Next Article in Journal
Influence of Eucalyptus Plantation on Soil Microbial Characteristics in Severely Degraded Land of Leizhou Peninsula
Previous Article in Journal
Enhancing Cold Tolerance Evaluation in Camellia sinensis and Camellia japonica Through Multimethod Analysis and Predictive Modeling
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 16
Issue 10
10.3390/f16101600
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Forecasting the Impact of Climate Change on Tetraclinis articulata Distribution in the Mediterranean Using MaxEnt and GIS-Based Analysis

by
Kaouther Mechergui
1,†,
Umer Hayat
2,*,†,
Muhammad Hammad Ahmad
3,†,
Somayah Moshrif Alamri
4,
Eman Rafi Alamery
4,
Khadeijah Yahya Faqeih
4,
Maha Abdullah Aldubehi
4 and
Wahbi Jaouadi
1,5,*
1
Laboratory of Forest Ecology, National Research Institute of Rural Engineering, Water, and Forestry, University of Carthage, Ariana 2080, Tunisia
2
State Key Laboratory of Integrated Management of Pest Insects and Rodents, Institute of Zoology, Chinese Academy of Sciences, Beijing 100864, China
3
GIS Analyst at National Engineering Services Pakistan, Lahore 54770, Pakistan
4
Department of Geography and Environmental Sustainability, College of Humanities and Social Sciences, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh 11671, Saudi Arabia
5
Laboratory of Sylvo-Pastoral Resources, The Silvo-Pastoral Institute of Tabarka, University of Jendouba, Tabarka 8100, Tunisia
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Forests 2025, 16(10), 1600; https://doi.org/10.3390/f16101600
Submission received: 19 September 2025 / Revised: 14 October 2025 / Accepted: 16 October 2025 / Published: 18 October 2025
(This article belongs to the Special Issue Climate Change Impacts on Forest Dynamics: Use of Modern Technology)
Browse Figures
Review Reports Versions Notes

Abstract

Climate change threatens Tetraclinis articulata, a Mediterranean plant endangered by habitat loss, logging, and aridification. This study used the MaxEnt model to analyze factors affecting its distribution under current and future climate scenarios (SSP1-2.6 to SSP5-8.5) for 2040–2100, highlighting its vulnerability to drought and urgent conservation needs. Results showed that: (a) the model demonstrated excellent predictive power with an AUC of 0.92; (b) the highly suitable habitat for T. articulata is projected to expand by 6.5%–6.7% (5.24–5.38 million km2) by 2100 under SSPs 2-4.5, 3-7.0, and 5-8.5, compared to current conditions (6.1%, 4.92 million km2); (c) the centroid of suitable habitats shifts from northwest Algeria (1.394° N, 33.538° E) to various locations under future climate scenarios: west Morocco (SSP1-2.6, −3.429° S, 33.588° E), east Tunisia (SSP2-4.5, 11.091° N, 32.501° E), northwest Morocco (SSP3-7.0, −1.947° S, 34.098° E), and southwest Morocco (SSP5-8.5, −2.985° S, 34.707° E); (d) key environmental variables influencing T. articulata distribution include annual precipitation (bio12, 41.7%), mean annual temperature (bio1, 27.9%), and precipitation during the driest month (bio14, 16.1%). This study concluded that climate change significantly influenced the distribution of T. articulata in the Mediterranean, highlighting the urgent need for conservation strategies to mitigate the risk of local extinction driven by both anthropogenic activities and climate impacts.

1. Introduction

Climate change is a significant factor profoundly affecting biodiversity at all levels [1,2,3]. Changes in plant distribution represent a fundamental reaction to novel climatic conditions [4,5], as has been observed in prior eras of climate change. Pollen analysis findings suggest altitudinal and latitudinal shifts in species distribution over centuries and decades [6].
Tetraclinis articulata Vahl. Mast., commonly known as the thuya/sandarac gum tree, is a type of evergreen conifer belonging to the Cupressaceae family. It can grow to a height of 6–12 m and is characterized by its flattened branches, scale-like leaves, and reddish-brown aromatic trunk [7]. This multipurpose tree is renowned for its premium root burlwood, which renders it extremely desirable in the handicraft sector [8]. The essential oil derived from T. articulata has antibacterial efficacy against Staphylococcus aureus and Micrococcus luteus [9]. This monospecific species predominantly has a North African distribution, with the sole European occurrence of T. articulata, aside from Malta and Cyprus, in southeastern Spain, particularly in the coastal areas of the Murcia Province [10]. The inherent characteristics and restricted geographic range of the species render it particularly significant for biodiversity conservation [11]. This tree is pivotal in the socioeconomic framework of North Africa, as it offers pastureland for animals and generates items for local consumption [12]. It is ideal for the soil conservation and restoration of forests, as it can adhere to rocks on the steepest inclines owing to its robust and flexible root structure [11,13]. T. articulata forests offer significant benefits, including soil erosion mitigation, biodiversity preservation, and carbon dioxide sequestration, making them appropriate for afforestation initiatives in dry or semiarid regions where few species can thrive, and in places with poorly degraded soils [14].
T. articulata is a Mediterranean conifer that can survive long periods of dry weather and has many uses, including timber, resin, and essential oils, also important for controlling erosion. In a fragile ecological environment, understanding the potential spatial distribution of species and the underlying environmental conditions is crucial for resolving ecosystem management issues and predicting impacts of global change [15]. Its distribution is fragmented and relict, which means that it is only found in a few places around the world, such as its small native range in SE Spain and Malta [7,16]. The Mediterranean is a well-known area where climate change is happening quickly; it is warming about 20% faster than the rest of the world. This makes it more likely that drought-sensitive forest mosaics, where T. articulata grows, would be affected (MedECC/UNEP, 2020–2024) [17]. Previous studies on T. articulata’s predicted range responses primarily focused on the Iberian Peninsula within older climate frameworks, emphasizing its susceptibility to warming and aridification, but failing to address basin-wide, CMIP6-based evaluations [14,18].
T. articulata forests present a particularly intriguing scenario for evaluating the anticipated impact of climate change on biodiversity. A common method for evaluating the possible effects of climate change on biodiversity involves employing species distribution models (SDM) and climate change scenarios to project future habitat suitability for various species [2,19,20]. SDM predicts environmental suitability and species distribution by integrating species distribution data with environmental variables such as climate and habitat [20]. Among the prominent SDM’s, CLIMAX, Biomod2, GARP, and MaxEnt, which are based on the maximum entropy model, are recognized for their accuracy, operational robustness, and efficiency [21]. It has an enhanced predictive capability compared with other models, particularly in its robustness against sample bias [20]. Most studies are based on the Coupled Model Intercomparison Project Phase 5 (CMIP5), whereas the newer version, CMIP6, demonstrates heightened climate sensitivity and enables a more comprehensive analysis of potential scenarios [22]. Thus, CMIP6 provides an improved study of the environmental factors influencing species distribution and enables a more accurate prediction of invasive alien pest spread.
Presence-only species distribution models (SDMs) like MaxEnt are well-tested for predicting climate suitability and finding the main factors that affect distributions in Mediterranean forests [23,24]. Recent applications in North Africa validate MaxEnt’s efficacy for tree taxa and advocate for its expansion to incorporate future climate scenarios and larger ensembles to mitigate bias [15]. Our study employs MaxEnt within a GIS framework integrated with CMIP6 SSPs (SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5) to provide the inaugural Mediterranean-wide, multi-horizon projections for T. articulata, thereby addressing the geographic and scenario deficiencies identified in previous CMIP5-era or sub-regional analyses.
T. articulata has experienced considerable fragmentation and a reduction in its population [25]. This resource is endangered by fragmentation resulting from human activities, extensive logging, and government negligence, which may result in permanent losses [26]. The effects of escalating aridification and heightened drought frequency on species tolerance and legacy responses to climate change remain unclear. Additionally, a study of its distribution pattern under different climate change scenarios has not yet been conducted. For better management of T. articulata forests in these semi-arid territories, knowledge of this species is essential. This information is essential for future conservation and management plans [27]. We hypothesized that, like many other species of plants and insects, climate change would have a significant influence on the distribution pattern of T. articulata under different climate change scenarios. This study aimed to identify the main climatic factors influencing the distribution of T. articulata under current and future climate change scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5) for 2040, 2060, 2080, and 2100, respectively.

2. Materials and Methods

2.1. Study Area

In the Mediterranean region, T. articulata is primarily located in countries such as Tunisia, where it serves as one of the main habitats for this species, especially in forests and mountainous areas. It is also found in various regions of Algeria, particularly in mountainous and arid regions. In Morocco, this species thrives in the Atlas Mountains and other elevated terrains. Additionally, T. articulata has been reported in other Mediterranean areas, including Spain, and islands, such as Malta (Figure 1). It is indigenous to the mountain ranges of northern Africa (Morocco, Algeria, and Tunisia, up to 1800 m a.s.l.), with scattered populations mostly found in eastern Spain (near Cartagena, the Murcia region) and Malta [28].

2.2. Species Occurrence Data

For occurrence data, we compiled all relevant published works from the library covering the period January 1952 to 2024. Occurrence record data were gathered from updated databases [27,30,31,32,33]. Additional data was obtained from the Global Biodiversity Information Facility (https://www.gbif.org/, accessed on 10 May 2025), EPPO (https://gd.eppo.int/, accessed on 10 May 2025), CABI (https://www.cabi.org/, accessed on 10 May 2025), and Euro + Med PlantBase (https://europlusmed.org/, accessed on 10 May 2025) databases. A total of 23,564 occurrence points were collected that were further refined to 15,000 points by applying spatial thinning of keeping one point in a 1 km radius in ArcMap (version 10.8.1).

2.3. Climatic Data

Nineteen WorldClim bioclimatic variables (Bio1–Bio19), averaged from 1950 to 2000 and possessing a spatial resolution of 2.5 arc minutes, were acquired from the WorldClim Database (http://www.worldclim.org, accessed on 10 May 2025) for MaxEnt modelling for the current scenario. To thoroughly assess the alterations in the probable acceptable habitats of T. articulata in the future, especially in the years 2040, 2060, 2080, and 2100, we utilized the global climate model (GCM) MIROC6. We used the GCM from the Sixth Assessment Report (AR6) of the IPCC. Four shared socioeconomic pathways (SSPs) were selected for each year: SSP1-2.6 [Sustainability, low forcing: Globally cooperative, equity-focused development with rapid clean-tech adoption, efficient resource use, and strong land protection. Emissions peak early and decline fast, stabilizing at ~2.6 W/m2 by 2100]; SSP2-4.5 [Middle of the Road, intermediate forcing: Continuation of current trends—moderate growth, uneven progress, mixed policies. Emissions plateau mid-century then slowly decline, landing near ~4.5 W/m2 by 2100]; SSP3-7.0 [Regional Rivalry, high forcing: Fragmented world with weak international cooperation, slower economic growth in many regions, high population in developing areas, low tech transfer, limited mitigation. Emissions keep rising, reaching ~7.0 W/m2 by 2100]; SSP5-8.5 [Fossil-Fueled Development, very high forcing: Rapid, energy-intensive growth driven by abundant fossil fuel use and material consumption; strong tech progress but little mitigation. Emissions surge to ~8.5 W/m2 by 2100]. The reason to incorporate MIROC6 because it is an updated version of the CMIP6 GCM that has been shown to work better than MIROC5 in terms of the mean state and internal variability. This is because it has new processes for aerosols and clouds, land surfaces, and the stratosphere [34]. MIROC6 is often used in Mediterranean temperature and extreme forecasts and demonstrates satisfactory performance when integrated with conventional bias-adjustment and ensemble techniques [35].

2.4. Species Distribution Model (SDM) and Variable Selection

MaxEnt v3.4.4 was utilized to forecast the distribution range of T. articulata across different climatic scenarios. MaxEnt operates by identifying correlations between recorded occurrence points and background data within defined parameters, as outlined by [36]. MaxEnt is distinguished for its ability to integrate several factors such as meteorological data, provide different metrics for assessing model performance, and provide user-friendly software functionality, making it extensively utilized in species distribution prediction [37]. A complete array of 19 bioclimatic variables was obtained from the WorldClim website and first utilized as a prospective predictor. Multicollinearity among bioclimatic factors may adversely affect the relationship between species distributions [20]. To create a high-performance model with reduced variables, the Pearson correlation coefficients of the cross-correlations among the 19 bioclimatic variables were computed using the SDM toolbox (Version 2.6) extension in Arc Map (version 10.8.1) to eliminate highly correlated variables based on Pearson correlation at p ≤ 0.8 [20]. Six variables (p ≤ 0.8) were selected for the investigation based on Pearson correlation values (Table 1). To test the influence of variable selection, we constructed the two models, one with all 19 bio variables and the other one with the selected 6 bio variables, and to determine the statistical difference between them, we conducted the ANOVA and t-test. We quantified the contribution of the 19 predictors using MaxEnt’s built-in jackknife procedure on the same dataset used for model fitting (presence records + background), with a 25% random test split and all other settings at defaults. The jackknife runs three models per predictor: (i) with only that variable, (ii) with all variables, and (iii) with that variable omitted. We examined changes in regularized training gain (and inspected AUCs) to interpret importance: a variable yielding the highest gain when used alone contains the most useful stand-alone information; a variable whose omission causes the largest drop in gain contributes the most unique information not captured by others. We report these results alongside percent contribution/permutation importance and use them to identify the dominant climatic drivers and to guide interpretation of species–environment relationships.

2.5. Model Accuracy Assessment and Map Generation

The final model was run using a 10-fold bootstrapping crossover technique for current and future conditions. The efficacy of the MaxEnt model was assessed using the Receiver Operating Characteristic (ROC) curve, with the area under the curve (AUC) computed as an indicator of predictive accuracy. The area under the receiver operating characteristic (ROC) curve (AUC) was adopted for model evaluation [38]. AUC values over 0.5 indicate a superior match compared to random chance, whereas values between 0.9 and 1 denote exceptional predicting capability. To enhance species distribution modeling with MaxEnt, we evaluated eight Feature Class (FC) auto features—auto class, L: Linear, Q: Quadratic, LQ: Linear + Quadratic, LQH: Linear + Quadratic + Hinge, LQT: Linear + Quadratic + Threshold, LQHP: Linear + Quadratic + Hinge + Product, LQHPT: Linear + Quadratic + Hinge + Product + Threshold—alongside Regularization Multipliers (RM) varying from 0.2 to 2. The AUC score was used to find the most effective feature class and RM pair by assessing the degree to which the model executed [38,39].
The best model choice was centered on getting the highest AUC while still being able to generalize. After the modeling process was finished, we used MaxEnt’s results to make habitat suitability maps for T. articulata for both the present and the future. The maximum training sensitivity plus specificity logistic threshold (MTSS) was used to figure out how good a habitat was for species. The models’ MTSS values came out to be 0.10.
By executing lowest presence threshold (LQT) feature class to delineate the appropriate and unsuitable locations [20], we categorized habitat into four categories: unsuitable (MTSS < 0.1), moderately suitable (MTSS > 0.1–<0.3), suitable (MTSS > 0.3–<0.6), and highly suitable (MTSS > 0.6). All layers were produced in ArcMap (version 10.8.1) with ArcToolbox under Spatial Analyst Tools, namely, the Reclassify function, according to the LQT criteria.

3. Results

3.1. Model Accuracy and Contributions of Predictor Variable

To mitigate overfitting, we performed a series of screenings on the distribution sites of T. articulata. The efficacy of several FC combinations under varied RMs was evaluated for T. articulata. AUC ratings reached a maximum of RM = 1 for all evaluated models, with LQT demonstrating the best predictive accuracy (Figure 2a). The ideal configuration was LQT at RM = 1, with an AUC of 0.92. The findings indicate that LQT is the most efficacious feature class for simulating T. areticulata distributions across varying environmental conditions. Furthermore, the results demonstrated that the model executed with the selection of shortlisted variables by correlation analysis yielded more significant (df = 21; F = 10.29; p < 0.01 for AUC; df = 21; F = 9; p < 0.01 for TSS) and precise outcomes, achieving a higher AUC and TSS of 0.92 and 0.83 in contrast to the model utilizing all variables (AUC = 0.87 and TSS = 0.77) (Figure 2b).
The AUC values obtained from the 10-fold cross-validation of the final model with selected variables are illustrated in Figure 3. The AUC value was 0.92, indicating the robust efficacy of the MaxEnt model in forecasting the possible distribution zones of T. articulata. MaxEnt prediction results revealed that all occurrence records were within the anticipated range, suggesting that the model’s predictions were highly accurate.
The primary environmental factors correlated with the spread of T. articulata were rated as follows: seasonality of annual precipitation, annual mean temperature, and precipitation during the driest month. However, the overall variables contributed an average of 41.7%, 27.9%, 16.1%, 7.7%, 4.5%, and 2.1% to the model (Figure 4).
The jackknife tests of variable relevance further validated that the annual precipitation seasonality (bio12), mean annual temperature (bio1), and precipitation during the driest month (bio14) had superior predictive abilities compared with the others. Their elevated training, test gain, and AUC values clearly demonstrate their substantial impact on the predictive accuracy of the model (Figure 5 and Figure 6). The presence of T. articulata showed distinct trends based on mean annual temperature (bio1). It abruptly increases from 10 to 16 °C, peaks at 17 °C, and drops rapidly beyond 17.5 °C. Based on the mean diurnal range (bio2), the presence of T. articulata increased at 6 °C, peaked at 10 °C, and started declining at 13 °C. Furthermore, regions exhibiting moderate year-round temperature stability, where seasonal fluctuations substantially influence yearly temperature variance, have an increased probability of T. articulata occurrence (bio4). Areas where the mean temperature of the wettest quarter ranged between 5 and 13 °C exhibited a high presence of T. articulata (bio8). Similarly, regions characterized by significant differences between precipitation rates or comparatively high seasonal fluctuations in precipitation rates displayed increased probabilities of T. articulata presence (bio12). Areas where the precipitation rate of the driest month ranged between 0 and 10 mm displayed increased probabilities of T. articulata presence (bio14).

3.2. Current Suitable Habitats

Under present climatic circumstances, 22.02% (17.68 million km2) of the total land area of the Mediterranean areas under investigation might provide habitats conducive to the survival of T. articulata (Figure 7 and Figure 8). Suitable conditions are predominantly found in the coastal regions, especially in Morocco, Algeria, and Tunisia. These locations display highly suitable environments owing to their Mediterranean climate, which is characterized by moderate precipitation and appealing temperatures. Coastal regions of these nations offer essential environmental conditions, including well-drained soil and regular humidity, which are vital for tree growth. In Algeria, the northern coastal zone has wide areas of high and moderate suitability, whereas the middle and southern parts are deemed unsuitable because of the dry climate and insufficient precipitation. Tunisia’s coastline zones also demonstrate favorable circumstances, with substantial sections rated as extremely appropriate, in contrast to the interior parts that are primarily unsuited. The Canary Islands, Cyprus, and Malta exhibit unique favorability patterns. The coastal areas of the Canary Islands are exceptionally conducive, indicating the island’s consistent temperate environment. Cyprus has a significant prevalence in both suitable and extremely suitable regions, particularly in the central and southern sectors of the island. Despite its small size, Malta presents ideal conditions throughout, establishing it as a reliable habitat for T. articulata. Marginal areas were dispersed in the peripheries of suitable zones, signifying transitional regions in which climatic or soil conditions may restrict ideal growth. This distribution highlights the significance of coastal Mediterranean conditions in supporting T. articulata, whereas inland desert regions are predominantly unsuitable for their growth.

3.3. Future Suitable Habitat Under Climate Change Scenarios

Habitat suitability under future climate change scenarios (SSP 1-2.6, SSP 2-4.5, SSP 3-7.0, and SSP 5-8.5) for the years 2040, 2060, 2080, and 2100 have been forecasted using MaxEnt (Figure 9). Each scenario forecasts possible alterations in habitat suitability predicted by climate change projections encompassing Northern Africa and the Mediterranean region. Throughout all the SSPs and timeframes, suitable and highly suitable locations were predominantly located in the coastal and northern regions. A distinct trend of habitat growth or shrinkage was observed based on the SSP. The suitable and highly suitable regions exhibited relative stability until 2100 in SSP 1-2.6, indicating few changes. Moderate emission scenarios indicated an expansion of highly suitable regions in SSP 2-4.5 and SSP 3-7.0, especially in the coastal region, accompanied by significant changes in the distribution pattern over time. High-emission scenarios demonstrated considerable growth in highly appropriate regions in SSP 5-8.5, indicating notable environmental advantages for the species, especially in the long-term (2100). In Algeria, the northern coastal regions exhibited high suitability across all SSPs and periods. Under SSP 3-7.0 and SSP 5-8.5, there is a noticeable expansion of highly suitable inland areas by 2080 and 2100, respectively. The marginal areas in the central region remained largely unchanged, but coastal zones remained consistent hotspots for potential infestations. In Tunisia, the northern and eastern coastal areas show high concentrations of suitable and highly suitable habitats. Under all SSPs, the extent of the highly suitable zones increased over time, particularly under SSP 5-8.5. The inland regions remained largely marginal, but coastal areas were consistently vulnerable. In Morocco, coastal regions, particularly in the north, are highly suitable. Habitat suitability is projected to expand southward under SSP 3-7.0 and SSP 5-8.5. By 2100, significant inland shifts in suitable habitats were noted, indicating an increased risk over a larger area. In Libya, the northern coastal areas exhibit marginal to suitable habitats, with highly suitable areas limited. Highly suitable areas are projected to increase slightly along the coast, particularly under the SSP 5-8.5. However, much of the country’s interior remains unsuitable. In Cyprus, the entire island showed significant suitability for habitat, with large portions categorized as highly suitable. Across all SSPs, Cyprus remained a consistent hotspot for infestation with no significant changes observed. High suitability was sustained for 2100, indicating persistent risk. In Malta, the island was highly suitable for the target species. Consistently, high suitability was maintained in all SSPs by 2100, suggesting an ongoing risk. In Spain and the Canary Islands, some islands show marginal suitability, whereas others have patches of high suitability. The marginal areas remained unchanged, but highly suitable zones persisted in specific locations, especially under higher emission scenarios. The highly suitable area coverage percentages under future climatic conditions for various scenarios, in comparison to the existing coverage (6.12%), are as follows: SSP1–2.6: 6.16% (2060) > 6.15% (2080) > 6.10% (2100) > 5.97% (2040); SSP2–4.5: 6.58%(2100) > 6.24%(2080) > 6.06%(2060) > 5.92%(2040); SSP3–7.0: 6.53%(2100) > 6.40%(2060) > 6.31%(2080) > 5.92%(2040); and SSP5–8.5: 6.70%(2100) > 6.63%(2080) > 5.97%(2040) > 5.95%(2060).

3.4. Area-Wise Comparison Between Current and Future Projections

3.4.1. Algeria

Currently, the optimal areas are primarily restricted to the northern coastal zone, whereas the central and southern regions are classified as marginal or unsuitable. However, future predictions imply a progressive inland expansion of highly appropriate zones, especially under the higher-emission scenarios (SSP 3-7.0 and SSP 5-8.5). By 2080 and 2100, the range of these highly suitable habitats will expand, whereas marginal coastal zones will diminish and convert to suitable or highly suitable regions.

3.4.2. Tunisia

The northern and eastern coastal regions of Tunisia currently display a significant concentration of suitable and highly appropriate habitats; however, the middle and southern areas are predominantly unsuitable. Future projections, particularly under SSP 5-8.5, suggest a significant increase in the highly suitable areas along the coastline.

3.4.3. Morocco

Currently, Morocco’s coastal and northern areas are deemed highly suitable with little inland extension. The central and southern regions of the nation are unsuitable. Future predictions suggest the progressive inland expansion of both suitable and highly suitable zones, particularly under high-emission scenarios. By 2100, the proliferation of highly suitable habitats will become increasingly evident, suggesting more available land for possible woodlands.

3.4.4. Libya

The current scenario in Libya indicates that suitable habitats are restricted to limited places along the northern coastline, whereas most inland regions are regarded as unsuitable. Future projections suggest a marginal increase in suitable areas near the coastline, particularly under elevated emissions scenarios. Notwithstanding this, inland areas remain predominantly inadequate, with a minimal increase in highly suitable regions.

3.4.5. Cyprus

Cyprus has a distinctive situation wherein the entire island is presently categorized as very appropriate, signifying considerable infestation danger. Future forecasts indicate that this high suitability remains consistent across all SSPs and timeframes, with no significant alterations in the geographical distribution. This consistency indicates that most Cyprus areas are extremely conducive to the establishment and development of T. articulata.

3.4.6. Malta

Like Cyprus, Malta is primarily categorized as highly suitable, with only slight variance in suitability levels. Future possibilities suggest that the island’s high adaptability remains stable throughout all trajectories, highlighting its continual suitability for the expansion and development of T. articulata.

3.4.7. Spain and Canary Islands

The present situation in Spain and the Canary Islands indicates that several islands possess highly suitable environments that exhibit regional differences. Future forecasts suggest that these highly suited regions will either remain stable or show moderate expansion, particularly under elevated emissions scenarios. This indicates positive conditions of the Canary Islands and southwestern Spain for the development and expansion of T. articulata.
The area changes in the T. articulata suitability classes (Unsuitable, Marginal, Suitable, Highly suitable) under four climate change scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5) from the current period to 2100 are presented in Table 2. These data revealed substantial shifts in habitat suitability with increasing emissions. Under low-emission scenarios (SSP1-2.6), marginal declines were observed in highly suitable and suitable areas, while unsuitable areas expanded moderately. However, under high-emission scenarios (SSP3-7.0, SSP5-8.5), unsuitable areas increased significantly, particularly by 2100, with SSP5-8.5 showing a dramatic rise of 1.34 million km2. The marginal and suitable areas declined steadily under higher emissions, indicating severe habitat degradation. Highly suitable areas exhibited mixed trends, with minor increases under SSP3-7.0, but significant reductions under SSP5-8.5. These projections emphasize the potential loss of optimal habitats for T. articulata, with higher emissions exacerbating the shift toward unsuitable conditions, posing a critical threat to the species’ future distribution.

3.5. Tetraclinis Articulata Centroid Shift

The projected centroid shifts in suitable T. articulata habitats under different climate change scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5) by 2100 are presented in Table 3 and Figure 10. The current centroid is located at 1.394° N and 33.538° E, with notable shifts observed across all future scenarios. Under SSP1-2.6, the centroid shifted westward to −3.429° S, 33.588° E, with 536.32 km. Conversely, SSP2-4.5 predicted an eastward shift to 11.091° N, 32.501° E, spanning the longest distance of 1084.32 km. Under SSP3-7.0 and SSP5-8.5, the centroids shifted northwest, with coordinates of −1.947° S, 34.098° E (376.68 km) and −2.985° S, 34.707° E (503.96 km), respectively.
These results highlight the potential redistribution of suitable habitats, driven by climate change. The eastward shift under SSP2-4.5 suggests a latitudinal expansion into northern areas, whereas the westward and northwestward shifts under SSP1-2.6, SSP3-7.0, and SSP5-8.5, indicate potential habitat contraction or displacement. The varying directions and distances emphasized the complex responses of T. articulata to climatic pressures, with high-emission scenarios leading to significant habitat instability. These findings underscore the urgent need for climate mitigation and conservation efforts to preserve suitable habitats of this species in the face of accelerating environmental changes.

4. Discussion

This study provides the first predicted potential habitat distribution map for a plant species (T. articulata) in the Mediterranean region. We conducted an initial investigation on the likely distribution range of T. articulata in Mediterranean regions under climate change scenarios using MaxEnt software, literature evidence, data sourced from GBIF, and an extensive field survey in these areas. The performance of the current MaxEnt model was validated to be satisfactory through pertinent screening of available occurrence records, optimization of bioclimatic variables, and adjustment of model parameters. The ROC curve, TSS, and omission rates (Figure 3) support these findings [20]. The results demonstrated a robust correlation between the performance of the MaxEnt model and modeled distributions of T. articulata. The predicted effectiveness of a model is influenced by species features, including geographical distribution and habitat location, as well as the nature of the model [40].
Suitable habitats for T. articulata are predominantly found across all countries and islands in the Mediterranean region, with the most extensive coverage in Morocco (−1.713° S, 34.128° E to −10.476° S, 28.472° E), Spain and the Canary Islands (−3.048° S, 40.091° E to −5.61° S, 36.021° E), Algeria (−1.631° S, 33.619° E to 8.266° N, 36.093° E), Tunisia (10.133° N, 37.211° E to 9.235° N, 33.934° E), Malta (14.337° N, 35.977° E to 14.542° N, 35.813° E), and Cyprus (34.559° N, 35.659° E to 34.399° N, 34.782° E) (Figure 7). Under future climate change scenarios, the current pattern of suitable distribution shifts slightly toward the Northern Hemisphere, with an increase of 0.46 million km 2 in the highly suitable area in SSP5-8.5, by the end of the century, compared to all other scenarios (Table 2, Figure 8). Climate change profoundly affects the distribution of T. articulata, as outlined by Monzón et al. [41] and Hacket-Pain et al. [42], who asserted that climate change is a global phenomenon that has severely impacted forest production along with the distribution and abundance of plant species.
Plants are particularly vulnerable to the effects of climate change, which has led to an increase in research aimed at forecasting plant responses to environmental changes [43]. Our investigation revealed that annual precipitation (bio12, 41.7%), mean annual temperature (bio1, 27.9%), precipitation during the driest month (bio14, 16.1%), and mean temperature of the wettest quarter (bio8, 7.7%) significantly affected the distribution of T. articulata (Figure 4). Rozas et al. [27] emphasized that the radial expansion of T. articulata is significantly associated with precipitation, with water availability in the preceding winter being a crucial factor in growth. In lowland locations, growth was affected by rainfall up to 33 months earlier, whereas in mountainous areas, this influence persisted for 26 months. These data underscore the significant reliance of T. articulata on precipitation, demonstrating its adaptation to fluctuating water availability and its tolerance to arid conditions. Forecast temperature increases and precipitation decreases of up to 30% in certain areas within the Mediterranean basin are predicted to profoundly affect tree biology [44]. Among the climatic variables, precipitation is particularly significant, as it predominantly dictates the distribution of diverse tree species in the Mediterranean region [45,46].
This study found annual precipitation (bio12) to be a key factor influencing the potential habitat range of T. articulata, corroborating the observations of Salvà-Catarineu et al. [47], who noted a comparable pattern for Juniperus phoenicea. Seasonal summer dryness has traditionally resulted in increased mortality rates in juniper populations in several Mediterranean [47,48]. Castaño-Santamaría et al. [43] identified temperature-related factors, specifically isothermality, annual mean diurnal range, temperature seasonality, and mean temperature during the wettest quarter, as the predominant variables affecting beech species in Northwest Spain. Reduced precipitation during spring and summer is anticipated to exacerbate aridity [49], perhaps leading to elevated tree mortality rates in the Mediterranean region [50].
T. articulata, a species with restricted dispersion capabilities, has difficulty adjusting to climate change, and this is exacerbated by habitat fragmentation [51,52]. These findings indicate that climate change may variably affect T. articulata based on the prevailing scenario, making this species a crucial indicator for assessing climate impacts on biodiversity in arid Mediterranean habitats. Owing to the effects of climate change in the Mediterranean area, both in situ and ex situ conservation strategies are strongly recommended [18].
Our CMIP6 forecasts show that the area that is favorable for Tetraclinis articulata to thrive in would shrink across the whole basin. In Spain, Malta, and portions of North Africa, the loss of about 8% of the area will make already patchy stands more likely to break up. The species is worldwide Least Concern, but the European subpopulations are small and very vulnerable. This means that even minor percentage losses can put local persistence and genetic representation at risk, especially in areas with a lot of fire and grazing pressure. Targeted activities include protecting existing nuclei, strengthening connections and corridors along expected shift vectors, and putting aided regeneration and ex situ seed banking at the top of the list to protect against random loss [53].
T. articulata prefers rocky, shallow, often calcareous substrates (SE Spain, Malta) and a wider mix of acidic to alkaline soils in North Africa, usually stony soils with low water-holding capacity that limit competitors but match the species’ xeric adaptations [54]. When choosing reinforcing or restoration locations, management should consider climate appropriateness and edaphic filters (shallow Leptosol-like profiles, high coarse fragments/fissures). With managed browsing and post-fire erosion mitigation, pilot or enrichment plants are possible on well-drained slopes and limestone formations.
In fragile, isolated European stands, an ~8% reduction in acceptable climate space can lead to microrefugia loss or decreased recruitment windows, increasing local extinction risk despite resprouting capacity [53]. Similar decreases might diminish wood/resin yields and tighten harvest limitations in North Africa, affecting artisan income and rural employment, emphasizing the need for sustainable harvest, controlled coppice cycles, and livelihood diversification.
MaxEnt, a presence-only SDM, infers probable climatic suitability—not occupancy—and is vulnerable to sampling bias, background selection, and threshold selections. Omitting essential processes (deforestation rate, fire, land use, competition, dispersal, soil features) can decrease transferability under unfamiliar climates [55]. We used bio climatic variables and inherited GCM/SSP uncertainty. We did not explicitly describe demography, genetics, or dispersal lags. These cautions suggest scenario-robust interpretation. We propose integrating demography/dispersal with SDMs to translate suitability into persistence, pairing models with genetic structure/provenance trials to guide seed sourcing and assisted gene flow, adding fire regime, grazing pressure, land use, and soil layers at operational scales to refine planting maps, and evaluating bias-correction and ensemble modelling to bracket structural uncertainty.

5. Conclusions

This study presents the inaugural Mediterranean-wide, CMIP6-based, multi-horizon projection of Tetraclinis articulata utilising a MaxEnt + GIS framework, quantifying the influence of climatic factors—particularly precipitation seasonality, mean annual temperature, and driest-month precipitation—on current suitability and prospective changes. We demonstrate a gradual reduction in climatically suitable areas from around 8.6% currently to 8.4% (SSP1-2.6) and 8.3/7.5/7.2% (SSP2-4.5/SSP3-7.0/SSP5-8.5) by the end of the century, accompanied by an increase in unsuitable zones (to around 78.6%–79.9%). We also see directional centroid movements (west–northwest for SSP1-2.6/SSP3-7.0/SSP5-8.5; eastward for SSP2-4.5), which means that key habitats are changing shape and the likelihood of fragmentation is growing.
Our analysis goes beyond just summarizing declines. It (i) finds the main climatic factors that affect T. articulata at the basin level, (ii) provides scenario-specific shift vectors to help with corridor design and cross-border planning, and (iii) finds relative refugia and emerging suitability that can help with assisted regeneration, seed sourcing, and ex situ conservation while putting less emphasis on edges that are very vulnerable.
We suggest the following to turn climate suitability into persistence outcomes: (1) combining SDMs with demography/dispersal (integral projection or meta-population models); (2) using genetic structure and provenance trials to help gene flow; (3) adding fire regime, land-use change, and soil constraints to make realized habitat more accurate; (4) testing model ensembles and threshold sensitivity to lower structural uncertainty; and (5) evaluating management interventions (e.g., drought-hardening, mixed-species plantings) in priority refugia and possible corridors. These efforts will make climate-smart strategies stronger so that the species’ genetic diversity and ecological services can stay strong across Mediterranean regions.

Author Contributions

K.M.: conceptualization, experimental design, model running, first draft writing. U.H.: conceptualization, experimental design, model running, first draft writing, revision, project administration, supervision. M.H.A.: literature review, methodology, revision. S.M.A.: conceptualization, experimental design, first draft writing. E.R.A.: literature review, methodology, revision. K.Y.F.: conceptualization, experimental design, first draft writing. M.A.A.: literature review, revision. W.J.: conceptualization, experimental design, first draft writing, revision, project administration, visualization. All authors have read and agreed to the published version of the manuscript.

Funding

Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2025R674), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding authors.

Acknowledgments

The authors acknowledge support from the Princess Nourah bint Abdulrahman University Researchers Supporting Project (number PNURSP2025R674), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.

Conflicts of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

  1. Hayat, U.; Akram, M.; Kour, S.; Arif, T.; Shi, J. Pest Risk Assessment of Aeolesthes sarta (Coleoptera: Cerambycidae) in Pakistan under Climate Change Scenario. Forests 2023, 14, 253. [Google Scholar] [CrossRef]
  2. Hayat, U.; Kour, S.; Akram, M.; Shi, J.; Wiarta, R. Assessing the Global Pest Risk of Aeolesthes sarta with Regards to the Host Specie Populus alba under Climate Change Scenarios. Forests 2023, 14, 1260. [Google Scholar] [CrossRef]
  3. Hayat, U. City longhorn beetle (Aeolesthes sarta): A review of the species, its distribution, ecology, damage, prevention and control. J. For. Sci. 2022, 68, 199–212. [Google Scholar] [CrossRef]
  4. Svenning, J.-C.; Skov, F. Potential impact of climate change on the northern nemoral forest herb flora of Europe. Biodivers. Conserv. 2006, 15, 3341–3356. [Google Scholar] [CrossRef]
  5. Lenihan, J.M.; Bachelet, D.; Neilson, R.P.; Drapek, R. Response of vegetation distribution, ecosystem productivity, and fire to climate change scenarios for California. Clim. Chang. 2008, 87, 215–230. [Google Scholar] [CrossRef]
  6. López de Heredia, U.; Carrión, J.; Jiménez, P.; Collada, C.; Gil, L. Molecular and palaeoecological evidence for multiple glacial refugia for evergreen oaks on the Iberian Peninsula. J. Biogeogr. 2007, 34, 1505–1517. [Google Scholar] [CrossRef]
  7. García-Castaño, J.L.; Balao, F.; Lorenzo, M.T.; Véla, E.; Hadjadj-Aoul, S.; Mifsud, S.; Terrab, A. A complex genetic structure of Tetraclinis articulata (Cupressaceae) in the western Mediterranean. Bot. J. Linn. Soc. 2021, 197, 420–438. [Google Scholar] [CrossRef]
  8. Abdelwahed, F.; Mohamed, R.; Bousselham, K.; Mohsine, Z.; Fatima, E.B.-C.; De Marsan, M.; Abdelwahed, F.F.; Abderrahim, F. Natural durability of Tetraclinis articulata (Vahl) Masters woods against wood decay fungi: Laboratory test. Wood Res. 2015, 60, 865–872. [Google Scholar]
  9. Bourkhiss, M.; Hnach, M.; Bourkhiss, B.; Ouhssine, M.; Chaouch, A. Composition chimique et propriétés antimicrobiennes de l’huile essentielle extraite des feuilles de Tetraclinis articulata (Vahl) du Maroc. Afr. Sci. Rev. Int. Sci. Technol. 2007, 3, 61267. [Google Scholar] [CrossRef]
  10. Esteve-Selma, M. Bosques de Tetraclinis articulata. In Manual de Directrices Ecológicas Para la Conservación de Los Tipos de Habitat de Interés Comunitario en España; Dirección General de Medio Natural y Política Forestal, Ministerio de Medio Ambiente y Medio Rural y: Madrid, Spain, 2009. [Google Scholar]
  11. Sghaier, T.; Cañellas, I.; Calama, R.; Sánchez-González, M. Modelling diameter distribution of Tetraclinis articulata in Tunisia using normal and Weibull distributions with parameters depending on stand variables. iforest-Biogeosci. For. 2016, 9, 702. [Google Scholar] [CrossRef]
  12. Kouider, H.; Assia, L. Synthèse bibliographique sur le thuya de berbérie [Tetraclinis articulata (Vahl) Mast.]. Geo-Eco-Trop 2017, 41, 13–27. [Google Scholar]
  13. Ayache, F. Les Résineux Dans la Région de Tlemcen: Aspect Écologique et Cartographie (Thesis). Magister en Ecologie et Environnement. Master’s Thesis, Université abou Bekr Belkaid Tlemcen, Tlemcen, Algérie, 2014. Available online: http://dspace.univ-tlemcen.dz/handle/112/6642 (accessed on 25 May 2025).
  14. Esteve-Selma, M.; Martínez-Fernández, J.; Hernández, I.; Montávez, J.; Lopez, J.; Calvo, J.; Robledano, F. Effects of climatic change on the distribution and conservation of Mediterranean forests: The case of Tetraclinis articulata in the Iberian Peninsula. Biodivers. Conserv. 2010, 19, 3809–3825. [Google Scholar] [CrossRef]
  15. Ghallab, A.; Boubekraoui, H.; Laaribya, S.; Ben-Said, M. Potential natural vegetation pattern based on major tree distribution modeling in the western Rif of Morocco. iforest-Biogeosci. For. 2024, 17, 405. [Google Scholar] [CrossRef]
  16. Khatib, S.; Sobeh, M.; Bouissane, L. Tetraclinis articulata (vahl) masters: An insight into its ethnobotany, phytochemistry, toxicity, biocide and therapeutic merits. Front. Pharmacol. 2022, 13, 977726. [Google Scholar] [CrossRef] [PubMed]
  17. Marini, K. Climate and environmental change in the Mediterranean–main facts. MedEC Erişim 2018, 1, 2019. [Google Scholar]
  18. Esteve-Selma, M.; Martínez-Fernández, J.; Hernández-García, I.; Montávez, J.; López-Hernández, J.; Calvo, J. Potential effects of climatic change on the distribution of Tetraclinis articulata, an endemic tree from arid Mediterranean ecosystems. Clim. Chang. 2012, 113, 663–678. [Google Scholar] [CrossRef]
  19. Kharouba, H.M.; Nadeau, J.L.; Young, E.; Kerr, J.T. Using species distribution models to effectively conserve biodiversity into the future. Biodiversity 2008, 9, 39–46. [Google Scholar] [CrossRef]
  20. Hayat, U.; Shi, J.; Wu, Z.; Rizwan, M.; Haider, M.S. Which SDM Model, CLIMEX vs. MaxEnt, Best Forecasts Aeolesthes sarta Distribution at a Global Scale under Climate Change Scenarios? Insects 2024, 15, 324. [Google Scholar] [CrossRef]
  21. Phillips, S.J.; Dudík, M.; Schapire, R.E. A maximum entropy approach to species distribution modeling. In Proceedings of the Twenty-First International Conference on Machine Learning, Banff, AB, Canada, 4 July 2004; p. 83. [Google Scholar]
  22. Zhang, X.; Zhao, J.; Wang, M.; Li, Z.; Lin, S.; Chen, H. Potential distribution prediction of Amaranthus palmeri S. Watson in China under current and future climate scenarios. Ecol. Evol. 2022, 12, e9505. [Google Scholar] [CrossRef]
  23. Phillips, S.J.; Anderson, R.P.; Schapire, R.E. Maximum entropy modeling of species geographic distributions. Ecol. Model. 2006, 190, 231–259. [Google Scholar] [CrossRef]
  24. Elith, J.; Phillips, S.J.; Hastie, T.; Dudík, M.; Chee, Y.E.; Yates, C.J. A statistical explanation of MaxEnt for ecologists. Divers. Distrib. 2011, 17, 43–57. [Google Scholar] [CrossRef]
  25. Sánchez-Gómez, P.; Jiménez, J.F.; Vera, J.B.; Sánchez-Saorín, F.J.; Martínez, J.F.; Buhagiar, J. Genetic structure of Tetraclinis articulata, an endangered conifer of the western Mediterranean basin. Silva Fenn. 2013, 47, 1073. [Google Scholar] [CrossRef]
  26. Makkaoui, M.; Abbas, Y.; EL ANTRY-TAZY, S.; Medraoui, L.; Alami, M.; Rabani, S.; Filali-Maltouf, A. Genetic diversity of ten Moroccan populations of Tetraclinis articulata as revealed by inter simple sequence repeat (ISSR) markers. BOIS For. DES Trop. 2020, 345, 15–25. [Google Scholar] [CrossRef]
  27. Rozas, V.; Garcia-Cervigon, A.I.; Garcia-Hidalgo, M.; Rodriguez-Garcia, E.; Olano, J.M. Living on the edge: Legacy of water availability on Tetraclinis articulata secondary growth under semiarid conditions in Morocco. Dendrochronologia 2021, 68, 125853. [Google Scholar] [CrossRef]
  28. Charco, J. Guía de los Árboles y Arbustos del Norte de África; Agencia Española de Cooperación Internacional: Madrid, Spain, 2001. [Google Scholar]
  29. Caudullo, G.; Welk, E.; San-Miguel-Ayanz, J. Chorological maps for the main European woody species. Data Brief 2017, 12, 662–666. [Google Scholar] [CrossRef]
  30. Meryem, M.; Younes, A.; Leila, M.; Mohammed, A.; Salwa, E.A.-T.; Abdelkarim, F.-M. Development of a Core Collection for Tetraclinis articulata Using ISSR Markers and Maximization Strategy. Plant Mol. Biol. Report. 2023, 41, 427–439. [Google Scholar] [CrossRef]
  31. El Mechri, O.; Belgherbi, B.; Benaradj, A.; Berkane, I. Spatial distribution of Tetraclinis articulata (Vahl) Mast. formations in north-western Algeria. Biodivers. Res. Conserv. 2024, 74, 43–52. [Google Scholar] [CrossRef]
  32. Asbabou, A.; Hanane, T.; Gourich, A.A.; Siddique, F.; Drioiche, A.; Remok, F.; Saidi, S.; Adadi, I.; Khamar, H.; Almaary, K.S. Phytochemical profile, physicochemical, antioxidant and antimicrobial properties of Juniperus phoenicea and Tetraclinis articulate: In vitro and in silico approaches. Front. Chem. 2024, 12, 1397961. [Google Scholar] [CrossRef]
  33. Khatib, S.; Mahdi, I.; Drissi, B.; Fahsi, N.; Bouissane, L.; Sobeh, M. Tetraclinis articulata (Vahl) Mast.: Volatile constituents, antioxidant, antidiabetic and wound healing activities of its essential oil. Heliyon 2024, 10, e24563. [Google Scholar] [CrossRef]
  34. Tatebe, H.; Ogura, T.; Nitta, T.; Komuro, Y.; Ogochi, K.; Takemura, T.; Sudo, K.; Sekiguchi, M.; Abe, M.; Saito, F. Description and basic evaluation of simulated mean state, internal variability, and climate sensitivity in MIROC6. Geosci. Model Dev. 2019, 12, 2727–2765. [Google Scholar] [CrossRef]
  35. Babaousmail, H.; Ayugi, B.; Rajasekar, A.; Zhu, H.; Oduro, C.; Mumo, R.; Ongoma, V. Projection of extreme temperature events over the Mediterranean and Sahara using bias-corrected CMIP6 models. Atmosphere 2022, 13, 741. [Google Scholar] [CrossRef]
  36. Aidoo, O.F.; Souza, P.G.C.; da Silva, R.S.; Santana, P.A., Jr.; Picanço, M.C.; Kyerematen, R.; Sètamou, M.; Ekesi, S.; Borgemeister, C. Climate-induced range shifts of invasive species (Diaphorina citri Kuwayama). Pest Manag. Sci. 2022, 78, 2534–2549. [Google Scholar] [CrossRef]
  37. Lu, Y.; Zhao, Q.; Cheng, L.; Zhao, L.; Zhang, H.; Wei, J. The potential global distribution of the white peach scale Pseudaulacaspis pentagona (Targioni tozzetti) under climate change. Forests 2020, 11, 192. [Google Scholar] [CrossRef]
  38. Hayat, U.; Ahmad, M.H.; Sattar, W.; Zhu, Z. A host availability-driven model for predicting the distribution of Red Palm Weevil in China. Crop Prot. 2025, 192, 107176. [Google Scholar] [CrossRef]
  39. Merow, C.; Smith, M.J.; Silander, J.A., Jr. A practical guide to MaxEnt for modeling species’ distributions: What it does, and why inputs and settings matter. Ecography 2013, 36, 1058–1069. [Google Scholar] [CrossRef]
  40. Zare, M.; Moameri, M.; Ghorbani, A.; Sahragard, H.P.; Mostafazadeh, R.; Dadjou, F.; Biswas, A. Modeling habitat suitability of Dorema ammoniacum D Don. in the rangelands of central Iran. Sci. Rep. 2024, 14, 16185. [Google Scholar] [CrossRef] [PubMed]
  41. Monzón, J.; Moyer-Horner, L.; Palamar, M.B. Climate change and species range dynamics in protected areas. Bioscience 2011, 61, 752–761. [Google Scholar] [CrossRef]
  42. Hacket-Pain, A.J.; Cavin, L.; Friend, A.D.; Jump, A. Consistent limitation of growth by high temperature and low precipitation from range core to southern edge of European beech indicates widespread vulnerability to changing climate. Eur. J. For. Res. 2016, 135, 897–909. [Google Scholar] [CrossRef]
  43. Castaño-Santamaría, J.; López-Sánchez, C.A.; Obeso, J.R.; Barrio-Anta, M. Modelling and mapping beech forest distribution and site productivity under different climate change scenarios in the Cantabrian Range (North-Western Spain). For. Ecol. Manag. 2019, 450, 117488. [Google Scholar] [CrossRef]
  44. Giorgi, F.; Lionello, P. Climate change projections for the Mediterranean region. Glob. Planet. Chang. 2008, 63, 90–104. [Google Scholar] [CrossRef]
  45. Allen, C.D.; Macalady, A.K.; Chenchouni, H.; Bachelet, D.; McDowell, N.; Vennetier, M.; Kitzberger, T.; Rigling, A.; Breshears, D.D.; Hogg, E.T. A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. For. Ecol. Manag. 2010, 259, 660–684. [Google Scholar] [CrossRef]
  46. Matías, L.; Jump, A.S. Interactions between growth, demography and biotic interactions in determining species range limits in a warming world: The case of Pinus sylvestris. For. Ecol. Manag. 2012, 282, 10–22. [Google Scholar] [CrossRef]
  47. Salvà-Catarineu, M.; Romo, A.; Mazur, M.; Zielińska, M.; Minissale, P.; Dönmez, A.A.; Boratyńska, K.; Boratyński, A. Past, present, and future geographic range of the relict Mediterranean and Macaronesian Juniperus phoenicea complex. Ecol. Evol. 2021, 11, 5075–5095. [Google Scholar] [CrossRef]
  48. Elmahdy, S.I.; Mohamed, M.M. Factors controlling the changes and spatial variability of Junipers phoenicea in Jabal Al Akhdar, Libya, using remote sensing and GIS. Arab. J. Geosci. 2016, 9, 478. [Google Scholar] [CrossRef]
  49. Paparrizos, S.; Maris, F.; Matzarakis, A. Integrated analysis and mapping of aridity over Greek areas with different climate conditions. Glob. NEST J. 2016, 18, 131–145. [Google Scholar]
  50. Walas, Ł.; Sobierajska, K.; Ok, T.; Dönmez, A.A.; Kanoğlu, S.S.; Dagher-Kharrat, M.B.; Douaihy, B.; Romo, A.; Stephan, J.; Jasińska, A.K. Past, present, and future geographic range of an oro-Mediterranean Tertiary relict: The Juniperus drupacea case study. Reg. Environ. Chang. 2019, 19, 1507–1520. [Google Scholar] [CrossRef]
  51. Pearson, R.G.; Dawson, T.P. Long-distance plant dispersal and habitat fragmentation: Identifying conservation targets for spatial landscape planning under climate change. Biol. Conserv. 2005, 123, 389–401. [Google Scholar] [CrossRef]
  52. Heller, N.E.; Zavaleta, E.S. Biodiversity management in the face of climate change: A review of 22 years of recommendations. Biol. Conserv. 2009, 142, 14–32. [Google Scholar] [CrossRef]
  53. Gómez, P.S.; Stevens, D.; Fennane, M.; Tahar, S.; Gardner, M.F.; Thomas, P.I. Tetraclinis Articulata. Available online: https://threatenedconifers.rbge.org.uk/conifers/tetraclinis-articulata (accessed on 18 May 2025).
  54. Grimshaw, J.; Bayton, R. Tetraclinis articulata. Available online: https://www.treesandshrubsonline.org/articles/tetraclinis/tetraclinis-articulata/ (accessed on 18 May 2025).
  55. Kramer-Schadt, S.; Niedballa, J.; Pilgrim, J.D.; Schröder, B.; Lindenborn, J.; Reinfelder, V.; Stillfried, M.; Heckmann, I.; Scharf, A.K.; Augeri, D.M. The importance of correcting for sampling bias in MaxEnt species distribution models. Divers. Distrib. 2013, 19, 1366–1379. [Google Scholar] [CrossRef]
Forests 16 01600 g001
Figure 1. The study area map and known distribution of Tetraclinis articulata [29].
Figure 1. The study area map and known distribution of Tetraclinis articulata [29].
Forests 16 01600 g001
Forests 16 01600 g002
Figure 2. (a) Evaluation of AUC scores for different Feature Class (FC) combinations across Regularization Multipliers (RMs) for T. articulata leveraging MaxEnt modeling. Validated FCs encompass Auto Features, LQ (Linear + Quadratic), LQH (Linear + Quadratic + Hinge), and LQHPT (Linear + Quadratic + Hinge + Product + Threshold). The intersection of red dashed lines indicates the most promising model for T. articulata, computed according to the highest AUC value. (b) Influence of variable set all vs. selected on the SDMs’ AUC and TSS scores.
Figure 2. (a) Evaluation of AUC scores for different Feature Class (FC) combinations across Regularization Multipliers (RMs) for T. articulata leveraging MaxEnt modeling. Validated FCs encompass Auto Features, LQ (Linear + Quadratic), LQH (Linear + Quadratic + Hinge), and LQHPT (Linear + Quadratic + Hinge + Product + Threshold). The intersection of red dashed lines indicates the most promising model for T. articulata, computed according to the highest AUC value. (b) Influence of variable set all vs. selected on the SDMs’ AUC and TSS scores.
Forests 16 01600 g002
Forests 16 01600 g003
Figure 3. Model evaluation based on (A) the averaged omission and predicted area T. articulata and (B) the ROC curve calculated by MaxEnt as averaged sensitivity versus 1-specificity for Tetraclinis articulata.
Figure 3. Model evaluation based on (A) the averaged omission and predicted area T. articulata and (B) the ROC curve calculated by MaxEnt as averaged sensitivity versus 1-specificity for Tetraclinis articulata.
Forests 16 01600 g003
Forests 16 01600 g004
Figure 4. Percentage contribution of bioclimatic variables used for MaxEnt modelling.
Figure 4. Percentage contribution of bioclimatic variables used for MaxEnt modelling.
Forests 16 01600 g004
Forests 16 01600 g005
Figure 5. Relative importance of environmental variables based on the jackknife test. The figure represents the contribution of each variable to regularized training, test gain, and AUC in the Tetraclinis articulata model.
Figure 5. Relative importance of environmental variables based on the jackknife test. The figure represents the contribution of each variable to regularized training, test gain, and AUC in the Tetraclinis articulata model.
Forests 16 01600 g005
Forests 16 01600 g006
Figure 6. Response curves of predictors of Tetraclinis articulata in the MaxEnt model.
Figure 6. Response curves of predictors of Tetraclinis articulata in the MaxEnt model.
Forests 16 01600 g006
Forests 16 01600 g007
Figure 7. Potential distribution of Tetraclinis articulata under current climatic conditions. Habitat suitability is presented in four colors: pink, highly suitable; blue, suitable; yellow, marginally suitable; and gray, unsuitable.
Figure 7. Potential distribution of Tetraclinis articulata under current climatic conditions. Habitat suitability is presented in four colors: pink, highly suitable; blue, suitable; yellow, marginally suitable; and gray, unsuitable.
Forests 16 01600 g007
Forests 16 01600 g008
Figure 8. Predicted suitable area percentage for Tetraclinis articulata under present and future climatic conditions.
Figure 8. Predicted suitable area percentage for Tetraclinis articulata under present and future climatic conditions.
Forests 16 01600 g008
Forests 16 01600 g009
Figure 9. Potential distribution of Tetraclinis articulata under future climatic conditions for four climate change scenarios. Habitat suitability is presented in four colors: pink, highly suitable; blue, suitable; yellow, marginally suitable; and white, unsuitable.
Figure 9. Potential distribution of Tetraclinis articulata under future climatic conditions for four climate change scenarios. Habitat suitability is presented in four colors: pink, highly suitable; blue, suitable; yellow, marginally suitable; and white, unsuitable.
Forests 16 01600 g009
Forests 16 01600 g010
Figure 10. Centroid shift in Tetraclinis articulata under different climate change scenarios. For each scenario, we take an average point for 2100 as the centroid point of each year (2040, 2060, 2080, and 2100) for each scenario overlapping each other.
Figure 10. Centroid shift in Tetraclinis articulata under different climate change scenarios. For each scenario, we take an average point for 2100 as the centroid point of each year (2040, 2060, 2080, and 2100) for each scenario overlapping each other.
Forests 16 01600 g010
Table 1. Selected bioclimatic variables based on person correlation analysis of Tetraclinis articulata in this study.
Table 1. Selected bioclimatic variables based on person correlation analysis of Tetraclinis articulata in this study.
VariableVariable DescriptionVariable Retainedp-Value
Bio1Annual mean temperature✔<0.8
Bio2Mean diurnal range✔<0.8
Bio3Isothermality (Bio2/Bio7) (×100) >0.8
Bio4Temperature seasonality✔<0.8
Bio5Max temperature of the warmest month >0.8
Bio6Min temperature of coldest month >0.8
Bio7Temperature Annual Range (Bio5–Bio6) >0.8
Bio8Mean temperature of the wettest quarter✔<0.8
Bio9Mean temperature of the driest quarter >0.8
Bio10Mean temperature of warmest quarter >0.8
Bio11Mean temperature of the coldest quarter >0.8
Bio12Annual precipitation✔<0.8
Bio13Precipitation of Wettest Month >0.8
Bio14Precipitation of driest month✔<0.8
Bio15Precipitation Seasonality (Coefficient of Variation) >0.8
Bio16Precipitation of Wettest Quarter >0.8
Bio17Precipitation of Driest Quarter >0.8
Bio18Precipitation of Warmest Quarter >0.8
Bio19Precipitation of Coldest Quarter >0.8
Table 2. Area change (Million km2) of T. articulata under different climate change scenarios.
Table 2. Area change (Million km2) of T. articulata under different climate change scenarios.
SSP1-2.6SSP2-4.5
Current-20402040-20602060-20802080-2100Current-2100Current-20402040-20602060-20802080-2100Current-2100
Unsuitable0.31−0.150.120.170.46−0.330.260.350.381.12
Marginal−0.190.09−0.09−0.17−0.360.27−0.04−0.32−0.42−0.87
Suitable0.00−0.09−0.030.03−0.090.21−0.33−0.17−0.23−0.61
Highly Suitable−0.120.15−0.01−0.04−0.02−0.150.110.140.270.37
SSP3-7.0SSP5-8.5
Current-20402040-20602060-20802080-2100Current-2100Current-20402040-20602060-20802080-2100Current-2100
Unsuitable−0.900.540.580.191.52−1.480.590.590.111.34
Marginal1.01−0.23−0.59−0.29−0.980.87−0.41−0.07−0.11−0.70
Suitable0.43−0.700.08−0.07−0.881.05−0.16−1.07−0.05−1.10
Highly Suitable−0.530.39−0.070.180.33−0.45−0.020.540.060.46
Table 3. Centroid shift in suitable Tetraclinis articulata habitats under different climate change scenarios.
Table 3. Centroid shift in suitable Tetraclinis articulata habitats under different climate change scenarios.
ScenarioLatitudeLongitudeRegionShift Direction
Current1.394° N33.538° EAlgeriaNorthwest
SSP1-2.6–2100−3.429° S33.588° EMoroccoWest
SSP2-4.5–210011.091° N32.501° ETunisaEast
SSP3-7.0–2100−1.947° S34.098° EMoroccoNorthwest
SSP5-8.5–2100−2.985° S34.707° EMoroccoNorthwest
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Mechergui, K.; Hayat, U.; Ahmad, M.H.; Alamri, S.M.; Alamery, E.R.; Faqeih, K.Y.; Aldubehi, M.A.; Jaouadi, W. Forecasting the Impact of Climate Change on Tetraclinis articulata Distribution in the Mediterranean Using MaxEnt and GIS-Based Analysis. Forests 2025, 16, 1600. https://doi.org/10.3390/f16101600

AMA Style

Mechergui K, Hayat U, Ahmad MH, Alamri SM, Alamery ER, Faqeih KY, Aldubehi MA, Jaouadi W. Forecasting the Impact of Climate Change on Tetraclinis articulata Distribution in the Mediterranean Using MaxEnt and GIS-Based Analysis. Forests. 2025; 16(10):1600. https://doi.org/10.3390/f16101600

Chicago/Turabian Style

Mechergui, Kaouther, Umer Hayat, Muhammad Hammad Ahmad, Somayah Moshrif Alamri, Eman Rafi Alamery, Khadeijah Yahya Faqeih, Maha Abdullah Aldubehi, and Wahbi Jaouadi. 2025. "Forecasting the Impact of Climate Change on Tetraclinis articulata Distribution in the Mediterranean Using MaxEnt and GIS-Based Analysis" Forests 16, no. 10: 1600. https://doi.org/10.3390/f16101600

APA Style

Mechergui, K., Hayat, U., Ahmad, M. H., Alamri, S. M., Alamery, E. R., Faqeih, K. Y., Aldubehi, M. A., & Jaouadi, W. (2025). Forecasting the Impact of Climate Change on Tetraclinis articulata Distribution in the Mediterranean Using MaxEnt and GIS-Based Analysis. Forests, 16(10), 1600. https://doi.org/10.3390/f16101600

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop