from public policies to cultural practices (Colsaet et al., 2018; Kati et al., 2023). Guastella et al. (2017) support the hypothesis that larger municipalities are more efficient in managing land take. Their research revealed that the marginal land consumption per new household is inversely related to the size of the municipality. This finding is consistent with the notion that, as more space is often available, small municipalities are less inclined to allocate institutional attention to the issue of land take. Consequently, they internalize less the environmental externalities associated with land use. In Spain, urban land expansion has been particularly notable over the past three decades, especially around major cities and coastal areas, where tourism and construction have had a significant impact. It is estimated that between 2000 and 2018, Spain consumed 2474.12 km 2 of natural land, mainly for urban and infrastructure use (EEA, 2019b). The planning background will be further developed in section 2.1. It would be reasonable to assume that more urbanization and town growth have resulted in demographic growth and socioeconomic development. However, the expansion of urbanization has not always been accompanied by such changes. Despite the growth of urbanized land, rural communities continue to lose population. It is, therefore, important to assess the relationship between urbanization through soil sealing and demographic changes in rural municipalities. This study aims to determine if there is a correlation between soil imperviousness, as an indicator of urbanization processes and demographic changes in rural municipalities in Catalonia. This allows us to either refute or support the idea that building more is the solution to economic stagnation, which has been the guiding principle in Spain for at least the last 40 years (Naredo, 2014). The study period, between 2009 and 2018 covers a singular phase of Spanish urbanization, which is the period after the real estate crisis of 2008 and the COVID pandemic. Urbanization, when considered alongside other factors, such as population growth or decline, could serve to identify trends and make Fig. 1. Land converted to urban areas in the EEA-39, 2000–2018. Source: EEA. 1 The National Institute of Statistics of Spain has developed a methodology for building this indicator. However, results are still not available. See https:// www.ine.es/dyngs/ODS/objetivo.htm?id=4907. R. d’Avila et al. Habitat International 162 (2025) 103443 2 comparative studies (Grabska-Szwagrzyk, Hashemvand Khiabani, Pesoa-Marcilla, Chaturvedi, & De Vries, 2024). While urban growth does result in increased revenue for municipal administration in the form of property tax, this is not the case when coupled with the current trend of population decline in rural areas. This results in significant maintenance costs for underutilized infrastructure and services and a considerable environmental impact. We are currently facing a paradigm shift that involves changes in the environmental sphere, as well as in strategies and objectives that promote less extensive city models, in line with the objectives of the 2030 Agenda, or the European ‘No net land take’ objectives for 2050. These programs advocate reducing the consumption of undeveloped land, to reduce the loss of environmental services essential for the sustainability of human life in the territory. Therefore, it seems that the changing times we are experiencing demand urgent solutions to ensure the adaptation of planning that has been excessively static. This is particularly relevant to land for development and unconsolidated urban land and, as we intend to investigate in this article, to rethink territorial strategies that promote urban planning that allows us to deal with scenarios of demographic decline (Amat, 2015; Bl´ azquez, 2006) and reduce land consumption. The following section outlines the developed methodology, detailing remote sensing (RS) data and demographic indicators to develop an index for comparing land consumption across rural municipalities. The results section presents an analysis of changes in urbanized areas. The discussion section offers two perspectives on these findings. First, it contrasts individual trajectories with regional patterns. Second, it examines the relationship between land consumption and vitality in rural municipalities. Third, we examine the transformation in relationship to the public policies that act as drivers and the ones that try to stop land consumption. Finally, conclusions summarize the main findings of the study in relation to the initial question, highlighting the paradox of rural areas losing population while consuming more land. 2. Materials and methods 2.1. Study area: background and current situation Catalonia is one of Spain’s 17 autonomous communities, with a land area of just over 32,000 km 2 and a population of approximately 8 million. The region is subdivided into 947 municipalities, 595 of which are considered rural by Law 45/2007 of December 13, 2007 for the sustainable development of the rural environment (hereinafter LDSMR). The law defines rural municipalities as those with a resident population of less than 5000 inhabitants and integrated into rural areas, defined as geographical spaces with a density of less than 100 inhabitants per km 2 . Fig. 2 demonstrates the percentage of municipalities with fewer than 5000 residents. To create this cartography, the nine territorial planning areas into which the Catalan territory is subdivided have been taken into account. The data indicate that 72 % of the territory is inhabited by only 9 % of the population (INE, 2024). 2 In Catalonia, there has been a clear trend towards population concentration on the coast over the last hundred years, where the greatest economic dynamism is located. Likewise, in the last 20 years, there has been a progressive growth of the county capital cities in the interior. The regressive trend of the rural population that began in the mid-20th century eased from the mid-1990s until 2008, thanks to foreign immigration and its wide territorial distribution (Bayona Carrasco & Gil Alonso, 2011; Pujadas & Bayona, 2016). A cultural change related to a return to the natural environment and the growth of tourism, the reorientation of traditional activities and the increasing protection of natural areas also contributed to slowing down the trend. However, the economic crisis of 2008 brought this trend to a halt, causing more than half of the Catalan municipalities to experience a decline in population due to continuous migration from rural and small municipalities to larger ones (Aldom` a, 2022). This trend was also followed by foreign immigrants, who had contributed substantially to the demographic growth of the smaller enclaves. Consequently, a regressive dynamic has been observed in many rural municipalities in recent years, particularly pronounced in smaller municipalities (Gil & Bayona, 2021). This further accentuates the territorial disparities between the Catalan coastal area and the inland regions. Despite the economic recovery that took place in the five years before the health crisis of 2020, no general demographic recovery was observed for the whole Catalan territory (Aldom` a, 2022). In this context, most rural municipalities in Catalonia are experiencing an aging population, a low birth rate and low immigration (IDESCAT, 2023), leading to a demographic trend towards stagnation or even decline in a significant portion of the territory. However, given this demographic situation, there has been an increase in the construction of new housing in rural municipalities. Concerning urban planning, the 1956 Land Law is the precursor of the current urban planning model in Spain (although successive land laws were subsequently drafted in 1976, 1992, 1998, 2007–2008, 2013 and 2015, which is currently in force). The 1956 Law classifies land into urban, urban reserve (land for development) and rustic categories. Urban classification has been and continues to be a determining factor in the entire legal regime of land and in the definition of the territorial model, since it implicitly predetermines and directs urban development actions towards land that is ‘apt’ to be transformed for urban development (Agudo, 2010). Land classification was the main mechanism of municipal urban planning practice during the so-called ‘prodigious decade’, between 1997 and 2007 (Amat, 2015). It was a strategy based on promoting excessive patterns of urban land consumption, under the ideology of land liberalisation. The argument was that planning had been very Fig. 2. Catalonia map highlighting the rural municipalities and planning regions (in colors, on the right side) defined by the Catalan Government. 3 1 Source: Own elaboration based on data from INE and ICGC. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) 2 Of the 32.115 sq km of Catalonia, rural municipalities occupy approx. 23.180 sq km. Of the 8.034.743Catalonia inhabitants, 788.015 live in rural municipalities. Source: Instituto Nacional de Estadística, 2024 R. d’Avila et al. Habitat International 162 (2025) 103443 3 restrictive and by making more land available for development, more housing would be built and therefore the price of housing would fall (de Santiago, E. and Gonz´ alez, I. 2021a). This strategy was widely used until the 2008 real estate crisis in Spain. It was in large cities where its consequences became more visible, with an oversized land consumption, far beyond the mere satisfaction of the expected demand, as several researches have shown (G´ orgolas, 2019; Gaja, 2008; Calder´ on Calder´ on & García Cuesta, 2017). However, the practice of land classification also had important consequences in small municipalities, whose impact has been much less studied than in large ones (de Santiago, E. and Gonz´ alez, I. 2021b). Given this situation, it is possible that new principles of urban planning, or greater versatility of urban regulations, may be necessary to adapt to a horizon of no-growth, of urban containment, that adjusts to the real needs of the different types of settlements, and that uses the accumulated housing and urbanisation stock (Gaja, 2015). There are few studies that have analyzed what has happened to developable land and undeveloped urban land (Zamb´ on et al., 2017), despite the fact that the question of what to do with this heritage has been raised for years (Nel⋅lo, 2012). 2.2. Data The analysis was conducted by combining two types of data, as detailed in Table 1. The first was remote sensing data (in raster format), and the second was statistical data (tables). The raster images provide a combination of data obtained through remote sensing, offering valuable insights into changes in land cover between 2009 and 2018, as well as the uses of the land. The statistical data on population variation for the study period was determined based on the availability of data on land impermeabilization so the study period covers from 2009 to 2018, addressing a particularly interesting phase in Spain’s urbanization dynamics, as the 2008 real estate bubble burst, marking the end of the socalled “prodigious decade” during which there was an accelerated process of housing and infrastructure construction (Burriel de Orueta, 2008). From 2009 onwards, the economic crisis deepened, with signs of recovery only emerging from 2015 onwards. All data used in this research are open and accessible, allowing administrations to obtain a clear picture of the situation in each area at a supramunicipal level. To enhance and contextualize data visualization, the data is projected onto the digital terrain model obtained from the Cartographic and Geological Institute of Catalonia. 2.2.1. Land cover dataset from ICGC The ICGC Land Cover dataset provides detailed cartographic information on land cover types in Catalonia, including urbanized areas, agricultural land, forests, and water bodies, with a spatial resolution of 1 m and a thematic breadth of 41 categories. This dataset is available for 2009 and 2018, providing valuable insights into land use changes over time. The 2009 land covers are a simplification of the Catalonia Land Cover Map v4 of 2009, developed by CREAF and adapted to the data model approved by the Cartographic Coordination Commission of Catalonia. On the other hand, the 2018 land covers are an update derived from the photo interpretation of the 2018 ICGC orthophotos, incorporating changes identified through comparing images, topographic base data, forest fire databases, and other sources. The ICGC Land Cover dataset maintains methodological consistency across study periods through photointerpretation and digital screen digitization, supported by cartographic resources. The 2018 updates integrate manual photointerpretation with semi-automatic change detection using radiometric indices, deep learning, and GIS-based agricultural parcel data (SIGPAC). 4 High-resolution ICGC orthophotos (natural color and infrared) ensure classification accuracy. These validation processes enhance dataset reliability in detecting land cover changes. In this study, the ICGC land cover dataset was used to analyze changes in urbanized areas. Therefore, all types of areas classified as urbanized areas were selected for this study. By intersecting these urbanized areas with municipalities (polygons), it was possible to calculate the total urbanized area per municipality for 2009 and 2018. 2.2.2. IMCC Degree of Imperviousness change classified This information, provided by Copernicus, illustrates changes in imperviousness or land sealing on a pan-European scale with a spatial resolution of 10 m in raster format. The categories defined are: new imperviousness, imperviousness loss, no change, increased imperviousness, decreased imperviousness, unclassifiable, and no imperviousness. Changes in imperviousness or land sealing are primarily based on NDVI (Normalised Difference Vegetation Index) analysis. These layers are derived products from the base information called IMD Degree of Imperviousness. The time series of the High-Resolution Imperviousness Degree product started in 2006, with updates every three years. Throughout this time series, there have been repeated methodological adjustments and technical improvements. 5 To ensure the homogeneity and continuity of the High-Resolution Imperviousness Degree (HRL IMD) time series, a reprocessing of the historical products of 2006, 2009 and 2012 was carried out in 2015, using the most advanced methodology. With the change in spatial resolution from 20 m to 10 m in the recent production of the 2018 products, another significant methodological adjustment was made. In addition, using high-quality data from Sentinel-1 and Sentinel-2, with their high spatial and temporal resolution, allowed a significant improvement in quality compared to historical products. This improvement results in a spatial differentiation between built and unbuilt areas, which was not possible in the previous production due to the limited quality of the input data. Therefore, with the increased spatial resolution, the current HighResolution IMD 2018 is no longer directly comparable with historical layers (2015 and earlier). Direct comparability and analysis of changes are only guaranteed for the IMCC Degree of Imperviousness change product at 20 m resolution, which distinguishes between actual changes and technical improvements due to the increase in spatial resolution of input data and final products. As the proposed analysis only requires a distinction between whether there has been an increase in imperviousness or not, and does not require knowledge of the degree of change, we have chosen to use the categorized change layer (IMCC Degree of Imperviousness change classified). Table 1 Data used, sources, and period availability. Source: Own elaboration. Dataset Type Source Spatial resolution Years Land cover Raster ICGC 1 m 2009 y 2018 IMCC Degree of Imperviousness change classified Raster Copernicus 20 m 2009, 2012, 2015 y 2018 Population CSV INE Anual, 2009; 2018 Digital Elevation Model (for visualization) Raster ICGC 5m 2020 4 Technical specifications of the IGCG’s Catalonia land cover dataset: htt ps://www.icgc.cat/ca/Territori-sostenible/Cobertes-del-sol. 5 Imperviousness Change User Manual: https://land.copernicus.eu/en/pro ducts/high-resolution-layer-imperviousness/imperviousness-classified-change2009-2012. R. d’Avila et al. Habitat International 162 (2025) 103443 4 The datasets used in this study cover different aspects of urbanization, ranging from the construction of new housing and public facilities to the development of new infrastructure. Overall, they focus more on change than on land classification. In this respect, following the concepts of the UN-Habitat report on land use efficiency, which defines land consumption as the “conversion of land from other uses to urban functions” (UN-Habitat, 2018), this study uses data from different sources and methodologies to find a reliable way to detect changes in land cover in Catalonia (Fig. 3). 2.2.3. Statistical data The statistical data used in this study come mainly from the Spanish National Statistics Institute (INE). This dataset includes the annual population census. 2.3. Methods The latest population census data has revealed which municipalities have experienced a decline in population over recent years. In addition, studies on Catalonian ground cover have delineated urbanized areas in both 2009 and 2018. By cross-referencing these datasets, we were able to identify the municipal areas that experienced a loss of population while still undergoing urbanisation. Subsequently, studies on Imperviousness Classified Change, conducted by Copernicus and available for all of Europe, provided enhanced visualization of areas that underwent changes in imperviousness during specific periods (2009–2012, 2012–2015, and 2015–2018). To analyze the relationship between urbanisation and demographic changes, we adapted key indices, including the Population Growth Rate (PGR), which tracks the percentage increase or decrease in population over time; the Land Consumption Rate (LCR), which measures the rate at which urban land expands; and the Land Consumption per Capita (LCP), which assesses the amount of land used per person. The UN Habitat Land Use Efficiency Report (2018) introduces these key indicators. When combined, these metrics provide a comprehensive framework for evaluating sustainable urbanisation, highlighting areas where land use is efficient or excessive relative to population growth. PGR =(LN(Pop(t2)/Pop(t1)) / (y) Where: Popt1 is the total population within the urban area in t1 (initial year), Popt2 is the total population within the urban area in t2 (final year) and y is the number of years between the two measurement periods. LCR =(LN(Urb(t2)/Urbt1) / (y) Where: Urbt1 is the total area covered by the urban area in the initial year t1; Urbt2 is the total area covered by the urban area in the final year t2 and y is the number of years between the two measurement periods (t1 and t2) LCPCt1 =UrBUt1 / Popt1 Where: UrBut1 is the total built up area within the defined t1 urban boundaries; Popt1 is the total population within the t1 urban boundaries. Which culminates in final indicators, such as: 6 LCRPGR (LUE 6 ) =(Annual Land Consumption rate) / (Annual Population growth rate) % Change in LCPC (t1-t2) =(LCPCt2 - LCPCt1) / LCPCt1 x 100 In previous studies, results have been analyzed in different ways. Schiavina et al. (2022) divided the data into categories for a more nuanced understanding of land consumption in each municipality. They explain that “an efficient development trajectory occurs when 0 <LUE ≤1, where PGR >LCR, and four classes characterize inefficient behaviors where LCR >PGR: LUE <−1; −1 <LUE ≤0; 1 <LUE ≤2; and LUE >2.” However, these calculations do not account for negative values, such as when the LCR is lower than zero, posing challenges when applying the indices to municipalities with declining land consumption. Moreover, the categorization in their study limits the findings to “worsening” or “improvement,” which may be useful on a global scale but less applicable in regional analyses like this case. On the other hand, Fujimura et al. (2022) incorporate negative values but also rely on categorical distinctions (Fig. 4). Their study, also on a global scale, uses categorization to interpret results that may be similar mathematically. For instance, positive numbers on the index can indicate both population increase with land consumption increase (upper right quadrant), as well as population decrease with land consumption decrease (lower left quadrant). While this approach works for some analyses, it was essential in our case to create a comparable parameter that could evaluate the best and worst cases within Catalonia. To solve this, we introduced a constant into the calculations, allowing us to rank all municipalities, from best to worst, across the studied indicators. This constant shifted all results to positive values, enabling us to create a clear and unified ranking of municipalities based on their performance. We kept information about land or population increase as separate indicators to maintain the original value of the data, while the value of the result itself (LUE) was changed by the constant. In that regard, we could keep all the information needed and yet compare and enlist all Catalonia municipalities from worst to best in land use efficiency. 3. Results 3.1. Understanding land use efficiency in rural Catalonia As a result of applying the methodology described, we analyzed which of these municipalities continued to expand, as evidenced by the construction of new impervious areas (high LCR), despite experiencing population decline (low PGR). Starting from this unbalanced situation, our research sought to explore the expression of this phenomenon throughout Catalonia (Fig. 5). This perspective highlighted several municipalities in different counties with strong economic links that influenced the expansion of urban settlements and infrastructure. The results illustrate a pronounced disparity between coastal and inland Catalonia, with the former exhibiting a concentration of the majority of the population and a consistent demographic growth trajectory, contributing to an upward shift in the land-use efficiency index. In contrast, inland areas are experiencing a general decline in population, which is reflected in lower index rates. Furthermore, numerous rural municipalities in Catalonia are experiencing severe depopulation, with over 300 municipalities having less than 1000 inhabitants, particularly in the most remote areas. Regional capitals and intermediate cities, such as Tortosa, Lleida, Solsona, and Vielha, represent exceptions to this situation, serving as population, economic, and service nodes amidst the broader negative trends observed in the surrounding interior regions. Furthermore, significant infrastructure initiatives (such as the A2 that connects Barcelona do Lleida) which facilitate connectivity between pivotal urban centers within Catalonia and beyond, contribute to the positive land consumption indices observed in their immediate vicinity. The municipalities situated near these transportation corridors exhibit higher population densities, even when they are classified as rural or small towns. This evidence illustrates that accessibility to regional urban centers is a critical factor in maintaining population levels. A more detailed examination of the data at the municipal level 6 LUE =Land Use Efficiency indicator, as stated at Schiavina et al. (2022) which is the same as the LCRPGR index as stated at Fujimara et al. (2022) and the UN Habitat Land Use Efficiency Report (2018). R. d’Avila et al. Habitat International 162 (2025) 103443 5 allows identifying the most and least favorable cases within the study (see Table 2). Notable examples can be found in key regions such as Ponent (Lleida), Terres de l’Ebre (Tortosa), and Comarques Centrals (Solsona). These municipalities are notable for their negative performance concerning land consumption, which is driven by high levels of depopulation and large-scale expansion of impervious surfaces. When these results are considered alongside the regional map, broader patterns and dynamics emerge. For example, Cervi` a de les Garrigues, a small municipality near the provincial capital of Lleida, exhibits a significant population loss, which reflects the rural-to-urban migration trend. Furthermore, it demonstrates one of the highest land consumption rates in Catalonia, largely due to the construction of a major ring road as part of a regional logistics and mobility plan for the Ponent region. Conversely, more favorable examples are concentrated in the province of Girona and the Barcelona Metropolitan Area, predominantly in major urban regions where the equilibrium between land consumption and population growth is more stable. These areas illustrate a more sustainable approach to urbanization, whereby population density is maintained without excessive land expansion (see Fig. 5). Fig. 3. Image showing each dataset, its level of detail and accuracy, and how they relate and complement each other. Source: Own elaboration based on previously detailed datasets. Fig. 4. Diagram of quadrants showing negative and positive values on the LCRPGR (LUE) index. Source: Adapted from Fujimura et al. (2022). Fig. 5. Land Use Efficiency in Catalonia (Green =Better, Pink =worst). Source: Own elaboration. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) R. d’Avila et al. Habitat International 162 (2025) 103443 6 3.2. Higher land consumption per capita in rural areas From a regional perspective, the map in Fig. 6 highlights another key index: the change in land consumption per capita between 2009 and 2018. This index illustrates, municipality by municipality, the relationship between urbanized land and population, indicating where there has been an improvement (a decrease in land consumption per capita) and where there has been a negative change (an increase in land consumption). Dark circles represent municipalities in poor conditions, where fewer people occupy more urbanized space, and the size of the circles reflects the size of the municipality. A notable concentration of dark circles is evident in the vicinity of Lleida (Ponent) and Tortosa (Terres de l’Ebre), highlighting the adverse land consumption dynamics in the inland areas surrounding these pivotal cities. Conversely, the coastal regions display a concentration of lighter and larger circles, which may be indicative of more favorable conditions. It is also worth noting that the darkest circles are relatively small, indicating that the most adverse conditions are found in smaller, rural municipalities. While new construction is often regarded as an indicator of economic growth, in small municipalities experiencing depopulation, this phenomenon presents a paradox. These rural areas are experiencing a net loss of population while simultaneously expanding their urban footprint. In contrast, larger urban municipalities are demonstrating greater Table 2 Land consumption in Representative cases. Source: Own elaboration. Municipality Group Planning Region Land consumption Pop. Growth PGR LCR Ch_LCPC LCRPGR Cervi` a de les Garrigues 500–2.000 Ponent 57 % −23 % 0,9704 1,0502 1,0500 1,0822 Godall 500–2.000 Terres de l’Ebre 3 % −30 % 0,9608 1,0037 0,4700 1,0446 Vilanova de la Barca 500–2.000 Ponent 15 % −18 % 0,9774 1,0151 0,4000 1,0385 Ginestar 500–2.000 Terres de l’Ebre −1 % −29 % 0,9627 0,9992 0,3900 1,0380 Vilaller 500–2.000 Alt Pirineu i Aran 0 % −25 % 0,9860 1,0022 0,3400 1,0333 T´ ermens 500–2.000 Ponent 21 % −10 % 0,9883 1,0210 0,3400 1,0330 Flix 2.000–5.000 Terres de l’Ebre 6 % −15 % 0,9822 1,0066 0,2500 1,0248 Artesa de Segre 2.000–5.000 Ponent 7 % −11 % 0,9872 1,0078 0,2000 1,0209 Cardona 2.000–5.000 Comarques Centrals 1 % −10 % 0,9879 1,0016 0,1300 1,0139 Rajadell 500–2.000 Comarques Centrals 20 % −9 % 0,9912 1,0092 0,1000 1,0105 Almenar 2.000–5.000 Ponent 2 % −6 % 0,9933 1,0022 0,0800 1,0090 Castellet i la Gornal 2.000–5.000 Pened` es 4 % 1 % 1,0012 1,0041 0,0300 1,0028 Peralada 500–2.000 Comarques Gironines 4 % 3 % 1,0038 1,0049 0,0100 1,0011 Soses 500–2.000 Ponent 3 % 3 % 1,0031 1,0028 0,0000 0,9998 Castellolí 500–2.000 Pened` es 12 % 21 % 1,0208 1,0129 −0,0700 0,9923 Fig. 6. Population by municipality (size of the circle), and Land consumption per capita change (Dark Grey >0, Light Grey <0). R. d’Avila et al. Habitat International 162 (2025) 103443 7 efficiency in land use, with a reduction in the amount of built space per person –even though this lack of expansion is few times due to the unavailability of more land for development. The findings of this study provide a comprehensive examination of land consumption trends in Catalonia’s rural municipalities, elucidating the complexities of achieving sustainable land use in the context of population decline. 4. Discussion 4.1. Individual trajectories and regional patterns To gain a deeper understanding of these dynamics, rural municipalities were classified according to their planning regions (Fig. 2), and visual representations of population sizes were incorporated for each municipality. This approach enables us to identify “positive” and “negative” cases, as depicted in previous maps, and to uncover whether certain trends are more prevalent in larger or smaller municipalities. This provides valuable insights into emerging land-use patterns. The scatter plot (Fig. 7) provides a visual representation of trends within each planning region, while also identifying outliers. The points, representing municipalities, originate from a central point, with each quadrant reflecting a different trend. The upper-left quadrant shows positive population growth and reduced land consumption, representing the best cases. The upper-right quadrant shows positive values for both indexes. The lower-left quadrant reflects negative values for both indexes. The lower-right quadrant highlights the worst cases, with positive land consumption paired with population decline. Notably, Cervi` a de les Garrigues, T´ ermens, and Vilanova de la Barca (all in the Ponent region) stand out negatively due to high land consumption. Following, Terres de l’Ebre, represented by Godall and Ginestar, shows a predominantly declining population paired with minimal urban expansion, also resulting in a low land-use efficiency rate. Regions like Pened` es (Castellolí) and Comarques Centrals (Rajadell) have experienced both high population growth and high land consumption. However, in Comarques Centrals, there is an imbalance, with land consumption outpacing the population growth rate. In a different context, Alt Pirineu i Aran (Vilaller) has seen one of the sharpest declines in population, though without significant land consumption concerns, which can indicate severe economic stagnation. The Metropolitan Area demonstrates a centralized trend with positive correlations between population growth and land consumption. Similarly, though with fewer urban municipalities, Comarques Gironines (Forti` a) exhibits population growth along with modest reductions in land consumption, a trend also seen in several rural areas (Table 3). It is a well-established fact that some urban areas experience population growth, which in turn necessitates the construction of new housing to meet the rising demand. However, the situation in rural municipalities presents a different and, in many cases, more complex challenge. Many of these rural areas do not experience population growth, yet they continue to build and consume land. Fig. 8 presents a second scatter plot, in which higher points on the vertical axis indicate more populated municipalities. Movement from left to right on the horizontal axis represents increasing amounts of urbanized areas, which correlates with higher levels of land consumption. The red line represents Catalonia average number of population growth and land consumption (0.1 Ha per inhabitant in the years studied), reflecting a pattern of urban concentration as a reference point. By examining the trajectory of each municipality’s data points, we can identify variations in both population growth and land consumption. The lines connecting pairs of points for each municipality provide a clearer illustration of these trends. Municipalities that deviate the most from sustainable growth patterns are those with nearly horizontal lines, Fig. 7. Relative land consumption and population change between 2009 and 2018 of rural municipalities in Catalonia. Source: Own elaboration. R. d’Avila et al. Habitat International 162 (2025) 103443 8 indicating disproportionate land development in relation to population growth (Soses, Castellet i la Gornal y Peralada). Of greater concern are municipalities with downward-sloping lines, which represent areas where the population has decreased but land consumption has continued to rise (Almenar, Cardona, Flix y Artesa de Segre). This unsustainable development pattern suggests that in some rural municipalities, significant land is being consumed despite a shrinking or stagnant population. In this plot, although we lack a regional pattern spatial visualization like the previous one, it highlights some individual cases that can be analyzed as individual phenomena, or even to identify their impacts on surrounding regions and counties. Thus, in order to gain a comprehensive understanding of land consumption patterns it is crucial to shift the focus from individual municipalities to a regional scale analysis. 4.2. The specificity of rural territories The results demonstrate that rural municipalities exhibit distinct characteristics compared to medium-sized regions and metropolitan areas. The rural context often includes less dense populations and fewer economic opportunities, which presents an additional challenge in balancing the costs associated with land consumption and infrastructure development. It is therefore necessary to adopt a tailored approach in rural areas, which must be distinct from the broader regional frameworks that might apply to urban centers. The Spanish government is currently contributing to the evaluation of land consumption by publishing an annual index. However, results are provided for municipalities with 20,000 inhabitants or more, provinces, and autonomous communities and cities. Therefore, results should Table 3 Population growth and land consumption by municipality group size and planning region. Source: Own elaboration. Groups Population in 2009 Population in 2018 Pop. Growth Urb. Area in 2009 (Ha) Urb. Area in 2018 (Ha) Land consumption <500 86.267 81.146 −5,94 % 14.456,25 14.364,61 −0,63 % 500–2.000 260.957 256.258 −1,80 % 25.254,48 25.054,16 −0,79 % 2.000–5.000 420.386 428.296 1,88 % 27.905,43 28.043,08 0,49 % >5.000 6.707.810 6.834.365 1,89 % 119.428,64 119.266,01 −0,14 % Regional plan Alt Pirineu i Aran 76.828 71.888 −6,43 % 5.483,15 5.445,03 −0,70 % Camp de Tarragona 513.580 515.095 0,29 % 20.740,38 20.598,79 −0,68 % Comarques Centrals 399.621 403.624 1,00 % 17.087,85 17.346,67 1,51 % Comarques Gironines 732.918 747.464 1,98 % 32.510,64 32.275,25 −0,72 % Metropolit` a 4.744.774 4.849.691 2,21 % 63.255,20 62.712,47 −0,86 % Pened` es 457.918 473.408 3,38 % 17.471,99 17.665,62 1,11 % Ponent 358.921 360.497 0,44 % 20.447,86 20.641,75 0,95 % Terres de l’Ebre 190.860 178.398 −6,53 % 10.047,73 10.042,29 −0,05 % Total general 7.475.420 7.600.065 1,67 % 187.044,80 186.727,87 −0,17 % Fig. 8. Relative land consumption and population change between 2009 and 2018 of rural municipalities in Catalonia. Source: Own elaboration. R. d’Avila et al. Habitat International 162 (2025) 103443 9