Urban Sprawl Inside and Outside Natura 2000 Sites (SPAs) in Mediterranean EU States: The Case of Cyprus

Abstract

Land-use change associated with scattered (isolated) housing in the countryside remains largely underestimated in conventional European land-use datasets due to spatial resolution and minimum mapping unit constraints. This study quantifies low-density urban sprawl at the building level in Cyprus for the period 1993–2022, both within and outside Special Protection Areas (SPAs) of the Natura 2000 network. Situating the analysis within a broader Mediterranean EU planning context, the paper examines how local spatial patterns reflect wider development trajectories, including tourism-driven growth and second-home demand. Results reveal a fivefold increase in isolated housing outside development planning zones, from 2440 units in 1993 to 12,640 in 2022 (+418%). Significant increases occurred within agricultural zones (Γ: +568%) and even in protection zones (Z1: +438%). Within SPAs, isolated houses rose from 341 to 1556 (+356%), while total building premises within these areas increased from 955 to 3649 (+282%), indicating statistically significant encroachment. Although Natura 2000 designation appears to have moderated development rates compared to the broader countryside, it has not prevented sprawl. The findings demonstrate substantial cumulative impacts on landscapes, ecosystems, and land-use planning objectives, highlighting the urgent need for stricter regulation of dispersed houses and auxiliary buildings both within protected and non-protected rural areas.

1. Introduction

One of the biggest challenges of our times is land use change and its wide range of impacts. Urban land expansion and urban sprawl are major phenomena of our times in many parts of the world, including within the European Union. Biodiversity impacts and conservation implications are apparent, and urban land expansion is a contributing driver of habitat loss for around one-third of the species assessed [1].

Urban expansion into rural and agricultural areas has been widely recognized as a major driver of habitat fragmentation and degradation of ecosystem functions. As land is converted for built infrastructure, this process contributes to biodiversity loss, disrupts ecological connectivity, and reduces the capacity of landscapes to deliver essential ecosystem services [2]. Land take—defined as the conversion of agricultural, forest, or semi-natural land to artificial surfaces—reduces soil permeability and compromises the landscape’s ability to buffer climate extremes, including floods and heatwaves [2].

Between 2000 and 2018, the EU experienced an average annual land take of around 539 km², primarily affecting arable land and pastures [3]. While in general the rate of land take has slowed compared to earlier decades, its cumulative impacts remain significant, especially in peri-urban areas. In response, the European Commission has adopted a “no net land take” policy target by 2050 as part of its broader Biodiversity Strategy and Soil Strategy, aiming to halt further soil sealing and ecosystem loss [3].

Land take is the loss of undeveloped land to human-developed land and can also be defined as the loss of agricultural, forest, and other semi-natural land to urban and other artificial land development [4]. Since the 1950s this has been largely driven by urban sprawl and the conversion of land from non-urban to urban use. Sprawl is characterized by a decrease in urban density, a decentralization of urban functions, and the transformation of a compact urban form to an irregular, discontinuous, and dispersed pattern [5].

Furthermore, the monitoring of land take poses significant challenges, as the basic EU dataset used, CORINE Land Cover (CLC), underestimates urban fabric [6]. The CLC was built to map European land cover and land use using a minimum mapping unit (MMU) of 25 hectares and linear features over 100 m wide (i.e., 100 × 100 m in the best of cases, based on CLC [7,8]. Therefore, many land use changes at the building level are not included in the analysis provided. To address these limitations, the Copernicus Land Monitoring Service was set up and became operational in 2015, but still the minimum mapping unit is 1 ha in rural areas [6]; thus, many land use changes and particularly discontinued land use and spatial changes in the countryside are not recorded. Even the Copernicus Pan-European high-resolution data (20 × 20 m) on the degree of land imperviousness does not fully quantify urban sprawl at the building level. Therefore, existing valuations and impacts of land take and land use changes are largely underestimated. This study overcomes this limitation by recording all building premises regardless of their size.

Studies have shown that land artificialization continues to occur both outside as well as within Natura 2000 sites [9,10], but the analysis undertaken is not at the building level. Here we examine one major component of residential and urban sprawl at the detailed building level, that of dispersed/isolated housing development and other building premises in the Cypriot countryside and its implications.

Development activity in Cyprus outside city development boundaries is regulated mainly through the Policy Statement [11], whereas several countryside areas are regulated by specific Local Plans that retain more or less the same approach. This policy states that one of its main objectives, at least in theory, is the control and regulation of dispersed development in the countryside while protecting the environment, as well as the protection of the traditional and aesthetic character of rural settlements. A key pillar of this policy Statement is the avoidance of dispersed construction.

Dispersed development and the policy of isolated housing were initially introduced in Cyprus during the 1990s as a social measure to assist less privileged, land-owning households in acquiring affordable housing. This was known as the “memonomeni katoikia” (single dwelling isolated housing) policy. Thus, the construction of houses outside city or community (village) development boundaries was, in general, cheaper and gave the opportunity to many individuals who owned land property outside community development boundaries to construct houses.

But the adoption of this policy decision disregarded the environmental problems of dispersed houses and the encroachment of building activity in the wider landscape. The wide range of negative spatial problems have been largely disregarded and not taken into consideration.

This aligns with observations in other contexts: in the areas of Cyprus not under the effective control of the Republic of Cyprus, communities have even embraced sprawling housing development for its short-term benefits, sacrificing social cohesion and environmental quality in the process [12].

Nevertheless, various review reports of the Cyprus Ministry of Interior, which is the responsible ministry for the policy statement and its implementation, identify the challenge of isolated and dispersed development in Cyprus [11]. It clearly reiterates the necessity to achieve compact and concentrated building development as well as to discourage scattered development with the aim of preventing negative impacts on the environment, biodiversity, ecosystems, and the landscape. Especially with respect to isolated housing, attention is given with the aim to discourage scattered development in the countryside and to preserve environmentally important areas and productive agricultural land, at least in theory.

Uncontrolled urban sprawl has been widely documented to degrade natural habitats, reduce soil fertility, increase water runoff and pollution, and contribute to the permanent loss of productive agricultural land [2,13]. These impacts not only threaten biodiversity and ecosystem services but also undermine long-term food security and landscape resilience [2].

Although the problem of scattered/isolated housing was identified, it has never been quantified in a systematic manner at the detailed building level; therefore, this is what we have aimed to do in this paper. We have also quantified the environmental impacts of this policy and the total building encroachment that has occurred within Natura 2000 sites, and particularly within the 32 Special Protection Areas (SPAs), which are the areas with a terrestrial component that have been designated in Cyprus under the Birds Directive (2009/147/EC). These areas constitute 27.1% of the total terrestrial area of the island, which is under the effective control of the government of Cyprus [3].

The Natura 2000 network is the centerpiece of European Union nature and biodiversity policy, established to ensure the long-term survival of Europe’s most valuable and threatened species and habitats. It was founded under two landmark pieces of legislation: the 1979 Birds Directive (Directive 2009/147/EC), which establishes Special Protection Areas (SPAs) for wild birds, and the 1992 Habitats Directive (Directive 92/43/EEC), which designates Special Areas of Conservation (SACs). Together, these directives form an ecological network of protected areas across all EU member states, representing the largest coordinated network of its kind in the world. Its primary purpose is not to prohibit human activity entirely, but to ensure that human activities, including plans and projects, as well as land use, remain compatible with the conservation of biodiversity, maintaining “favorable conservation status” for key European ecosystems.

Development pressures within Natura 2000 SPAs in Europe:

Special Protection Areas (SPAs), which are part of the Natura 2000 network in the EU, face various development pressures that threaten their conservation objectives. These pressures stem from human activities that encroach upon these protected areas, potentially undermining their ecological integrity and biodiversity. The primary development pressures include urbanization, agricultural activities, and socio-economic challenges in rural areas. Each of these pressures presents unique challenges to the management and conservation of Natura 2000 sites.

Urbanization: Urban sprawl is increasingly affecting Natura 2000 sites, particularly small, high-value areas near urban clusters [10].

Agricultural Activities: Intensive cultivation and related practices threaten key habitat types within the network, especially coastal and halophytic zones [14].

Socio-Economic Challenges: Depopulation and limited rural opportunities fuel land-use conflicts and resistance to conservation, which is perceived as restrictive [15].

Urban sprawl poses a particularly insidious threat, as growth can occur incrementally and indirectly—through gradual infrastructure expansion, second homes or other built premises, or land fragmentation—without immediately breaching legal thresholds. This “silent” urbanization can degrade ecological connectivity and edge habitats, especially in peri-urban Natura 2000 sites or regions lacking clear zoning enforcement. Addressing urban sprawl thus requires both spatial planning reforms and proactive landscape-scale management to preserve ecological coherence across protected and surrounding areas.

While dispersed housing and urban sprawl have begun to receive increased research attention in Cyprus and the broader Mediterranean, a “dialogue” between current studies reveals a critical monitoring gap. For instance, whereas macro-scale EU analyses [10] suggest only “incipient sprawl” within Natura 2000 sites due to coarse data mapping, more granular regional studies in Spain [16] and in areas of Cyprus not under the effective control of the Republic of Cyprus [12] indicate intense development pressures that these broad datasets fail to capture. Our research enters this debate by comparing these coarse-grained findings with our building-level quantification, addressing the “invisibility” of scattered development that current literature identifies but rarely quantifies. Consequently, this study does not merely list new data but demonstrates how current monitoring standards [10] fundamentally underestimate habitat degradation in SPAs. By contrasting the doubling of urbanized areas in Spanish parks [16] with the lack of master plans in areas of Cyprus not under the effective control of the Republic of Cyprus [12], we highlight the importance of examining these cumulative impacts through a high-resolution lens in the Cypriot countryside specifically.

The phenomenon of urban sprawl is characterized by the gradual and scattered expansion of urban areas into rural lands, leading to habitat fragmentation, loss of biodiversity, and increased human pressures on protected areas. This issue is exacerbated by socio-economic factors and inadequate urban planning, which fail to mitigate the adverse effects of urban expansion. To address these challenges, various mitigation measures have been proposed and implemented across Europe. These measures aim to balance urban development with the conservation of biodiversity within Natura 2000 sites.

Development Pressures from Urban Sprawl:

• Extent of Urbanization: Urban cover within Natura 2000 sites is significantly lower than outside, yet the growth rate of urbanization within these sites is slightly higher, indicating an incipient sprawl into protected areas [10]. This sprawl is particularly pronounced in areas surrounded by densely populated urban clusters [10].
• Impact on Biodiversity: Urban sprawl leads to habitat fragmentation and loss, which are detrimental to the species and ecosystems that Natura 2000 sites aim to protect [17]. The spread of urban areas increases landscape fragmentation and the spread of invasive species, further threatening biodiversity [18].
• Socio-Economic Drivers: Factors such as population density, road and railway density, and economic development are significant drivers of urban sprawl [18]. These factors contribute to the pressure on Natura 2000 sites as urban areas expand to accommodate growing populations and infrastructure needs.

While these pressures are significant, it is important to note that in general Natura 2000 sites are somewhat better protected against urbanization compared to non-protected areas. However, the effectiveness of these protections varies, and there is room for improvement in legal enforcement and management strategies to better address these development pressures (Concepción, 2021) [10].

Mitigation Measures:

• λlaws are necessary to prevent urban sprawl into Natura 2000 sites, especially those near urban clusters [10]. The EU Biodiversity Strategy emphasizes the need for comprehensive threat management programs to mitigate human pressures on protected areas [17].
• Urban Planning and Management: Implementing sustainable urban growth management practices can help mitigate the effects of urban sprawl. This includes the use of geographic information systems and remote sensing for environmental monitoring, as well as the enforcement of planning policies [19].
• Conservation Strategies: The mitigation hierarchy approach, which involves avoiding, reducing, and offsetting negative impacts, is crucial for maintaining habitat connectivity and preventing biodiversity loss [20]. Spatial graph analysis can aid in identifying and prioritizing actions needed to maintain connectivity in urban landscapes [20].

Evidence from different regional case studies highlights the environmental consequences of urban sprawl, including habitat fragmentation, land degradation, and loss of agricultural land, as summarized in

Table 1. Environmental Impact of Urban Sprawl in Selected Regions.

Region	Key Findings	Citation
Famagusta, Cyprus	Loss of fertile land, habitat fragmentation, increased greenhouse gas emissions	[12,21]
Nicosia, Cyprus	Scattered housing growth in buffer zone, environmental vulnerability due to climate change	[22]
Cyprus	Land use change, environmental threats, loss of agricultural land, fragmentation	[23]
Austria	Loss of forest and agricultural land to settlement expansion	[24]
Mediterranean Region	Soil degradation, biodiversity loss, water resource degradation, increased emissions	[25,26]
EU level	Urban sprawl in Natura 2000 network, Land transformation	[10]
EU level	Land lost to artificial surfaces, land imperviousness inside and outside Natura 2000 sites	[9]

.

Scattered housing and Natura 2000 in Mediterranean EU countries:

This section synthesizes available evidence on scattered/low-density residential development affecting Natura 2000 sites in Mediterranean EU countries, with emphasis on (i) how the phenomenon is described and detected, (ii) reported pressures and impacts, (iii) governance and enforcement constraints, and (iv) key evidence gaps that limit cross-country comparability [10,27,28,29,30].

Across Mediterranean countries, dispersed settlement and second-home development intersect with coastal attractiveness, tourism-led demand, and long-standing planning/enforcement challenges. The phenomenon is often framed through country-specific terms that capture both the spatial form and the governance dimension of low-density expansion, as illustrated in

Table 2. Indicative Mediterranean framings of dispersed development (terminology and dominant patterns).

Country	Common Term/Framing (Indicative)	Characteristic Pattern (as Reported in Literature)
Spain	“Urbanización”	Coastal resort/second-home enclaves; pressure on protected landscapes [31].
Italy	“Abusivismo edilizio”	Persistent illegal/unauthorized coastal construction; enforcement/regularization tensions [32].
Greece	Unregulated expansion/coastal artificialization	Tourism- and second-home-driven sprawl, including island contexts; continued impervious expansion in coastal Natura 2000 areas [28,33].
France	Mitage/dispersed development	Dispersed coastal/rural development; governance barriers in protected coastal systems [29].
Portugal	Dispersed settlement (limited evidence in supplied sources)	General recognition in governance debates, but limited Natura-specific quantitative evidence in reviewed material [30].

[28,29,31].

What the empirical evidence shows in Mediterranean settings:

Available Mediterranean evidence is dominated by (a) pan-European assessments of urbanization trends within Natura 2000 and (b) case studies using land-use modelling, imperviousness metrics, or landscape fragmentation indicators Key empirical findings from selected Mediterranean studies are summarized in

Table 3. Selected Mediterranean evidence on development pressures and impacts in/near Natura 2000.

Geography/Case	What Is Measured	Key Finding (as Reported)	Main Implication
Europe-wide Natura 2000 (pan-European)	Urban cover and growth rates using consistent datasets	Urban cover is lower inside Natura 2000 than outside, but growth rates are slightly higher—suggesting incipient sprawl into protected areas [10].	Designation reduces absolute urbanization but does not eliminate development pressure.
Spain (Madrid region protected areas)	Land-use scenarios/modelling of built-up growth	Built-up expansion continues within protected areas; scenario modelling indicates substantial future land take under baseline trends [27].	Protected status alone may be insufficient under strong metropolitan pressure.
Greece (Mediterranean coast)	Impervious land expansion in coastal settings	Continued impervious surface expansion in coastal Natura 2000 areas, associated with renewed development cycles [28].	Soil sealing and hydrological impacts can accumulate inside/near Natura sites.
Greece (infrastructure–Natura overlap)	Land artificialization linked to wind-energy siting	Substantial artificialization risk from spatial planning shortcomings where projects overlap Natura 2000 areas [34].	Large projects can create step changes in land take when planning integration is weak.
Italy (Sardinia, terrestrial Natura 2000)	Rural Buildings Fragmentation Index + mesh density	Rural buildings contribute measurably to fragmentation; relationships vary across landscape units [35].	Low-density structures can be important fragmentation drivers even without “urban fabric” expansion
Italy (detection methods)	AI-driven detection of unauthorized buildings	Evidence of emerging technical approaches to identify unauthorized buildings in fragile territories [36].	Monitoring capacity is expanding, but institutional uptake remains uneven.

[10,15,28].

Residential Development in Natura 2000 Areas: Legal and Planning Constraints:

Residential development of any form within Natura 2000 areas, particularly in planning zones not designated for development, constitutes a critical spatial-planning and environmental concern. According to the regulatory framework established by the EU Birds and Habitats Directives, plans or projects that may have a significant adverse effect on the integrity of a Natura 2000 site need to be evaluated, and if they pose a threat or may have a negative impact, as a rule, they should not be permissible [37]. This principle applies irrespective of the scale or form of development, including isolated or low-density housing.

Prior to approval, residential development proposals within Natura 2000 sites must undergo an appropriate assessment evaluating potential impacts on habitats and species, including cumulative effects, and examining alternative solutions [37]. In several member states, this assessment is complemented by parallel environmental procedures and site-specific opinions issued by competent environmental authorities [38]. Empirical evidence from European case law and planning practice demonstrates that even small-scale or scattered housing developments can generate significant cumulative pressures, legal disputes, and long-term spatial conflicts within protected areas, as illustrated by cases such as Valdecañas [39].

While the protective regime is not absolute, authorizations or derogations are subject to strict conditions. Development that could have a negative impact on Natura 2000 sites may only be authorized following the procedure set out in Article 6 of the Habitats Directive, provided that no feasible alternatives exist, that there are imperative reasons of overriding public interest, and that appropriate compensatory measures are implemented to safeguard the overall coherence of the Natura 2000 network [40,41]. The definition, implementation, and effectiveness of such compensatory measures remain complex and are widely debated in the scientific and legal literature [41].

Overall, European experience indicates that housing development in non-development planning zones within Natura 2000 areas poses substantial risks to biodiversity conservation objectives. In contexts where national planning frameworks are particularly fragmented or weakly enforced, the literature consistently highlights the need for a precautionary spatial planning approach, early environmental screening, and transparent assessment of alternatives and compensatory measures as prerequisites for any form of approval [38,42].

Measures to Address Scattered Housing in Natura 2000 Areas:

The literature identifies a set of policy instruments and measures aimed at addressing scattered and low-density residential development within Natura 2000 areas. While these instruments formally exist at both EU and national levels, their effectiveness is primarily constrained by weaknesses in implementation and enforcement [10,29,30].

At the EU level, the main regulatory framework is the Habitats Directive, which also applies to the Birds Directive. Article 6 requires appropriate assessment of plans and projects that may have a negative impact on the habitats and species. The European Commission also activates infringement procedures against noncompliant member states.

At national and regional levels, instruments typically include spatial planning legislation, zoning and building permit systems, protected area management plans, and enforcement agencies responsible for inspections. Despite this comprehensive framework, multiple studies converge on the assessment that even though policy instruments do exist, these are weakly implemented in practice [10,29,30].

Policy Measures Proposed in the Literature:

Across Mediterranean case studies, recurring recommendations focus on five main areas [10,27,28,29,30]:

• Strengthening enforcement through increased inspection capacity, deterrent penalties, timely prosecution of violations, and avoidance of repeated legalization or amnesty schemes;
• Improving planning integration, including mandatory consideration of Natura 2000 objectives in local plans, establishment of buffer zones, connectivity-oriented spatial planning, and enhanced regional coordination;
• Enhancing monitoring by deploying systematic high-resolution observation, early detection mechanisms, public reporting tools, and regular compliance audits;
• Targeted spatial interventions, prioritizing small and peri-urban Natura 2000 sites at higher risk, enforcing strict no-build zones in critical areas, restoring degraded habitats, and applying compensatory measures where impacts are unavoidable;
• Governance reform, aimed at better integration between conservation and planning authorities, strengthening local administrative capacity, improving stakeholder engagement, and ensuring transparency in decision-making.

Implementation Gaps in the Institutional Framework for Residential Development within Natura 2000 Areas in Cyprus:

Residential development of any form in non-development planning zones within Natura 2000 areas is not institutionally foreseen under the applicable EU and national framework, unless it is assessed and demonstrated that no negative environmental impacts are foreseen. While residential, touristic, or any other spatial planning proposal at the planning zoning level is environmentally assessed using the existing national and/or EU legislation instruments, isolated houses and/or auxiliary buildings outside development planning zones are not assessed. Therefore, the planning policy of ‘memonomeni katoikia’ is essentially a planning distortion policy and a loophole that largely allows urban sprawl outside development zones without any assessment. Its practical application in Cyprus is also characterized by limited transparency, incomplete regulation, and weak enforcement. Consequently, such development reflects structural implementation gaps rather than a coherent or legally robust system of permitted exceptions [43].

The persistence of isolated housing in Cyprus mirrors broader Mediterranean “ex-post” legalization trajectories, where the formalization of informal or unauthorized structures serves as a recurring planning tool. As seen in countries like Greece, Italy, and Spain—which share the Mediterranean framings detailed in Table 2—periodic amnesty schemes often transition from “exceptional measures” to predictable administrative cycles. In Cyprus, the misuse of agricultural warehouse permits and the legacy of “defrosting” old building permits function as a parallel mechanism for gradual legalization. This study’s data suggests that Cyprus follows a Mediterranean pattern where legal ambiguity and administrative tolerance, together with the planning distortion policy aspect of ‘memonomeni katoikia’, facilitate a “de facto” urban sprawl that is eventually absorbed into the “de jure” planning system. Consequently, the high building density recorded within Natura 2000 sites is not only merely a failure of oversight, but a reflection of an institutionalized culture of amnesty common across Southern Europe, where economic and social pressures frequently override long-term environmental conservation objectives.

Evidence on Effectiveness and Limitations:

Despite the wide range of proposed measures, the literature provides limited quantitative evidence regarding their relative effectiveness. Comparative assessments across countries, cost-effectiveness analyses, and long-term evaluations of policy outcomes remain scarce. Some studies cite Germany and Portugal as examples of comparatively stronger governance frameworks, while local enforcement improvements are occasionally reported as promising but insufficiently documented [30]. Overall, the lack of systematic evaluation limits the ability to identify which measures are most appropriate and relevant. Although the problem of scattered/isolated housing was identified, it has never been quantified in a systematic manner at the detailed building level; therefore, this is what we have aimed to do in this paper. Most effectively reduce scattered housing pressures within Natura 2000 areas.

Theoretical Framework and Research Contribution:

This study is situated at the intersection of three distinct theoretical domains: urban sprawl theory, the efficacy assessment of protected areas, and policy implementation theory. Traditional urban sprawl theory often relies on macro-scale indicators of land sealing; however, this research posits that “building-level quantification” is essential to capture the “invisible sprawl” of isolated structures that aggregate into significant environmental footprints. By shifting the scale of analysis from land-cover patches to individual building units, we challenge the limits of conventional datasets (like CORINE), which often fail to detect the granular, low-density patterns characteristic of Mediterranean landscapes.

The research integrates these perspectives into a cohesive framework:

• Urban Sprawl Theory: We utilize high-resolution data to redefine sprawl not just as an expansion of urban boundaries, but as a fragmented, pervasive encroachment into rurality.
• Efficacy Assessment Theory: We assess the “protective shield” of Natura 2000 (SPAs) by measuring building density as a proxy for conservation success or failure, testing whether legal designation translates into physical exclusion of development.
• Policy Implementation Theory: We examine the “implementation gap” between the social intent of the “memonomeni katoikia” (isolated housing) policy and its actual spatial outcome, identifying how institutional loopholes facilitate luxury-driven sprawl.

The core theoretical proposition of this study is that in the Mediterranean context, the conservation efficacy of protected networks (Natura 2000) is undermined by micro-scale policy exemptions, tolerance for small-scale building structures, and the planning policy distortion of ‘memonomeni katoikia’ that favors fragmented development. Building-level quantification thus provides the “missing link” in environmental governance, offering a precise metric to evaluate how small-scale planning decisions cumulatively compromise large-scale ecological networks. This study contributes to current theory by demonstrating that without granular spatial monitoring, the administrative “success” of protected area designation may mask a physical reality of persistent habitat degradation.

2. Materials and Methods

This study employs a diachronic geospatial analysis to identify and assess the expansion of isolated buildings across the Cypriot countryside from 1993 to 2022. The methodology is grounded in the interpretation of high-resolution orthophotographic images from 1993 and satellite imagery from 2008 and 2022, allowing for temporal comparison and spatial change detection. The 1993 orthophotographic images were taken with an RMK TOP 15 type camera in analog format. The raw material was then digitized at a resolution of 14 microns. The digitized material was used to produce the final product of the orthophoto maps, with a pixel size of 50 cm. The acquisition scale was 1:8000 for lowland/plateau areas (flight altitude ~1250 m amsl) and 1:15,000 for mountainous areas (flight altitude ~2300 m amsl).

For the 2008 images, the satellite QuickBird 2 was used, with a resolution of 60 cm/pixel, Panchromatic 405–1053 nm, and NIR 715–918 NIR.

For the 2022 images, the satellite Wordlview 3 was used, with resolution (GSD) 29 cm Panchromatic, 1.24 m Multispectral (8 VNIR bands), 3.7 m SWIR (8 bands), and 30 m CAVIS.

The analysis was performed using ArcMap 10, a Geographic Information System (GIS) platform that facilitated the digitization, classification, and spatial quantification of individual building structures. The methodology follows established practice in land use change studies where diachronic imagery is used to detect and quantify settlement expansion and rural land fragmentation [44,45].

All isolated buildings in Cyprus that fall under the effective control of the Republic of Cyprus and that fall outside the city or community (village) development planning boundaries have been recorded. A second round of recording was undertaken in order to ensure no buildings were left behind. Thus, houses and building premises that fall outside any residential and holiday house development planning zones (H, Π), touristic zones (Τ), industrial zones (E and B), and livestock zones (Δ1, Δ2), outside community development city or village boundaries, were recorded.

We recorded all isolated houses located within agricultural zones (i.e., Γ1, Γ2 and Γ3) as well as planning protection zones (Ζ1, Ζ2, Ζ3, and Ζ4-ΠΤ). These areas are often vulnerable to low-density, unregulated development, a trend observed in comparable studies in the Mediterranean region [46]. In this way, an attempt was made to mark only housing dwellings that are clearly the result of dispersed (isolated) housing construction, and not any small dwellings or other premises, like agricultural warehouses or auxiliary structures. It is also recognized that within some planning zones, including within livestock zones (Δ1, Δ2), a number of houses may have been constructed, possibly irregularly, but these were not taken into consideration for this analysis.

At a second level, within the designated SPAs, a more granular analysis was undertaken. Due to the ecological sensitivity of these sites, all types of building premises have been recorded, including isolated houses, warehouses, and auxiliary structures. This aimed to assess the full spectrum of anthropogenic pressure on biodiversity hotspots and to track long-term encroachment trends within these protected areas.

For the purposes of the urban sprawl analysis, 32 SPAs out of a total of 33 sites, were taken into consideration, as one site (Oceanis) consists of only marine area. Therefore, only Natura 2000 sites (N2K) with a terrestrial component were considered.

For the statistical analysis of the data within the SPAs, we have employed descriptive statistics, graphs, and ANOVA to investigate any statistically significant differences in the number of built premises between the years (1993, 2008, and 2022). For the analysis we are using the software package IBM SPSS V20.

This approach builds upon recent remote sensing and land use change research, which has demonstrated the value of combining time-series imagery with spatial analysis for detecting urban sprawl and dispersed rural development. Recent studies have highlighted the effectiveness of combining time-series remote sensing with GIS-based spatial analysis for monitoring land-use change and urban sprawl. Global analyses demonstrate the ecological consequences of urban expansion, with spatial models showing habitat encroachment in biodiversity-rich areas [1]. While automated classification techniques continue to advance, expert-supported manual digitization remains critical for accurately identifying dispersed rural housing, especially in heterogeneous landscapes [47].

This approach, based on high-resolution imagery and parcel-level GIS mapping, offers a more detailed picture of spatial transformation—particularly essential in fragmented and informally developed rural landscapes like those of Cyprus.

Finally, this study addresses key gaps in official EU datasets such as CORINE Land Cover (CLC), which has been shown to underestimate scattered development due to its large minimum mapping unit and focus on continuous land cover categories (i.e., 100 × 100 m in the best cases based on CLC or even using the Pan-European High Resolution Layers (20 × 20 m) on the degree of land imperviousness [10]. By contrast, our high-resolution, parcel-level analysis enables a more granular understanding of spatial change, particularly relevant for Cyprus, where land fragmentation and informal rural housing are prevalent.

Table 4. Comparison of Data Sources and Techniques.

Data Source/Technique	Description	Citation
High-Resolution Orthophotos	Used for detailed habitat mapping and updating of Natura 2000 habitat maps in Cyprus.	[48,49]
Sentinel-2 Data	Provides multispectral data for habitat classification and monitoring, with applications in NaturaSat software (V2.1.7).	[50,51]
Random Forest Classification	Achieves high accuracy (91–94%) in habitat mapping using Sentinel-2 and topographical data.	[49]
Object-Based Image Analysis (OBIA)	Effective for high-resolution orthophotos, enabling detailed land cover classification.	[52,53]
Drone Orthomosaic Images	Acquired at 1 m resolution, useful for small-scale features like Ecological Focus Areas.	[49]
NaturaSat Software	Utilizes semi-automatic segmentation and deep learning for habitat identification and monitoring.	[51]
LiDAR Data	Enhances habitat mapping with structural diversity indices, improving conservation status assessment.	[54]

summarizes the main remote sensing datasets and analytical techniques considered in recent studies for habitat mapping and monitoring in Natura 2000 contexts.

3. Results

Table 5. Descriptive results and confidence intervals for the time period examined for the 32 SPAs in Cyprus.

N = 32	Year	Mean	Std. Deviation	95% Confidence Interval for the Mean		Max
				Lower Bound	Upper Bound
Number of isolated houses	1993	10.66	31.39	−0.66	21.97	146.00
	2008	32.16	49.34	14.37	49.94	229.00
	2022	48.63	64.99	25.19	72.06	268.00
Number of warehouses & auxiliary buildings	1993	19.19	24.67	10.29	28.08	102.00
	2008	56.81	52.30	37.95	75.67	211.00
	2022	65.41	60.50	43.59	87.22	233.00
Total number of building premises	1993	29.84	53.23	10.65	49.03	248.00
	2008	88.97	89.69	56.63	121.31	332.00
	2022	114.03	111.94	73.67	154.39	381.00
Density of isolated houses	1993	0.51	1.40	0.00	1.01	6.79
	2008	2.10	3.73	0.75	3.44	14.77
	2022	2.79	4.58	1.14	4.44	21.43
Density of warehouses & auxiliary buildings	1993	1.04	1.49	0.50	1.58	5.38
	2008	3.59	5.48	1.61	5.56	24.26
	2022	3.52	5.45	1.56	5.49	25.85
Total density of built premises	1993	1.54	2.61	0.60	2.49	12.18
	2008	5.68	8.21	2.72	8.64	31.82
	2022	6.31	8.32	3.31	9.31	31.82

shows the mean, the standard deviation, the 95% confidence interval for the mean number of buildings, and the maximum number of buildings per SPA for the time period examined for the 32 SPAs.

From

Table 6. Total number of built premises within the SPA sites for the time period 1993–2022 (numbers and %).

	Year			% Increase/Year
	1993	2008	2022	Increase (%) 1993–2008	Increase (%) 2008–2022	Increase (%) 1993–2022
Number of isolated houses within SPA sites	341	1029	1556	201.76%	51%	356%
Other types of building premises (warehouses & auxiliary buildings) within SPA sites	614	1818	2093	196.09%	15%	241%
Total number of building premises (isolated houses and warehouses & auxiliary buildings) within SPA sites	955	2847	3649	198%	28%	282%

it is clear that there is an increasing trend in the number of built premises within the SPAs.

To investigate whether there are any statistically significant differences in the number of built premises between the three examined years (1993, 2008, and 2022), we employed ANOVA and post-hoc tests.

Table 7. Analysis of Variance for the comparison of the means between three different years (1993, 2008, 2022) within the designated SPA areas.

N = 32, K = 3	F (2, 93)	p-Value
Number of isolated houses	4.55	0.013 *
Number of warehouses & auxiliary buildings	8.28	0.001 *
Total number of built premises	7.66	0.001 *
Density of isolated houses	3.56	0.032 *
Density of warehouses & auxiliary buildings	3.27	0.042 *
Total density of built premises	4.49	0.014 *

* The mean difference is significant at the 0.05 level.

shows the results of ANOVA. Column two shows the F-statistic and the respective degrees of freedom of the test. F (between-groups degrees of freedom, within-groups degrees of freedom) = F (k − 1, N − 1) = F (2, 93), where k is equal to the number of groups (three years) and N is equal to the sample size (32 SPAs).

According to Table 7, there is at least one pair of years that we observe statistically significant change (increase in our case), for all categories (isolated houses, warehouses, auxiliary buildings, and total built premises).

To investigate during which years we have a significant increase, we employed the post-hoc test of Tukey’s HSD (Honestly Significant Difference). Tukey’s HSD is one of the most common post hoc tests for ANOVA and can provide all pairwise comparisons.

The results of the Tukey’s HSD post-hoc test are presented in

Table 8. Multiple pairwise comparisons for the mean differences between the three different years (1993, 2008, and 2022) using Tukey HSD test.

N = 32 Dependent Variable	(I) Year	(J) Year	Mean Difference (I − J)	p-Value	95% Confidence Interval
					Lower Bound	Upper Bound
Number of households	1993	2008	−21.50	0.21	−51.56	8.56
		2022	−37.97	0.01 *	−68.02	−7.91
	2008	1993	21.50	0.21	−8.56	51.56
		2022	−16.47	0.40	−46.52	13.59
	2022	1993	37.97	0.01 *	7.91	68.02
		2008	16.47	0.40	−13.59	46.52
Numbers of warehouses	1993	2008	−37.63	0.01 *	−66.40	−8.85
		2022	−46.22	0.001 *	−74.99	−17.45
	2008	1993	37.63	0.01 *	8.85	66.40
		2022	−8.59	0.76	−37.37	20.18
	2022	1993	46.22	0.001 *	17.45	74.99
		2008	8.59	0.76	−20.18	37.37
Total number of premises	1993	2008	−59.13	0.02 *	−111.72	−6.53
		2022	−84.19	0.001 *	−136.79	−31.59
	2008	1993	59.13	0.02 *	6.53	111.72
		2022	−25.06	0.50	−77.66	27.54
	2022	1993	84.19	0.001 *	31.59	136.79
		2008	25.06	0.50	−27.54	77.66
Density for houses	1993	2008	−1.59	0.17	−3.68	0.50
		2022	−2.28	0.03 *	−4.37	−0.19
	2008	1993	1.59	0.17	−0.50	3.68
		2022	−0.69	0.71	−2.78	1.40
	2022	1993	2.28	0.03 *	0.19	4.37
		2008	0.69	0.71	−1.40	2.78
Density for warehouses	1993	2008	−2.55	0.07 *	−5.25	0.16
		2022	−2.48	0.08 *	−5.19	0.22
	2008	1993	2.55	0.07 *	−0.16	5.25
		2022	0.07	1.00	−2.64	2.77
	2022	1993	2.48	0.08 *	−0.22	5.19
		2008	−0.07	1.00	−2.77	2.64
Density for total premises	1993	2008	−4.14	0.049 *	−8.26	−0.02
		2022	−4.77	0.02 *	−8.88	−0.65
	2008	1993	4.14	0.049 *	0.02	8.26
		2022	−0.63	0.93	−4.74	3.49
	2022	1993	4.77	0.02 *	0.65	8.88
		2008	0.63	0.93	−3.49	4.74

* The mean difference is significant at the 0.08 level.

. The analysis reveals a statistically significant increase in the number of households specifically between 1993 and 2022. Regarding the category of warehouses, a significant rise is observed across two comparison periods: 1993–2008 and 1993–2022. When aggregating all built premises (including both isolated houses and warehouses), the data shows a significant overall increase during these same timeframes (1993–2008 and 1993–2022), mirroring the trends found in the warehouse category.

To overcome the problem of scattered premises, we have created three new variables that represent the density of built premises. These variables are defined as the quotient of the number of premises over the area of the SPA site (in square kilometers). All results mentioned above are confirmed using the density of the built environment. The density of the houses, for example, has again significantly increased between 1993 and 2022 (p-value = 0.03). Similarly, when using the density of the warehouses, there is a significant change (increase) between the same pairs of years as above, that is, between 1993 and 2008 (p-value = 0.07) and 1993 and 2022 (p-value = 0.08). Finally, when we combine the density of the total premises (houses and warehouses), we observe that there is a significant change (increase) between the same pairs of years as above, that is, between 1993 and 2008 (p-value = 0.049) and 1993 and 2022 (p-value = 0.02).

3.1. Isolated Housing and Other Built Premises Within SPAs

Table 6 above describes the total number of built premises within the SPAs for the time period 1993–2022 (isolated houses, warehouses & auxiliary buildings, and total) and the % increase between the years 1993 and 2022.

Results (Table 6 and Figure 1) show that a relatively large number of houses and other built premises have also been constructed within the SPAs. In 1993 there were only 341 isolated houses present within these areas, while in 2022 the numbers have increased to a total of 1556 houses. It should be noted that most Natura 2000 site designations took place during the years (2004–2010); therefore, until that time, these areas did not enjoy the additional protection status of the Natura 2000 areas. The total number of isolated houses within the SPAs constituted 13.98% of the total number of all isolated houses during 1993, 15.78% in 2008, and 12.31% in 2022, respectively.

Comparing the total number of premises for the years 1993–2008 (mainly before N2K sites were designated), the total increase of built premises was 198%, while for the same areas, the respective increase for the years 2008–2022, after the sites were designated, was 28%.

Progress of isolated houses built within the SPA site ‘Koilada Xylourikos’ (CY 5000008). Figure 2 below shows an example for the SPA site, ‘Koilada Xylourikos’ and the progress of isolated houses that has occurred between 1993 and 2022.

3.2. Results of Urban Sprawl at the Country Level per Planning Zone (1993–2022)

Table 9. Number of isolated houses for the whole Cypriot countryside per planning zone outside development zones for the period 1993–2022 (number and % of the total).

Type of Planning Zone	Number of Isolated Houses (1993)	Number of Isolated Houses (2008)	Number of Isolated Houses (2022)	% Increase (1993–2008)	% Increase (2008–2022)	% Increase (1993–2022)	% of Total Houses (1993)	% of Total Houses (2008)	% of Total Houses (2022)
Γ (1, 2, 3, Γα)	1111	3371	7430	203.42%	120.41%	568.77%	45.53%	51.70%	58.78%
Z1	494	1497	2659	203.04%	77.62%	438.26%	20.25%	22.96%	21.04%
Z2	49	21	39	−57.14%	85.71%	−20.41%	2.01%	0.32%	0.31%
Z3	143	236	382	65.03%	61.86%	167.13%	5.86%	3.62%	3.02%
Z3-ΠΠ	16	23	49	43.75%	113.04%	206.25%	0.66%	0.35%	0.39%
Z4-ΠΚΦ	61	66	75	8.20%	13.64%	22.95%	2.50%	1.01%	0.59%
AZ	141	235	267	66.67%	13.62%	89.36%	5.78%	3.60%	2.11%
Other	425	1071	1739	152.00%	62.37%	309.18%	17.42%	16.43%	13.76%
Total	2440	6520	12,640	167.21%	93.87%	418.03%	100.00%	100.00%	100.00%

shows the number of isolated houses for the whole Cypriot countryside per planning zone outside development zones for the period 1993–2022 (number and % of the total).

The results (Table 9) show that isolated housing outside community development zones at the country level has shown a very large increase between 1993–2022. Thousands of isolated houses have been constructed within Agricultural zones (Γ), thus limiting the potential of valuable agricultural land. By the end of 2022 a total number of 7430 isolated houses have been constructed within Agricultural Planning Zones, while even within Planning Protection Zones (Z1, Z2, Z3-ΠΤ and Z4-ΠΦ), a total number of 3204 houses were constructed (Table 9).

The increase of isolated houses per planning zone is presented in Figure 3.

3.3. District Comparison of Isolated Houses for the Years 1993–2022

The number of isolated houses was also analyzed per district.

The spatial distribution of isolated housing reveals significant regional disparities across Cyprus, as illustrated in

Table 10. Total number of isolated houses per district for the time period 1993–2022.

District	Number of Houses (1993)	Number of Houses (2008)	Number of Houses (2022)	% Increase (1993–2008)	% Increase (2008–2022)	% Increase (1993–2022)	% of Total Houses (1993)	% of Total Houses (2008)	% of Total Houses (2022)
Nicosia	885	1406	2395	58.87%	70.34%	170.62%	36.27%	21.56%	18.95%
Limassol	668	1732	3289	159.28%	89.90%	392.37%	27.38%	26.56%	26.02%
Larnaca	442	1111	2264	151.36%	103.78%	412.22%	18.11%	17.04%	17.91%
Ammachostos	82	237	1116	189.02%	370.89%	1260.98%	3.36%	3.63%	8.83%
Paphos	363	2034	3576	460.33%	75.81%	885.12%	14.88%	31.20%	28.29%
Total	2440	6520	12,640	167.21%	93.87%	418.03%	100.00%	100.00%	100.00%

and Figure 4. Despite its relatively smaller land area, Paphos has emerged as the primary hub for isolated building development, closely followed by Limassol. The growth trajectories between 1993 and 2022 vary substantially by district; while Nicosia experienced a comparatively moderate expansion, coastal districts in general exhibited far more aggressive development patterns.

The most dramatic surges occurred in the Ammochostos and Paphos districts. This trend suggests that sprawl in these regions is also heavily influenced by their high tourism value and the resulting demand for seasonal or second-home residences, contrasting with the more stable, in comparative terms, administrative-led growth seen in the capital district of Nicosia.

4. Discussion

The escalating problem of dispersed and scattered housing development in Cyprus should not be seen in isolation; rather, it exists in a critical dialogue with broader European land-use trends. While EU-wide reports [55,56] document a general increase in built-up areas, our finding that Cyprus holds ‘negative firsts’ in sealed surface per capita validates the ‘Second-Home Sprawl’ thesis, often applied to Mediterranean coastal zones. However, our building-level data questions the conclusions of macro-scale studies [10] that identify only ‘incipient sprawl’ within Natura 2000 sites. By quantifying a 418% increase in isolated houses for the period 1993–2022, we demonstrate that what appears as moderate growth in coarse datasets [10,55] is, in reality, a systemic penetration of residential assets into the countryside including within protection zones.

This high rate of urban growth—one of the highest in the EU [10]—reveals a significant gap between administrative protection and physical land-use reality. While European averages [56] provide a baseline for comparison (387 m²/inhabitant), the fact that Cyprus triples this average (1087 m²/inhabitant) suggests a unique socio-spatial shift from necessity to luxury sprawl that challenges the efficacy of current EU soil-sealing mitigation strategies [55]. Consequently, the dialogue between our granular findings and existing EU metrics highlights that without high-resolution monitoring, the true extent of habitat fragmentation in Mediterranean Natura 2000 sites remains critically underestimated.

As shown through our analysis, the Cyprus countryside experienced expanding urban sprawl, which contributes to land sealing and land imperviousness. Thousands of dispersed, isolated houses and other built premises have been constructed throughout the Cypriot landscape, particularly over the last decades.

Although the isolated housing policy outside community development boundaries has been a controversial policy to begin with, the problematic nature of this policy was not apparent enough or was ignored. Also, its popular acceptance by the wider public didn’t leave room for major changes at the political level, while at the same time environmental and spatial concerns have been largely ignored and disregarded.

One of the major fallbacks of the dispersed housing policy is that it has also been exploited by many developers for profit, as for the most part this policy was equipped with very limited or no controls and checks. Therefore, it was possible for many non-poor individuals, both Cypriots and foreigners, to own a country house in the countryside (often a second house), and at a cheaper price. Thus, for the most part, no social criteria were taken into consideration in the decision-making process.

Moreover, although the potential negative impacts of urban sprawl and single-dwelling isolated housing are highlighted in the Cypriot Policy Statement, this remains only a statement and an admission of fact. An attempt to restrict and regulate this problematic housing policy to a maximum radius of 500 m from existing village boundary development zones and towns a few years ago, was overturned by the Council of Ministers after its unpopularity and an outcry by some members of Parliament [57].

Moreover, residential development within Natura 2000 areas may be considered unlawful and/or unjustified when it occurs without appropriate assessment, when it is approved despite negative impacts without a legally valid derogation, when it violates non-development planning zones, or when it proceeds without or in breach of development permits. Empirical studies document recurring conflicts, contested approvals, informal construction practices, and weaknesses in enforcement mechanisms, including limited sanctions and ex post legalization practices [43,58,59,60].

While residential, touristic, or any other spatial planning proposal at the planning zoning level is environmentally assessed using the existing national and/or EU legislation instruments, isolated houses and/or auxiliary buildings outside development planning zones are not assessed. Therefore, the existing planning policy of ‘memonomeni katoikia’ is essentially a planning distortion policy and a loophole that largely allows and facilitates urban sprawl outside development zones without any assessment or thorough evaluation. Similarly, Cyprus is characterized by limited transparency, incomplete regulation, and weak enforcement. Consequently, such development reflects structural implementation gaps rather than a coherent or legally robust system of permitted exceptions [43].

It should also be stressed that the existing policy of exempting certain types of development, including agricultural warehouses, from the obligation to obtain planning and building permits, regardless and irrespective of their location even within protected areas, has become a vehicle and a loophole for the construction of holiday/scattered houses throughout the landscape. This policy is extremely problematic due to the fact that these exemptions also apply within Natura 2000 sites, irrespective of potential negative environmental impacts.

The wider environmental problems of building activity encroachment in the wider landscape have been recorded. These are directly related to the degradation and alteration of habitats, disturbance, fragmentation of habitats, increase of road networks, and land use change in the countryside, among others. It negatively affects the densities of species, their populations, their territorial areas, the chances of nesting success, and generally the biodiversity of an area [61,62,63,64].

The analysis undertaken in this paper takes a further step and quantifies urban sprawl at the building level, which is not included in previous calculations. These results reinforce and make more apparent the observed negative spatial outcomes in Cyprus, which are additive to previous analysis of land use change and sealed surfaces.

The number of dispersed houses and the rate of increase observed in agricultural planning zones and landscapes is phenomenal. As a result, many agricultural areas are self-defeating and have been essentially transformed into sparsely populated residential areas.

Planning protection zones (Z) in many instances does not hinder dispersed buildings, with the exception of Z3 and Z3-ΠΤ planning protection zones.

The observed differences among districts as years progressed, with Paphos, Larnaka, and Ammochostos districts taking the lead and Limassol still constituting a big percentage of the absolute number of isolated houses, show that these districts possible due to tourism pressures and coastal proximity but also due to the desire of non-Cypriots to acquire private houses, have experienced more pressures compared to Nicosia District.

Of particular concern is the fact that, as demonstrated, hundreds of houses and other building premises have been built within protected N2K sites. The increase in the number of isolated houses as well as the total number of building premises and the density is statistically significant, with the mean number of total building premises per SPA site increasing from approximately 30 in 1993 to 114 in 2022.

Our results show that isolated houses within the SPAs increased by more than 3 times from 1993 to 2008 (from 341 to 1029—201% increase), whereas for the same period the rate of increase was observed to be even lower in the wider countryside (from 2440 to 6520—167% increase).

But the rate of increase from 2008 to 2022 was lower within these protected areas (from 1029 to 1556 houses—51% increase) compared to the rest of the areas outside N2K sites during that time period (94% increase). Similarly, when comparing the total number of premises for the years 1993–2008, which reflect the years mainly before N2K sites were designated, the total increase of built premises was 198%, while for the same areas, the respective increase for the years 2008–2022, which reflect the situation after the sites were designated, was 28%.

It is recognized that the differences when comparing both for the within N2K sites vs. outside N2K sites may also be affected by external factors like remoteness, attractiveness, coastal proximity, accessibility, etc. Nevertheless, as demonstrated by the results, and allowing for these parameters for the same areas inside the N2K sites, the designation of Natura 2000 sites certainly had an impact in limiting the rate of increase of urban sprawl within the designated N2K sites.

The majority of terrestrial SPAs sites have been designated by 2010; therefore, it is demonstrated that the designation of SPAs sites and the subsequent implementation of EU regulations and the status of Natura 2000 sites have contributed at least to some extend and have even shifted the above trend to a lower building rate of isolated houses within the N2K sites in comparison to the rest of the countryside.

Moreover, a more detailed and specific analysis at the site level of Natura 2000 sites is deemed to be necessary so as to determine, among other things, which sites are facing the most challenges, which are in a better position, and possible reasoning for this, as well as biodiversity impacts at the site level.

The findings are in line with related studies at the EU level, which quantified land transformation and imperviousness inside and outside Natura 2000 areas [9,10,65], where in general areas within protected areas are less impacted with respect to urban sprawl and land imperviousness compared to the rest of the areas.

At the same time as with previous studies [9,10,66], our results show that land use change, including urban sprawl at the building level, is evident and very worrying even within designated protected areas.

5. Conclusions

This study advances the monitoring of urban sprawl by utilizing a high-resolution, building-level methodology that overcomes the spatial limitations of conventional European datasets, such as CORINE Land Cover, which underestimate scattered development and urban sprawl. By situating the case of Cyprus within the broader Mediterranean EU context, the research identifies critical spatial trends that challenge existing conservation frameworks.

Main Findings:

The analysis reveals a phenomenal increase in urban sprawl within agricultural and planning protection zones, effectively transforming natural landscapes into sparsely populated residential and/or holiday areas. While Natura 2000 (SPA) designation has somewhat moderated the pace of isolated building development compared to the wider countryside, it has failed to halt it. A key regional driver identified is the transition from necessity-based housing to tourism-led and second-home demand, a trend that aligns Cyprus with other Mediterranean coastal regions like Sardinia. Furthermore, the study highlights the frequent misuse of “auxiliary” or agricultural warehouse building structures as a primary mechanism for bypassing existing building restrictions as well as tolerance for these kinds of developments.

Planning and Policy Implications:

Current spatial development patterns in Cyprus are fundamentally unsustainable and contradict both national and EU-level conservation strategies. The continued reliance on the “isolated housing” policy exacerbates habitat fragmentation and depletes limited natural resources essential for climate change adaptation and other policy targets related to nature conservation. The findings demonstrate that existing legal and planning instruments and planning protection zones (with few exceptions like Z3) are insufficient to deter urban sprawl, while the cumulative impacts are disregarded. Similarly, the exemptions policy which exists for developments like agricultural warehouses, even within protected areas, facilitates and reinforces a wide range of problems including urban sprawl. This suggests a systemic failure in enforcement and a planning framework that prioritizes short-term development over long-term ecological integrity.

Future Recommendations:

To safeguard the Mediterranean landscape, a shift from passive designation to active enforcement is required. Future efforts should focus on:

• Implementing stricter legal measures and financial instruments to curb the exploitation of planning loopholes and the abolition or strict tightening of isolated housing policy, which essentially is a planning distortion policy that appeases and facilitates urban sprawl.
• Standardizing high-resolution, building-level mapping across the EU to ensure precise cross-border monitoring, thus overcoming existing limitations and underestimation of urban sprawl.
• Conducting site-specific analyses within the Natura 2000 network to understand localized patterns at the site level.

Ultimately, this study serves as a critical evidence-based alert for Mediterranean land-use management. It underscores that without a radical revision of isolated housing policies, elimination of building exemptions which apply even within protected areas, strengthening of control and enforcement mechanisms, assessment of cumulative impacts and the adoption of fine-scale monitoring tools, the ecological objectives of the Natura 2000 network, as well as other crucial policy-related aspects like climate adaptation, halting of land sealing, and landscape protection, will remain compromised by the persistent pressure and ‘silent’ encroachment of low-density urban expansion.