Farmland Abandonment and High Nature Value Farming in Mediterranean Landscapes: Plant Biodiversity Outcomes and Biocultural Trade-Offs

Abstract

Mediterranean rural landscapes are globally important biocultural systems in which long-standing low-intensity farming, grazing, terracing, and agroforestry have historically maintained fine-grained habitat mosaics and high vascular-plant diversity. This review adopts a systematic scoping approach focused on farmland abandonment and High Nature Value (HNV) farming, with a search window extending from 1996 to April 2026. Following PRISMA 2020 and PRISMA-ScR principles, the review maps evidence on plant richness, diversity indices, floristic turnover, indicator species, and vegetation trajectories across Mediterranean agricultural and semi-natural systems. The literature shows that abandonment does not produce a single biodiversity outcome. Early abandonment can maintain or temporarily increase local richness through overlap among remnant grassland species, annuals, and colonizing shrubs, whereas prolonged abandonment more often drives shrub encroachment, woody dominance, and losses of open-habitat specialists. By contrast, HNV systems such as extensive grazing, agro-pastoral mosaics, wood pastures, and dehesa/montado-like agroforestry more consistently maintain habitat interfaces, beta diversity, and plant assemblages associated with cultural landscape continuity. The review argues that biodiversity-sensitive policy in the Mediterranean should move beyond the binary of abandonment versus conservation, and instead support landscape-specific combinations of passive succession, targeted grazing, agroforestry renewal, and rural livelihood viability.

1. Introduction

1.1. Mediterranean Rural Landscapes as Biocultural Systems

The Mediterranean Basin is both a globally recognized biodiversity hotspot and a long-inhabited cultural macro-region in which ecological patterns are inseparable from millennia of land use [1,2,3]. Rural landscapes across the basin are not simply residual “natural” habitats disturbed by human activity; rather, they are historically co-produced mosaics of pastures, open woodlands, terraces, orchards, small arable fields, fallows, scrublands, and field margins that have accumulated ecological value through repeated management, disturbance, and recovery cycles [4,5,6]. In this sense, Mediterranean agricultural landscapes are biocultural systems in which species pools, land-cover heterogeneity, practical knowledge, material heritage, and social meanings have co-evolved over long time spans [3,6,7]. Plant biodiversity is, therefore, embedded not only in soils, topography, and climate, but also in the persistence of extensive grazing, fallowing, pruning, understory clearing, low-input cultivation, and the social institutions that reproduced these practices across generations [8,9,10]. A representative example of this kind of Mediterranean biocultural rural landscape is shown in Figure 1.

1.2. Historical Role of Low-Intensity Farming and Grazing in Maintaining Plant Biodiversity

Low-intensity management has long been recognized as a major driver of biodiversity in European farmland, especially where semi-natural vegetation is maintained by recurrent but moderate disturbance [8,9,11]. In the Mediterranean, such management typically involved livestock mobility, multi-season grazing, rotational or intermittent cultivation, scattered trees, low fertilizer inputs, and fine-grained boundaries between productive and semi-natural land covers [3,12,13]. These practices created a shifting mosaic of microsites and habitat interfaces that supported annuals, geophytes, perennial grasses, ruderal forbs, and disturbance-tolerant shrub species within short distances, often resulting in high floristic richness and strong beta diversity [10,14,15,16]. HNV farming emerged as the policy shorthand for these systems because their biodiversity value depends not on production volume but on the continuity of extensive, place-adapted management and associated structural diversity [17,18,19,20].

1.3. Land-Use Transitions in the Mediterranean Basin

Over the last decades, Mediterranean rural areas have experienced rapid demographic, technological, and institutional restructuring. Marginal mountain farming, transhumant systems, rainfed terraces, and low-output wood-pastures have become increasingly vulnerable to outmigration, labour scarcity, weak profitability, uneven infrastructural integration, and policy regimes that often favour either intensification or administrative simplification over complexity-rich land stewardship [21,22,23,24,25]. One major trajectory is farmland abandonment, which may involve the cessation of cultivation, the interruption of grazing, or both, followed by secondary succession, litter accumulation, shrub encroachment, tree recruitment, and altered fire dynamics [26,27,28,29]. A second, partly opposing trajectory is the persistence, redesign, or targeted policy support of HNV farming and related extensive systems, including dehesas, montados, wood pastures, herb-rich grasslands, and diversified agro-pastoral mosaics [19,20,30,31,32]. The interaction between these trajectories is central to current biodiversity debates because both abandonment and continued use can be ecologically beneficial under some conditions and detrimental under others [33,34,35].

A third trajectory, agricultural intensification, is treated in this review as a cross-cutting comparator because it can conflict with both abandonment-prone extensive systems and the persistence of HNV farming. Intensification includes input increase, mechanization, irrigation, higher stocking pressure, land consolidation, field-margin removal, and the simplification of multi-use mosaics. These processes may reduce plant diversity through habitat homogenization, nutrient enrichment, and loss of semi-natural elements, even where agricultural activity continues [22].

1.4. Ecological Implications for Plant Biodiversity

The ecological effects of abandonment are strongly contingent on time, scale, and prior land-use intensity. Early succession may increase or maintain local richness when remnant grassland species persist while annuals and shrubs establish concurrently, producing transient compositional overlap rather than immediate homogenization [36,37,38,39]. However, many Mediterranean studies show that prolonged cessation of disturbance often reduces the persistence of open-habitat specialists, changes the functional composition of communities, and favours competitive grasses, shrubs, or woody dominants, especially where propagule sources, soil fertility, and fire regimes favour rapid structural closure [40,41,42,43,44,45]. Conversely, HNV systems can maintain high plant diversity through intermediate disturbance, patch turnover, and spatial heterogeneity, but they also depend on socio-economic viability and on grazing or cropping levels that avoid both intensification and neglect [12,13,15,46,47,48]. Thus, the central ecological problem is not whether abandonment is categorically beneficial or harmful, but under which combinations of landscape history, aridity, elevation, connectivity, and management continuity plant biodiversity is retained, transformed, or lost [33,35,49].

1.5. Knowledge Gaps in the Current Literature

Despite a large and diverse literature, four gaps remain especially important for plant-focused Mediterranean synthesis. First, studies of abandonment and studies of HNV farming have often developed in parallel rather than within a common analytical frame, which obscures their shared drivers and competing land-use futures [32,50]. Second, richness metrics are frequently reported without equivalent attention to floristic turnover, specialist decline, or the replacement of characteristic semi-natural communities by widespread colonists [33,44,45]. Third, ecological syntheses still underrepresent the biocultural implications of change—such as the erosion of sense of place, customary management knowledge, and landscape identity—even though these processes directly condition whether biodiversity-supporting practices can persist [5,6,7,51]. Fourth, the literature remains geographically uneven, with the western and northern Mediterranean—especially Iberia and Italy—far better represented than North Africa and parts of the eastern and southern basin [22,32,35].

1.6. Objectives and Review Questions

To address these gaps, this review adopts a systematic scoping approach to synthesize evidence on plant biodiversity outcomes in Mediterranean abandonment and HNV farming systems. The review asks: (i) how farmland abandonment affects plant richness, diversity, composition, and successional trajectories through time; (ii) which HNV systems most consistently support plant diversity and floristic distinctiveness; (iii) which moderators explain divergent biodiversity responses; and (iv) which trade-offs emerge between rewilding-oriented interpretations of abandonment and the maintenance of biocultural landscapes [35,52,53]. The emphasis is on plant biodiversity, but the review also uses policy and social-ecological literature where it clarifies how ecological outcomes are mediated by rural institutions, management incentives, and landscape values [54,55,56].

2. Materials and Methods

2.1. Search Strategy and Selection Criteria

This study followed PRISMA 2020 and PRISMA-ScR guidance for transparent reporting of evidence identification, screening, and inclusion and adopted a scoping review logic to map heterogeneous evidence on plant biodiversity responses to farmland abandonment and High Nature Value (HNV) farming in the Mediterranean Basin [57,58,59,60,61,62,63]. A scoping approach was selected because the evidence base includes observational vegetation surveys, chronosequences, comparative land-use studies, systematic reviews, meta-analyses, and policy-oriented assessments that are methodologically diverse and not directly commensurable [32,64].

This design should be understood as a systematic scoping review rather than as a bibliometric analysis or a quantitative meta-analysis. The purpose of the source counts is therefore descriptive and transparent: they document how the evidence corpus was assembled and structured, but they are not used as effect-size weights or as a statistical sample of Mediterranean land-use change.

The literature search covered the period from 1 January 1996 to 4 April 2026. Searches were structured around three concept blocks: (i) land-use trajectory; (ii) biodiversity response; and (iii) regional context. Representative search strings included combinations of (“farmland abandonment” OR “land abandonment” OR “grazing abandonment”) AND (“plant diversity” OR vegetation OR flora OR “species richness” OR “community composition”) AND (Mediterranean OR dehesa OR montado OR agro-pastoral), as well as (“High Nature Value farming” OR “low-intensity farming” OR agroforestry OR “extensive grazing”) AND (“plant diversity” OR vegetation OR flora) AND Mediterranean. Searches were conducted in Web of Science Core Collection (Clarivate Plc, London, UK), Scopus (Elsevier B.V., Amsterdam, Netherlands), Google Scholar (Google LLC, Mountain View, CA, USA), and CAB Abstracts (CABI, Wallingford, UK), and were complemented by targeted retrieval of European Environment Agency, European Commission, and related HNV policy and technical documents [17,18,19,57,59,62]. Backward citation tracing was used to identify foundational studies, while forward thematic triangulation was used to capture more recent reviews and syntheses.

Records were included if they met the following criteria: (i) they addressed Mediterranean Basin landscapes or Mediterranean-type rural systems in southern Europe and adjacent Mediterranean regions; (ii) they explicitly reported plant biodiversity outcomes or vegetation responses linked to farmland abandonment, grazing abandonment, HNV farming, low-intensity farming, or related extensive agroforestry/agro-pastoral systems; and (iii) they contributed evidence on one or more outcomes of interest, including vascular-plant richness, diversity indices, floristic composition, beta diversity, indicator species, community turnover, vegetation structure, or plant functional composition. Peer-reviewed journal articles formed the core evidence base, complemented by seminal methodological contributions and authoritative policy reports relevant to HNV farmland [17,18,19,20]. Records were excluded if they were duplicates, lacked thematic relevance, did not provide explicit plant- or vegetation-related evidence, or fell outside the ecological and geographic scope of the review.

The initial pool of records was screened in two stages: first by title and abstract, and then by full-text assessment. After removal of duplicates and exclusion of non-relevant records, the final evidence corpus comprised 86 sources, almost all published between 1996 and 2026. The full selection process is summarized in the revised PRISMA flow diagram (Figure 2 and Figure 3). Screening, eligibility assessment, and data charting were conducted by the author.

For clarity, the revised PRISMA flow distinguishes the 75 eligible sources retained through the main screening pathway from the 11 additional eligible sources identified through backward and forward citation tracing and targeted retrieval of policy or methodological documents. These 11 sources were not excluded full texts; they were added to the eligible corpus, giving a final total of 86 sources.

2.2. Thematic Framework and Data Charting

To operationalize the review and structure the synthesis, all eligible full-text records were coded against six predefined thematic aspects aligned with the Results section: (i) general structure of the evidence base; (ii) geographic distribution and thematic concentration; (iii) effects of farmland abandonment on plant biodiversity; (iv) biodiversity outcomes under HNV farming; (v) moderating factors shaping biodiversity responses; and (vi) comparative patterns and biocultural or policy-relevant interpretation across abandonment and HNV trajectories.

The six themes were selected a priori because they operationalize the review’s combined landscape-ecological and social-ecological framing. Themes (i) and (ii) establish the evidential structure and spatial transferability of the corpus; themes (iii) and (iv) correspond to the two principal land-use trajectories examined in the review; theme (v) captures the environmental and management moderators that explain context-dependent biodiversity responses; and theme (vi) integrates plant-biodiversity outcomes with biocultural and policy-relevant trade-offs. This structure follows the premise that Mediterranean plant communities are shaped jointly by disturbance, succession, landscape heterogeneity, management continuity, and rural institutions, rather than by land-cover change alone [4,5,6,20,32,53,56].

Accordingly, the thematic framework was used as an organizing and interpretive device for a heterogeneous evidence base, not as a rigid coding scheme intended to force all studies into equivalent quantitative categories. This approach allowed empirical vegetation studies, reviews, policy reports, and social-ecological papers to be compared transparently while preserving differences in scale, method, and evidential strength.

For each included record, the following information was extracted where available: bibliographic details (year, publication type), country or region, landscape type, land-use history, abandonment or management regime, approximate time since abandonment, grazing continuity, focal taxa or plant functional groups, biodiversity metrics, and environmental moderators such as elevation, aridity, fire regime, and landscape heterogeneity. Particular attention was paid to whether studies interpreted biodiversity in terms of total richness, compositional turnover, persistence of conservation-relevant open-habitat flora, vegetation structure, or functional change.

Because the included evidence was highly heterogeneous in design, scale, and response variables, a structured narrative synthesis was conducted rather than a quantitative meta-analysis. During synthesis, studies were first grouped according to the two principal land-use trajectories examined in this review-farmland abandonment and HNV farming and were then interpreted in relation to temporal dynamics, especially short-term versus long-term post-abandonment responses, as well as to the main ecological moderators identified across the literature. Greater interpretive weight was given to studies with clear site descriptions, explicit management contrasts, and direct plant-diversity measurements [40,44,45]. The distribution of included records by publication type and time period is presented in Figure 4.

For the sake of clarity and to make clear the connection between methods, results and questions, the charting framework was constructed as an analytic matrix, as described in Figure 4. Each question could be easily mapped onto the data fields relevant to the answer to that question and to the sections of the results or discussion where the data were summarized. In turn, this mapping helped ensure that the researchers wouldn’t treat abandonment, HNV farming, moderator or trade off findings as separate topics (Table 1).

Figure 4. Descriptive structure of the evidence corpus. Bars summarize the author’s charting of the 86 included sources by primary thematic focus and publication type. Counts are descriptive and not effect-size weighted; multi-coded outcome variables are summarized in Table 2.

Figure 4. Descriptive structure of the evidence corpus. Bars summarize the author’s charting of the 86 included sources by primary thematic focus and publication type. Counts are descriptive and not effect-size weighted; multi-coded outcome variables are summarized in Table 2.

Table 1. Alignment between the research questions, evidence-charting fields, and synthesis locations used in the revised manuscript.

Research Question	Main Analytical Focus	Data-Charting Fields	Synthesis Location
RQ1	Abandonment effects on plant richness, diversity, composition, and succession.	Land-use history; time since abandonment; grazing cessation; richness; diversity indices; floristic composition; vegetation structure.	Section 3.3; Section 4.1.
RQ2	HNV systems support plant diversity and floristic distinctiveness.	HNV system type; grazing regime; agro-pastoral mosaic structure; agroforestry attributes; indicator plants; habitat specialists.	Section 3.4; Section 4.3 and Section 4.5.
RQ3	Moderators explaining divergent biodiversity responses.	Elevation; aridity; fire regime; landscape heterogeneity; connectivity; prior management intensity; grazing continuity.	Section 3.5.
RQ4	Trade-offs between rewilding-oriented abandonment and biocultural landscape maintenance.	Ecosystem-service co-responses; biocultural values; policy instruments; livelihood viability; intensification risk.	Section 3.6; Section 4.2, Section 4.4 and Section 4.5.

Table 1. Alignment between the research questions, evidence-charting fields, and synthesis locations used in the revised manuscript.

Research Question	Main Analytical Focus	Data-Charting Fields	Synthesis Location
RQ1	Abandonment effects on plant richness, diversity, composition, and succession.	Land-use history; time since abandonment; grazing cessation; richness; diversity indices; floristic composition; vegetation structure.	Section 3.3; Section 4.1.
RQ2	HNV systems support plant diversity and floristic distinctiveness.	HNV system type; grazing regime; agro-pastoral mosaic structure; agroforestry attributes; indicator plants; habitat specialists.	Section 3.4; Section 4.3 and Section 4.5.
RQ3	Moderators explaining divergent biodiversity responses.	Elevation; aridity; fire regime; landscape heterogeneity; connectivity; prior management intensity; grazing continuity.	Section 3.5.
RQ4	Trade-offs between rewilding-oriented abandonment and biocultural landscape maintenance.	Ecosystem-service co-responses; biocultural values; policy instruments; livelihood viability; intensification risk.	Section 3.6; Section 4.2, Section 4.4 and Section 4.5.

Table 2. Descriptive charting of the 86 included sources and the principal variables detected during synthesis.

Charting Category	Coding Basis	Descriptive Charting Result	Interpretation
Primary thematic focus	Single-coded; n = 86	HNV farming/agroforestry/low-intensity systems = 30; farmland abandonment/land-use transition = 27; review methodology/general biodiversity/background evidence = 15; integrated policy/social-ecological/biocultural evidence = 14.	The corpus is balanced between abandonment and HNV farming evidence, with additional sources used for methods, policy interpretation, and biocultural framing.
Publication type or approach	Single-coded; n = 86	Empirical ecological/vegetation studies = 25; reviews/meta-analyses/syntheses = 20; policy, indicator and governance sources = 18; conceptual/biocultural papers = 13; dataset/scenario/modelling or methods papers = 10.	The review integrates direct vegetation evidence with secondary syntheses, policy sources, and social-ecological interpretation.
Biodiversity and vegetation variables detected	Multi-coded	Species richness/diversity indices = 33; floristic composition/turnover/indicator species/habitat specialists = 29; vegetation structure/succession/shrub encroachment/woody recruitment = 31.	These variables separate total richness from compositional change and open-habitat specialist decline.
Management, ecosystem-process and policy variables detected	Multi-coded	Grazing/management intensity/HNV indicators/mosaic structure = 28; soil/carbon/hydrology/fire/erosion/ecosystem services = 24; biocultural values/governance/policy instruments/livelihood viability = 17.	These variables support the assessment of management mechanisms, trade-offs, and policy relevance.

Note: The single-coded categories sum to 86 within each dimension. The variable categories are multi-coded and therefore do not sum to 86. The 86-source corpus, therefore, consists of 75 sources retained through the main screening pathway and 11 additional eligible sources retrieved through citation tracing or targeted policy/methodological searching.

Table 2. Descriptive charting of the 86 included sources and the principal variables detected during synthesis.

Charting Category	Coding Basis	Descriptive Charting Result	Interpretation
Primary thematic focus	Single-coded; n = 86	HNV farming/agroforestry/low-intensity systems = 30; farmland abandonment/land-use transition = 27; review methodology/general biodiversity/background evidence = 15; integrated policy/social-ecological/biocultural evidence = 14.	The corpus is balanced between abandonment and HNV farming evidence, with additional sources used for methods, policy interpretation, and biocultural framing.
Publication type or approach	Single-coded; n = 86	Empirical ecological/vegetation studies = 25; reviews/meta-analyses/syntheses = 20; policy, indicator and governance sources = 18; conceptual/biocultural papers = 13; dataset/scenario/modelling or methods papers = 10.	The review integrates direct vegetation evidence with secondary syntheses, policy sources, and social-ecological interpretation.
Biodiversity and vegetation variables detected	Multi-coded	Species richness/diversity indices = 33; floristic composition/turnover/indicator species/habitat specialists = 29; vegetation structure/succession/shrub encroachment/woody recruitment = 31.	These variables separate total richness from compositional change and open-habitat specialist decline.
Management, ecosystem-process and policy variables detected	Multi-coded	Grazing/management intensity/HNV indicators/mosaic structure = 28; soil/carbon/hydrology/fire/erosion/ecosystem services = 24; biocultural values/governance/policy instruments/livelihood viability = 17.	These variables support the assessment of management mechanisms, trade-offs, and policy relevance.

Note: The single-coded categories sum to 86 within each dimension. The variable categories are multi-coded and therefore do not sum to 86. The 86-source corpus, therefore, consists of 75 sources retained through the main screening pathway and 11 additional eligible sources retrieved through citation tracing or targeted policy/methodological searching.

3. Results

The Results are organized to follow the four research questions. Section 3.1 and Section 3.2 first describe the composition and geography of the evidence base. Section 3.3 addresses RQ1 on abandonment effects; Section 3.4 addresses RQ2 on HNV systems; Section 3.5 addresses RQ3 on moderators; and Section 3.6 addresses RQ4 by comparing abandonment and HNV trajectories.

3.1. General Structure of the Evidence Base

The resulting literature reveals a dense but uneven evidence base. Foundational work from the late 1990s and early 2000s established three enduring propositions: first, that low-intensity farming systems can sustain unusually high biodiversity; second, that abandonment is a major driver of rural landscape change; and third, that dehesas, montados, and other agroforestry or grassland mosaics represent key examples of biodiversity-dependent agricultural systems [8,9,11,12,21,65]. Subsequent literature expanded in two partly disconnected directions. One direction focused on abandonment as a succession process, documenting shifts in vegetation structure, soil properties, erosion, or carbon storage following management cessation [29,36,37,66,67]. The other focused on HNV systems and the policy architecture required to maintain them, particularly through typologies, indicators, and results-oriented support schemes [17,19,30,55,68,69]. Only more recent syntheses have directly juxtaposed these trajectories as competing outcomes of the same social-ecological transition [32,35,45].

To respond to the need for a more transparent description of the evidence pool, the 86 included sources were additionally charted by primary thematic contribution, publication type or analytical approach, and the main variables detected during data extraction. The first two dimensions were single-coded to describe the dominant role of each source in the review, whereas outcome variables were multi-coded because a single study could report richness, composition, grazing regime, soil properties, and policy-relevant information simultaneously. The resulting charting is summarized in Table 2 and Figure 4.

3.2. Geographic Distribution and Thematic Concentration

Geographically, the literature is clustered primarily in Spain, Portugal, Italy, southern France and Greece, with the strongest representation of the Iberian dehesa and montado systems, Mediterranean mountain grasslands and abandoned dryland mosaics [16,37,40,41,70,71]. However, this geographic concentration doesn’t suggest a concentration of land—use change patterns on HNV farmland in these regions, but the locations of some of the most iconic HNV systems, especially the western Mediterranean dehesa and montado and also the uneven capacity for research, data availability, EU policy monitoring and funding, which places most evidence on the northern side of the Mediterranean basin [18,19,20,32,46]. Consequently, the dominance of Iberian and other northern Mediterranean studies, the present summary should be viewed as an effect of an incomplete empirical evidence base than an indicator that the abandonment—HNV dynamic is stronger, more contingent or more ecologically important than on either the southern or eastern side of the Mediterranean basin. For instance, the plant—biodiversity literature available for North Africa and the Levantine countries is less dense in this respect than for European regions, while the few existing basin—scale syntheses draw predominantly on European cases and EU—developed concepts such as HNV farmland [18,19,20,32]. Thematic outcomes most frequently recorded when plants are the target of study include species richness, taxonomic and functional diversity, turnover in plant composition and the retention of grassland specialists, though only a few studies attempt to evaluate restoration thresholds or consider the role of dispersal limitation and long—term plant—community resilience in both abandonment and climate change, which only seems to be addressed with greater emphasis recently [39,49]. Figure 5 offers a schematic representation of the geographic clustering of the Mediterranean evidence base.

3.3. Effects of Farmland Abandonment on Plant Biodiversity

3.3.1. Short-Term Ecological Responses

Early post-abandonment phases often show transitional plant communities in which elements of the former managed flora remain visible while ruderal and shrub species establish. In old-field and dryland studies from Spain and elsewhere in the Mediterranean, this phase can produce stable or temporarily elevated local richness because residual grassland species overlap with early colonists before competitive sorting intensifies [36,37,38]. Similar short-term patterns have been observed where abandonment reduces direct disturbance but does not yet eliminate high-light microsites or seed-bank expression [39,40]. However, these transitional gains are ecologically unstable. Studies that distinguish total richness from community identity show that they often reflect temporary species stacking rather than long-term persistence of conservation-relevant open-habitat assemblages [33,44,72].

3.3.2. Long-Term Succession Trajectories

As abandonment persists, the dominant signal in many Mediterranean systems is directional succession toward denser shrubland or woodland, though the rate and endpoint vary with rainfall, soils, fire, and topographic accessibility [25,28,37]. Chronosequence and vegetation studies repeatedly document increased biomass, reduced soil-surface light, higher litter accumulation, and stronger woody recruitment after cessation of grazing or cultivation [36,39,41,43]. At the same time, broader environmental reviews show that abandonment can reduce some agricultural pressures while enhancing other processes such as fuel continuity, altered runoff generation, or the hydrological consequences of vegetation closure [29,53,73]. The systematic review by Hempel et al. confirms that, across agriculturally marginal grasslands in Europe, abandonment is far more likely than not to reduce plant and lichen biodiversity relative to traditionally managed systems [45].

3.3.3. Loss of Open-Habitat Specialists and Compositional Turnover

The strongest plant-conservation concern is not necessarily a decline in total species counts, but the loss of open-habitat specialists and characteristic semi-natural grassland communities. Grazing cessation tends to reduce the frequency or abundance of short-stature forbs, stress-tolerant annuals, and other light-demanding species, especially where litter accumulation and shrub cover suppress recruitment [40,41,42,44]. Functional studies show that abandonment may retain some richness while shifting communities toward taller, more competitive, or woody species, thereby weakening the distinctiveness of Mediterranean grassland assemblages [39,43]. This distinction between richness and composition is a key finding of both the Mediterranean meta-analysis and later syntheses: abandonment frequently redistributes biodiversity across successional states, but it often does so by replacing culturally maintained open-habitat flora with more widespread or structurally dominant taxa [33,35,45].

3.3.4. Soil and Ecosystem-Process Co-Responses

Although the main focus of this review is plant diversity, the plant evidence is intertwined with changes in soil and ecosystem processes. Several Mediterranean studies report increases in soil organic carbon after abandonment, especially over longer timescales, but these gains do not straightforwardly translate into higher conservation value for plant communities [67,73,74,75]. Likewise, post-abandonment vegetation recovery can produce stronger spatial heterogeneity in some semi-arid settings, yet also lead to denser shrub cover and lower suitability for open-habitat flora [38,76]. From a plant-biodiversity perspective, these co-benefits and trade-offs imply that carbon sequestration, erosion control, and floristic conservation may not be maximized by the same management trajectory [34,35].

3.4. Biodiversity Outcomes Under High Nature Value Farming

3.4.1. Extensive Grazing Systems

The HNV literature converges on the importance of moderate, recurrent grazing for maintaining open structure, preventing rapid shrub encroachment, and generating a shifting template of lightly and more heavily grazed microsites [8,10,20]. In Mediterranean grasslands and wood pastures, such heterogeneity supports coexistence among annuals, perennial herbs, and disturbance-tolerant species while maintaining habitat conditions required by many conservation-relevant plant assemblages [15,40,41,42]. However, the same literature also warns that “extensive” should not be confused with uniformly low pressure; understocking may accelerate abandonment-like trajectories, whereas local overgrazing can simplify vegetation and compromise regeneration [16,71,77]. The biodiversity value of HNV grazing, therefore, depends on management calibration, seasonal timing, and the spatial distribution of herbivore pressure rather than on a simple low/high grazing dichotomy [78,79].

3.4.2. Traditional Agro-Pastoral Mosaics

Traditional mosaics are among the strongest recurring predictors of high floristic diversity in the reviewed literature. Their value stems from landscape juxtaposition: pasture, fallow, field margin, shrub patch, orchard, terrace, and scattered-tree elements create many habitat interfaces within short distances, thereby increasing beta diversity and reducing the risk that a single management state dominates the entire landscape [4,14,15,47]. Reviews and HNV syntheses emphasize that these mosaics do not merely host more species; they retain plant communities associated with historical continuity, semi-natural habitats, and moderate disturbance regimes [20,30,32]. Their main vulnerability lies in socio-economic simplification: when labour-intensive mosaic management becomes unviable, landscapes tend to move toward either abandonment or spatially concentrated intensification, both of which reduce patch diversity [22,51].

3.4.3. Mediterranean Agroforestry Systems

Dehesas and montados remain the clearest emblematic HNV systems in the Mediterranean evidence base. These tree–grass–livestock systems combine open canopies, grazed understories, scattered microclimatic refugia, and heterogeneous nutrient redistribution, making them unusually rich in both ecological and cultural functions [12,13,46]. Plant-diversity studies from agro-silvo-pastoral systems show that mixed land uses and moderate management often create patchy vegetation patterns that sustain more varied herbaceous communities than simplified land covers do [14,15,80]. Yet these systems are also sensitive to tree decline, under-regeneration, overgrazing, and climate stress, which can uncouple their celebrated multifunctionality from actual biodiversity outcomes if management is not adapted [70,71,77,81]. Recent HNV and silvopastoral syntheses therefore treat agroforestry persistence as a question of both ecological management and institutional innovation [32,56,79].

3.4.4. Indicator Plants, Habitat Specialists and Monitoring Implications

The reviewed HNV literature repeatedly points to the diagnostic value of plant indicators. Species associated with nutrient-poor semi-natural grasslands, low-input pastures, and herb-rich wood-pasture understories are particularly informative because they are sensitive to both abandonment and intensification [10,16,31]. Plant indicators also help resolve a central problem in abandonment studies: stable total richness can obscure substantial conservation loss if habitat specialists decline while generalists expand [33,44]. This is one reason why modern HNV assessments increasingly combine structural metrics, farm-management descriptors, and biodiversity-sensitive indicators rather than relying solely on land cover or production intensity [48,68,82,83].

3.5. Moderating Factors

Across the literature, five moderators recur with particular consistency. First, time since abandonment strongly structures outcomes, with early and late stages often showing opposite biodiversity signals [33,36,39]. Second, grazing continuity matters: partial, seasonal, or low-density grazing can maintain open-habitat flora more effectively than complete cessation, especially in grassland- and scrub-prone landscapes [40,41,43]. Third, elevation and topography shape both the likelihood of abandonment and the pace of succession, which helps explain the prominence of mountain systems in the literature [21,24,25]. Fourth, aridity mediates vegetation closure and functional responses; in drier systems woody encroachment may be slower or more spatially uneven, while in subhumid systems the shift away from open habitats can be comparatively rapid [37,38,43]. Fifth, landscape heterogeneity and connectivity influence propagule supply, edge effects, and the retention of multiple habitat states within a single land unit [14,30,47]. Fire regime interacts with all of these moderators by linking biomass accumulation under abandonment to altered disturbance frequency and severity [29,84].

3.6. Comparative Patterns Across Abandonment and HNV Trajectories

Taken together, the comparative evidence suggests a recurring asymmetry. Abandonment can generate short-term floristic overlap, greater woody cover, and sometimes increases in local richness or carbon storage, but it often weakens the persistence of characteristic open-habitat flora and the cultural landscape structures that sustained it [33,35,67,75]. HNV systems, in contrast, more consistently maintain plant assemblages linked to moderate disturbance, patch interfaces, and long-term management continuity, although they remain vulnerable to both underuse and intensification [20,32,46,48]. The key practical implication is that biodiversity assessment should distinguish among total richness, specialist persistence, beta diversity, vegetation structure, and functional composition, because no single land-use trajectory optimizes all of them simultaneously [34,44,45]. A comparative summary of the dominant biodiversity patterns across abandonment and HNV-related trajectories is presented in Figure 6.

4. Discussion

4.1. Reframing Abandonment as a Context-Dependent Biodiversity Process

The Mediterranean evidence does not support a single, universal interpretation of farmland abandonment. Rather, abandonment is a context-dependent biodiversity process whose outcomes depend on prior management, spatial scale, and the conservation dimension being evaluated. Where the objective is structural recovery, carbon accumulation, or the expansion of woody habitats, abandonment may provide ecological opportunities [67,73,75]. Where the objective is the retention of open-habitat specialists or semi-natural grassland communities of high conservation value, the same process may be detrimental, particularly after the transient richness phase has passed [33,41,44,45]. The notion that abandonment is intrinsically restorative is therefore too coarse for Mediterranean rural landscapes, because many of the plant communities of highest conservation concern are historically linked to continued low-intensity use rather than to the cessation of all management [10,53,65].

4.2. Trade-Offs Between Rewilding and Cultural Landscapes

This review also highlights a conceptual tension between rewilding narratives and biocultural conservation. Rewilding frameworks often treat agricultural abandonment as an opening for spontaneous succession, trophic restoration, or reduced human pressure, and these perspectives can be valid in some local contexts [26,52]. Yet Mediterranean rural landscapes challenge any simple equation between reduced management and improved biodiversity because many open habitats, floristic assemblages, terraces, grazed understories, and field-edge communities are themselves historical products of cultural practice [4,5,6,51]. Abandonment can therefore entail not only ecological transformation but also the erosion of a sense of place, pastoral knowledge, and the social capacity needed to maintain biodiversity-supporting land use in the future [7,35]. In practical terms, the trade-off is not between “nature” and “culture” as separate domains, but between different ecological states, each with distinct implications for heritage, species persistence, and rural identity. The main interpretive synthesis of the review is presented conceptually in Figure 7.

4.3. Implications for HNV Policy and Agri-Environment Governance

The policy literature suggests that HNV farming remains one of the most useful frameworks for addressing these trade-offs because it links biodiversity outcomes to actual farming systems rather than to a simple land-cover category [17,18,20]. However, the reviewed studies also show that HNV implementation succeeds only when indicators, payments, advisory support, and local institutional arrangements are aligned with ecological reality [55,68,69]. Broad CAP payments that reward agricultural activity without differentiating biodiversity-relevant practices are unlikely to secure species-rich plant communities in abandonment-prone Mediterranean regions [54]. More promising approaches are those that target extensive grazing, mosaic maintenance, and wood-pasture regeneration, and that recognize the value of local innovation brokers, results-oriented schemes, and farm-scale ecological monitoring [48,56,79,83]. This is especially important in communal and seasonally used grazing systems, where collective action strongly shapes whether abandonment is prevented or merely displaced [85]. Recent HNV reviews and editorials reinforce this conclusion by showing that biodiversity-rich farming systems are shaped by interacting ecological and socio-economic drivers, not by a single indicator or policy lever [32,50,86].

4.4. Agricultural Intensification as a Third Land-Use Trajectory

Although this review focuses on abandonment and HNV farming, agricultural intensification is the third major trajectory against which both processes should be interpreted. Intensification includes mechanization, irrigation or fertilizer inputs, shrub and field-margin removal, simplified rotations, larger field units, higher stocking densities, and feed-based livestock production. It can erode many of the same biodiversity attributes that are lost through long-term abandonment, but by a different mechanism. Whereas abandonment tends to reduce open-habitat specialists through vegetation closure and litter accumulation, intensification more often reduces them through disturbance homogenization, nutrient enrichment, herbicide pressure, soil compaction, and the removal of semi-natural elements [44,54].

The Mediterranean region’s planning should therefore avoid assuming that ongoing cultivation is intrinsically good. High—value nature farming occupies a marginal space between underuse and overuse, retaining plant species not solely from continued farming itself, but due to continuing traditional low—input management methods, maintenance of structural and topographic variety and conservation of seminatural elements. Such intensity can, however, move abandonment elsewhere, forcing production and more easily accessible terrain to grow in input use while simultaneously leading to the desertion of terraced fields, upland pastures and dry Farming land. This polarization reduces landscape diversity in species between regions (beta diversity) while also eroding the social foundations that preserve extensive grazing, agroforest revival and the fragmented management of natural land [20,22,32,69]. Accordingly, the response to such a problem would entail differentiating three trends of the landscape: passive Succession where possible, intensive assistance to nature farming systems and restrictions against such intensification efforts that improve farmability at the cost of habitat diversity and an increased monoculture of plant species.

4.5. Implications for Biodiversity Conservation and Rural Planning

For conservation practice, the main lesson is that landscape-scale heterogeneity should be treated as the management objective, not simply an incidental by-product. Mediterranean plant diversity is frequently highest where open habitats, semi-open woodlands, shrub edges, and selectively managed transitional stages co-occur in functional proportions [13,14,47]. This does not imply freezing all landscapes in a historical state. Rather, it means that passive succession, targeted grazing, rotational clearance, wood-pasture regeneration, and selective restoration should be assembled in landscape-specific combinations that reflect ecological thresholds and social feasibility [28,78,84]. Where abandonment is unavoidable, intermittent management may be sufficient to prevent the rapid disappearance of open-habitat flora. Where HNV systems persist, conserving management continuity may yield greater plant-biodiversity benefits than attempting post hoc restoration after ecological and social simplification has already occurred [16,45,81].

4.6. Research Gaps and Future Directions

Several priorities emerge for future research. First, long-term plant-community datasets are still too scarce to disentangle temporary richness pulses from durable compositional change, particularly across different abandonment durations and grazing legacies [39,45]. Second, future syntheses should move beyond richness to integrate functional diversity, demographic processes, regeneration bottlenecks, and the persistence of narrow-range or culturally significant plant assemblages [42,43,49]. Third, underrepresented regions of the eastern and southern Mediterranean require much stronger empirical coverage if basin-wide conclusions are to avoid a western Mediterranean bias [32,35]. Fourth, climate change must be treated not as a background variable but as an interacting driver that can alter the biodiversity consequences of both abandonment and HNV persistence through drought stress, pasture-quality changes, and modified regeneration dynamics [49,77]. Finally, review methodology itself can be improved by combining ecological field studies with systematic social-ecological evidence on labour, institutions, and rural livelihood viability, because plant diversity in Mediterranean cultural landscapes ultimately depends on whether biodiversity-supporting management remains socially reproducible [56,62,63].

Further methodological limitations stem from the chosen corpus combining various forms of literature, such as empirical vegetation science literature (peer—reviewed papers) and meta—analyses and reviews. The rationale of this paper will clarify how various studies were selected and interpreted, but it’s worth mentioning from the outset that the combination represents a structured scoping review of evidence. It shouldn’t be taken as evidence on publication frequency nor on how the actual evidence on the effect of fire exclusion may stack up quantitatively.

5. Conclusions

Farmland abandonment and HNV farming are not simple opposites, but alternative trajectories within a shared Mediterranean social-ecological transition. The reviewed evidence shows that abandonment can deliver short-term gains in structural complexity, temporary richness, and some regulating services, yet often at the cost of open-habitat specialists and biocultural landscape continuity when management cessation persists. HNV systems more consistently sustain heterogeneous habitats, plant-community distinctiveness, and the ecological conditions required by many species-rich Mediterranean grasslands and agroforestry landscapes. Agricultural intensification should be recognized as a third trajectory in this debate because it can undermine HNV systems through habitat homogenization, nutrient enrichment, field-margin removal, and the simplification of low-intensity mosaics. Biodiversity-sensitive policy should therefore avoid both blanket opposition to abandonment and uncritical celebration of spontaneous succession, while also avoiding support for intensification that would erode HNVs. Instead, it should identify where passive recovery is ecologically acceptable, where continued low-intensity management, especially extensive grazing and mosaic agroforestry, remains essential for conserving Mediterranean plant diversity, and where safeguards are needed to prevent the conversion of biodiversity-rich farming systems into simplified production landscapes. These conclusions should nevertheless be interpreted with some caution, as the available evidence remains geographically uneven and is still weighted toward the western and northern Mediterranean, with comparatively limited plant-focused evidence from Mediterranean North Africa and parts of the eastern Mediterranean.