Research Paper
Print
Research Paper
Habitat characterization and mapping on the western slopes of Mount Hermon in Lebanon
expand article infoHicham El Zein, Lamis Chalak§, Safaa Baydoun|, Carla Khater, Dominique Choueiter#, Maher McKenna#
‡ Unaffiliated, Beirut, Lebanon
§ The Lebanese University, Beirut, Lebanon
| Beirut Arab University, Bekaa, Lebanon
¶ National Council for Scientific Research Lebanon, Beirut, Lebanon
# “STEPping up Nature Reserves Capacity” Project, United Nations Development Programme, funded by the Ministry of Foreign Affairs and International Cooperation of the Italian Republic through the Italian Agency for Development Cooperation (AICS), Beirut, Lebanon
Open Access

Abstract

Aims: Lebanon is renowned in the Levant for its distinctive vegetation types with some biodiversity hotspots as Mount Hermon, with rare and endangered endemic plant species. We aim to present the ecological characteristics and spatial distribution of habitat types present on its western slopes through the analysis of plant communities. Study area: Mount Hermon, Lebanon. Methods: We surveyed 169 plots, each spanning an area of 314 m², from 2020 to 2023, in the district of Rashaya, calculated compositional dissimilarity using the Bray–Curtis index, conducted hierarchical clustering analysis using the unweighted pair group method with arithmetic mean (UPGMA), applied the Nonmetric Multidimensional Scaling (NMDS) method to investigate the relationship between species frequency per site and environmental parameters, and identified significant diagnostic species for each group. Results: We recorded 383 taxa, including 27 narrow endemics. Ten habitat types are described; three at the oro-Mediterranean level: hedgehog-heaths of Astragalus echinus and Noaea mucronata, hedgehog-heaths of Tanacetum densum and Astragalus cruentiflorus, cliffs of Rosularia sempervivum subsp. libanotica; three at the supra-Mediterranean level: grasslands with Eryngium glomeratum, woodlands of Quercus infectoria, Q. coccifera and Crataegus azarolus, evergreen woodlands of Q. coccifera; four at the montane level: scree deciduous woodlands of Prunus korshinskyi and Lonicera nummulariifolia, woodlands of deciduous P. korshinskyi and evergreen Q. coccifera, shrublands of Astragalus gummifer, and deciduous woodlands of Quercus look and Acer monspessulanum subsp. microphyllum. Four environmental variables exhibited significant influences in shaping vegetation composition: elevation, mean annual temperature, slope and northness. Conclusions: Five habitats are novelties proposed as sub-types for the national typology. Floristic affinities with Mount Barouk are highlighted. The nature reserve on the western slopes of Mount Hermon encompasses the majority of the identified habitats. The insights from this study and the habitat map are useful for the development of a management plan and conservation measures.

Taxonomic reference: International Plant Names Index (IPNI 2023).

Abbreviations: EUNIS = European Nature Information System; NMDS = nonmetric multidimensional scaling; UPGMA = unweighted pair group method with arithmetic mean; WGS84 = World Geodetic System, 1984.

Keywords

cliffs, endemism, EUNIS, habitat type, hedgehog-heath, vegetation classification, Lebanese national typology, Levant, montane-Mediterranean, Prunus korshinskyi, Quercus look, oro-Mediterranean

Introduction

Within the eastern Mediterranean, the Levant stands as the junction point between Europe, Africa, and Asia, where three floristic regions meet: the Saharo-Sindian region, the Mediterranean region, and the Irano-Turanian region (Takhtajan 1986). In this biogeographical crossroad, complex biodiversity patterns have emerged (Krupp et al. 2009) and are distributed among the three prominent components of the Levant: the Mediterranean region, the Syro-African Rift, and the Syrian desert. Geographic and geomorphologic features allow the demarcation between the Northern and the Southern Levants. The Northern Levant, extending from the Nur Mountains northward to Galilee southward, is characterised by its distinctive mountainous topography (Abel 1933; Vaumas 1954; Suriano 2013). Mount Lebanon, Anti-Lebanon, and Mount Hermon are located in the Northern Levant, forming the highest ranges across the entire Levant (Vaumas 1954). These ranges have been recognized as regional hotspots of biodiversity within the larger Mediterranean Basin ‘macro’ hotspot, attributed to their remarkable floristic endemism which accounts for 12% of the total flora (Médail and Quézel 1997, 1999; Myers et al. 2000; Verlaque et al. 2001; Cañadas et al. 2014). At the Lebanese national scale, these mountain ranges are of high-priority conservation value as they have been identified as Important Plant Areas (Bou Dagher-Kharrat et al. 2018) and as Key Biodiversity Areas (El Zein et al. 2018) due to the simultaneous presence of rare and endangered endemic plant species.

Lebanon is renowned for its distinctive vegetation types in the Levant. Since ancient times, dating back to 2600 years BC, its mountains have been recognized for their remarkable and iconic conifer forests, composed of cedar (Cedrus libani), fir (Abies cilicica), and juniper (Juniperus excelsa) (Vaumas 1954; Beals 1965; Mikesell 1969). Nevertheless, other types of vegetation also occur in these mountains, forming unique associations of plant species (Post and Dinsmore 1932; Mouterde 1966, 1970, 1984). Several publications have described the vegetation of Lebanon. The description of the climactic series of vegetation of Lebanon and their spatial distribution constitutes one of the greatest achievements in this discipline for the country (Abi-Saleh 1982; Abi-Saleh and Safi 1988). Complementary efforts have provided detailed information about the phytosociological associations that compose some forest types on the western slopes of the Mount Lebanon range (Chouchani et al. 1975; Barghachoun 1976; de Foucault et al. 2013; Stephan et al. 2019). Recently, a typology of the natural and semi-natural terrestrial habitat types has been developed (El Zein et al. 2022) aligned to the hierarchical structure of the habitat classification of the European Nature Information System (EUNIS) (Davies et al. 2004) to facilitate efforts for conservation across the Mediterranean Basin and in neighboring countries. Moreover, the use of an international classification system improves the recognition of local diversity and natural patrimony.

However, detailed descriptions of the plant communities for each habitat type in Lebanon still require improvement and are continuously developing. The habitat types present on the western slopes of Mount Hermon in Lebanon were partly characterised through the vegetation series (Abi-Saleh and Safi 1988). Part of the area was designated as a nature reserve in December 2020, promoting the undertaking of comprehensive studies to document and conserve the natural patrimony present. The primary objective of our study was to address this knowledge gap and comprehend the ecological characteristics and spatial distribution of vegetation types. This paper consequently proposes to characterize and map the different habitats of western Mount Hermon in Lebanon based on their plant communities.

Study area

Mount Hermon, or Jabal al-Shaykh in Arabic, is administratively divided into four parts: the Lebanese western slopes, the Syrian eastern slopes, the United Nations buffer zone where the United Nations Disengagement Observer Force (UNDOF) operates, and the Israeli-occupied southern slopes that are part of the Golan Heights (Dar 1988). It stands as the second-highest mountain in the Levant, reaching an elevation of 2,814 m a.s.l., and is often considered as the southern continuation of the Anti-Lebanon mountain range. The massif extends 25 km from west to east and 45 km from north to south . The entire range spans an area of approximately 1,000 km2, with half of it located in Lebanon (Clermont-Ganneau 1903; Abel 1933).

Most of the upper stratum of Mount Hermon is dominated by Jurassic limestone (Dubertret 1955). In the mountain’s circumference, Cretaceous strata occur with occasional veins of basalt (Dubertret 1955). In most of Lebanon, the soils are young, characterized by poor consistency and shallowness, especially on sloping terrains (Clermont-Ganneau 1903; Vaumas 1954). These conditions contribute to the development of terra rossa soils (Gèze 1956). On the highest slopes, rough terrain features have been created by karstic erosion, such as crags, boulders or sinkholes. Precipitation is quickly absorbed through the porous rocky substrate and feed different watersheds. Westward and southward, the waters feed the headstreams of the Jordan River, while eastward it feeds other springs descending to the Damascus basin (Abel 1933).

Mount Hermon has a typically Mediterranean climate, with January as the coldest month, and July and August as the warmest (Abi-Saleh and Safi 1988). The annual average temperature in Rashaya, at the base of Mount Hermon at 1,200 m a.s.l., is 15.6°C (Baldy 1959). In Rachaya, the mean maximum temperature of the warmest month is 34.1°C, while the mean minimum temperature of the coldest month is 0.5°C (Baldy 1959). The mean annual rainfall in Rashaya is 675 mm (Baldy 1959), reaching around 1,500 mm at peak (YMCA Lebanon 2005; Ayalon et al. 2013). Approximately 80% of the annual rainfall takes place between November and March. The growing season spans from May to mid-September, during which the majority of plant species flower and set seed.

The western slopes of Mount Hermon, part of the Lebanese territory (Figure 1), have been recognized for their floristic diversity (Arnold et al. 2015; Baydoun and Arnold 2017) and the presence of at least 21 endangered endemic plant taxa (El Zein and Kahale 2022). The flora of the area remains relatively understudied. For instance, a species not previously reported in the flora of Lebanon was recently observed there (El Zein et al. 2023). The first checklist of the flora of western Mount Hermon (Arnold et al. 2015) highlighted the significant contribution of the region to traditional ethnobotanical knowledge held by the local communities. The prolonged historical interactions between the land and the successive communities living there have given rise to diverse traditions and cultural heritages (Dar 1993; Farra Haddad 2021). Moreover, the area constitutes a repository for wild crop relatives and a cradle for the domestication of wheat ancestors (Ghossain et al. 2023).

Figure 1. 

View on the western slopes of Mount Hermon in May 2019, Lebanon.

Methods

Vegetation sampling

This study was conducted annually from 2020 to 2023, between May and August, corresponding to the spring and summer seasons, on the Lebanese western slopes of Mount Hermon, in the district of Rachaya. The sampled area ranged between elevations of 1,000 and 2,500 meters. Geographic coordinates (WGS84) for each plot were directly recorded in the field. The coordinates and elevation of the plots are provided in Suppl. material 1. We surveyed 169 circular plots, each spanning an area of 314 m². Several studies recommend surveying plots of large size, reaching up to 400 m², across diverse habitat types, including deserts, shrublands, temperate and tropical forests (Gillison 2001; Archaux et al. 2007; Hunter and Hunter 2020; Hao et al. 2021; Montenegro-Hoyos et al. 2022, Zeballos et al. 2023). For security reasons, we had to maintain a specific distance from the country borders, limiting the accessible area of the study site. The distribution of the survey plots is mapped in Figure 2. In each plot, we documented all vascular plant species, noting their respective cover-abundance through the phytosociological method (Braun-Blanquet 1932), and subsequently transformed the data using the ordinal scale (van der Maarel 1979) as follows: r: 1; +: 2; 1: 3; 2m: 4; 2a: 5; 2b: 6; 3: 7; 4: 8; 5: 9.

Artificial habitats, such as urban areas, planted forests, and agricultural lands, were not surveyed. The flora of Syria and Lebanon (Mouterde 1966, 1970, 1984) served as the reference for species identification. Species nomenclature was based on the International Plant Names Index (IPNI 2023).

Figure 2. 

Map showing the geographic situation of Mount Hermon in Lebanon and vegetation plots surveyed between 2020 and 2023.

Environmental data

We extracted climatic variables, specifically the average annual precipitation and mean annual temperature calculated for the period from 1981 to 2010, from rasters provided by CHELSA database (Karger et al. 2017). The rasters have a resolution of 30 arc-seconds. Some studies demonstrated the significant role of these variables in influencing vegetation patterns and floristic composition (Dirnböck et al. 2002; Morris et al. 2016; Cabido et al. 2018; Zeballos et al. 2023). We extracted geomorphometric data, namely slope and aspect from SRTM data (Farr et al. 2007) available from the United States Geological Survey (USGS 2024), with a resolution of 1 arc-second. We transformed aspect into two components: eastness, represented by the sine, and northness, represented by the cosine. Eastness informs us about the slope orientation, ranging from west (sine = -1) to east (sine = 1). The sine value is 0 for slopes facing north or south. On the other hand, northness indicates the slope’s orientation to the north (cosine = 1) or south (cosine = -1), with a cosine value of 0 for slopes facing east or west (Joly and Gillet 2017; Cheng et al. 2023). Elevation was recorded in the field. We prepared maps with the QGIS software (QGIS Development Team 2023), and downloaded the shapefiles from the DIVA-GIS website (DIVA-GIS 2023).

Data analysis

Initially, we calculated the compositional dissimilarity coefficients between sampled sites using the Bray–Curtis index, also known as the Steinhaus index (Bloom 1981; Somerfield 2008). Then, we conducted a hierarchical clustering analysis to group the sites, employing the unweighted pair group method with arithmetic mean (UPGMA; Sokal and Michener 1958; Belbin and McDonald 1993). Secondly, we applied the Nonmetric Multidimensional Scaling (NMDS) method (Kruskal 1964; Legendre and Legendre 2012) for multivariate ordination to complement the hierarchical cluster analyses. Additionally, NMDS was employed to investigate the relationship between species cover abundance per plot and environmental parameters, namely elevation, eastness, northness, slope, mean annual mean temperature and annual precipitation. The species scores were expanded based on the Wisconsin double standardization of the input data. We performed the NMDS ordination in two dimensions (Legendre and Legendre 2012). The analysis was repeated 100 times using random starting configurations to allow the algorithm to explore a larger portion of the solution space, reducing the likelihood of converging to a local minimum. For visual representation, we plotted the NMDS ordination against the environmental gradients to illustrate the ecological characteristics of the different vegetation types. We tested the overall significance of the NMDS ordination by applying a one-way Analysis of Similarities (ANOSIM; Clarke 1993; Clarke and Warwick 1994).

In a third step, we identified diagnostic species significantly associated to the different clusters. This analysis calculates the probability of having discriminating species for each group of sampled sites. We arbitrarily set the number of clusters to 10, based on the number of groups obtained from the cluster analysis. For each taxa, we calculated the percentage frequency, which indicates the constancy, and the phi-coefficient of association, which represents the fidelity (Tichý and Chytrý 2006). The phi-coefficient ranges from -1, indicating maximum negative fidelity, to 1, indicating maximum positive fidelity. We assessed the statistical significance by direct calculation of the probability of observed taxa concentrations using Fisher’s exact test. Taxa with constancy ≥ 25% and significance p ≤ 0.001 in at least one group of sites were considered constant. Taxa with a phi > 0.6 were considered highly diagnostic, those with phi > 0.4 were deemed quite diagnostic, and those with phi > 0.23 were considered relatively diagnostic.

We carried out all statistical analyses, as well as the extraction of bioclimatic variables, using the R software (R Core Team 2022) and associated packages: BiodiversityR (Kindt 2023), devtools (Wickham et al. 2022b), dplyr (Wickham et al. 2023b), geoveg (von Lampe and Schellenberg 2023), ggfortify (Horikoshi et al. 2023), ggplot2 (Wickham et al. 2023c), ggvegan (Simpson 2023), labdsv (Roberts 2023), lattice (Sarkar et al. 2023), memoise (Wickham et al. 2021), metR (Campitelli 2023), permute (Simpson et al. 2022), remotes (Csárdi et al. 2021), rlang (Henry and Wickham 2023), scales (Wickham et al. 2022a), terra (Hijmans et al. 2023), tibble (Müller et al. 2023), tidyr (Wickham et al. 2023a), vegan (Oksanen et al. 2022), and withr (Hester et al. 2022).

We finally associated characterised habitat types to the national habitat typology of Lebanon (El Zein et al. 2022).

Results

Species richness and endemism

The fieldwork yielded a total of 3,120 observations, documenting 383 different plant taxa within the elevation range of 1,000 and 2,500 m a.s.l. The list of the species observed is provided in Suppl. material 2. The survey comprised 47 plant families, with Fabaceae, Asteraceae, Poaceae, Apiaceae, and Lamiaceae accounting for 56.7% of the total identified taxa (Figure 3). Fabaceae consisted of 67 taxa (17.5%), Asteraceae of 52 taxa (13.5%), Poaceae of 40 taxa (10.4%), Apiaceae of 30 taxa (7.8%), and Lamiaceae of 28 taxa (7.3%). The most diverse genera were Trifolium with 21 taxa, followed by Astragalus (10), Silene (7), Allium (6), Bromus (6), Crepis (6), Galium (6), and Vicia (6). In terms of life form, the taxa were distributed between 38 phanerophytes, 22 chamaephytes, 181 hemicryptophytes, 17 geophytes, and 125 therophytes. Among the phanerophytes, we found 11 tree taxa, comprising one conifer (Juniperus excelsa), one broadleaf evergreen tree (Quercus coccifera), and nine broadleaf deciduous trees (Acer monspessulanum subsp. microphyllum, Crataegus azarolus, Pistacia terebinthus, Prunus korshinskyi, P. cocomilia, Pyrus syriaca, Quercus infectoria, Q. look and Styrax officinalis). Twenty-four taxa were endemic to Mount Hermon and Mount Lebanon, including Allium feinbergii, A. libani, Astragalus coluteoides, A. cruentiflorus, A. hermoneus, Bellevalia hermonis, Centaurea drabifolia subsp. libanotica, C. hololeuca, Cousinia hermonis, Crepis robertioides, Draba antilibanotica, D. vesicaria, Galium libanoticum, Hypericum libanoticum, Lophiolepis lappacea subsp. hermonis, Marrubium globosum subsp. libanoticum, Papaver libanoticum, Phlomis brevilabris, Quercus look, Salvia rubifolia, Scilla libanotica, Scutellaria utriculata, Silene libanotica, and S. makmeliana; two taxa endemic to Mount Hermon and Anti-Lebanon, namely Centaurea iberica subsp. hermonis and Euphorbia erinacea; and one taxon endemic exclusively to Mount Hermon, Erysimum verrucosum.

Figure 3. 

Number of taxa per family surveyed in Mount Hermon.

Classification results

The classification is illustrated in Figure 4, revealing three main branches. The first branch, highlighted in red on the left, encompasses all sites surveyed above 2,000 m a.s.l. in elevation at the oro-Mediterranean belt. The second branch, colored in green, unites sites surveyed between 1,000 and 1,600 m a.s.l. in elevation at the supra-Mediterranean level. The last branch, boxed in yellow, includes sites sampled between 1,600 and 2,000 m a.s.l. in elevation at the montane-Mediterranean level. In total, ten clusters were formed from the 169 plots. Within the oro-Mediterranean branch, we identified two clusters of thorny hedgehog-heaths and one cluster of plots surveyed in cliffs. Within the supra-Mediterranean branch, one cluster comprises plots surveyed in grasslands, along with two additional clusters of plots surveyed in woodlands, one at the meso-Mediterranean level and another at the supra-Mediterranean level. Within the montane-Mediterranean branch, we observed one cluster representing plots surveyed in shrublands, and three clusters of plots surveyed in woodlands. These included scree deciduous woodlands, partially evergreen and deciduous woodlands and deciduous woodlands.

An abbreviated synoptic table, containing only diagnostic and constant species, is provided with each cluster (Table 1). It also includes the number of plots per cluster. The entire synoptic table is presented in Suppl. materials 3, 4.

Figure 4. 

Cluster dendrogram built using UPGMA from the dissimilarity matrix among plots of western slopes of Mount Hermon in Lebanon.

Table 1.

Abbreviated synoptic table of the ten groups obtained from the classification of 169 plots surveyed in the western slopes of Mount Hermon in Lebanon. The frequency (%) of occurrence and phi coefficient of all diagnostic and constant plant species in each group is shown. The species are sorted by decreasing fidelity (phi) within each group. phi > 0.6: dark grey; phi > 0.4: grey; phi > 0.23: light grey. Only species with constancy (Freq) ≥ 25% and significance p ≤ 0.001 in at least one group were included in the table. Vegetation types are: 1, Oro-Med. hedgehog-heaths of Astragalus echinus and Noaea mucronata; 2, Oro-Med. hedgehog-heaths of Tanacetum densum and Astragalus cruentiflorus; 3, Oro-Med. montane cliffs of Rosularia sempervivum subsp. libanotica; 4, Supra-Med. grasslands with Eryngium glomeratum; 5, Upper supra-Med. woodlands of Quercus infectoria, Q. coccifera and Crataegus azarolus; 6, Lower supra-Med. woodlands of Quercus coccifera; 7, Scree montane woodlands of Prunus korshinskyi and Lonicera nummulariifolia; 8, Montane woodlands of Prunus korshinskyi and Quercus coccifera; 9, Montane thickets of Astragalus gummifer; 10, Montane woodlands of Quercus look and Acer monspessulanum subsp. microphyllum. Med = Mediterranean.

Group number 1 2 3 4 5 6 7 8 9 10
Number of sites per group 30 10 13 15 22 22 19 14 6 18
Species Freq phi Freq phi Freq phi Freq phi Freq phi Freq phi Freq phi Freq phi Freq phi Freq phi
Prunus prostrata 93 0.46 80 0.36 62 0.23
Noaea mucronata 70 0.65
Astragalus echinus 60 0.63
Onobrychis cornuta 57 0.49
Cousinia hermonis 53 0.37
Eryngium heldreichii 50 0.65
Astragalus coluteoides 40 0.53
Acantholimon libanoticum 33 0.51
Verbascum cedreti 33 0.41
Veronica polifolia 30 0.39
Marrubium globosum subsp. libanoticum 43 0.34
Tanacetum densum 90 0.62 54 0.31
Bromus tomentellus 57 0.30 90 0.58
Alkanna orientalis 80 0.75
Acantholimon ulicinum 70 0.71
Astragalus cruentiflorus 60 0.63
Sabulina juniperina 50 0.49 31 0.26
Odontarrhena condensata subsp. flexibilis 30 0.53
Rosularia sempervivum subsp. libanotica 92 0.70
Ferulago trachycarpa 85 0.67
Arabis caucasica 77 0.79
Silene odontopetala 62 0.70
Scrophularia libanotica var. australis 46 0.61
Nepeta cilicica 46 0.41
Rhamnus libanotica 38 0.52
Lamium garganicum subsp. striatum 38 0.44
Arenaria deflexa 31 0.53
Campanula cymbalaria 31 0.53
Heracleum humile 31 0.53
Brunnera orientalis 31 0.50
Avena sterilis 87 0.52 82 0.48
Eryngium glomeratum 73 0.40 71 0.38
Echinops spinosissimus subsp. macrolepis 67 0.48
Trifolium boissieri 60 0.38 55 0.33 50 0.29
Picnomon acarna 47 0.46
Rhamnus punctata 33 0.37
Bituminaria bituminosa 27 0.50
Carthamus glaucus 27 0.45
Ononis spinosa subsp. leiosperma 27 0.41
Crataegus azarolus 100 0.59 59 0.27
Salvia rubifolia 68 0.34
Trifolium clusii 50 0.38 45 0.34
Chrysojasminum fruticans 27 0.33
Quercus coccifera 86 0.38 95 0.44 79 0.32
Hordeum bulbosum 93 0.28 91 0.27 100 0.33
Erysimum verrucosum 77 0.27 83 0.31
Dactylis glomerata 82 0.25 95 0.34 89 0.29
Trifolium stellatum 77 0.65
Quercus infectoria 64 0.44 77 0.57
Euphorbia hierosolymitana var. hierosolymitana 68 0.64
Galium libanoticum 64 0.32
Lathyrus aphaca 59 0.63
Trifolium purpureum 59 0.57
Trifolium grandiflorum 55 0.61
Bromus sterilis 55 0.37
Crepis reuteriana 50 0.58
Lagoecia cuminoides 50 0.45
Eryngium creticum 50 0.41
Rhagadiolus edulis 36 0.24 50 0.38
Poa bulbosa 50 0.38
Campanula rapunculus subsp. lambertiana 41 0.52
Lolium perenne 41 0.47
Achnatherum bromoides 41 0.41
Coronilla scorpioides 36 0.58
Poa pratensis 36 0.50
Klasea cerinthifolia 36 0.46
Pistacia terebinthus 36 0.43
Lapsana communis subsp. pisidica 36 0.42
Scutellaria brevibracteata 36 0.34 33 0.30
Trifolium scabrum 36 0.31
Hordeum spontaneum 32 0.54
Ononis natrix 32 0.54
Trifolium dasyurum 32 0.50
Vicia tenuifolia 32 0.46
Scabiosa palaestina 32 0.45
Fibigia clypeata subsp. clypeata 32 0.40
Trifolium plebeium 32 0.36
Melica inaequiglumis 27 0.45
Johrenia dichotoma 27 0.43
Salvia multicaulis 27 0.41
Scorzonera phaeopappa 27 0.39
Orlaya platycarpos 27 0.39
Allium stamineum 27 0.35
Trifolium pilulare 27 0.33
Galium incanum 42 0.41
Phlomis chrysophylla 59 0.30 64 0.34 86 0.51
Lonicera nummulariifolia 53 0.29 67 0.41
Prunus korshinskyi 47 0.28 57 0.37
Pseudoroegneria libanotica 47 0.28
Euphorbia erinacea 36 0.39
Phleum montanum 29 0.42
Astragalus gummifer 100 0.85
Rubia tenuifolia 82 0.25 100 0.37
Cephalaria stellipilis 84 0.38 71 0.29 89 0.41
Quercus look 43 0.24 83 0.61
Acer monspessulanum subsp. microphyllum 67 0.33
Eryngium billardierei 33 0.26 44 0.38

Ordination and environmental gradients

The arrangement of the sites in a two-dimensional NMDS ordination space (Figure 5) closely resemble the grouping obtained from the classification (K = 2; stress = 0.1541). Retaining the numbering and colors of the groups from the classification aids in an easy comparison with Figure 4. The degree of separation between groups, indicating differences in vegetation composition, was significant (ANOSIM R = 0.80; p < 0.0001). Community composition patterns per plot correlate with environmental gradients. The arrows in the ordination space represent the principal direction of variation and the strength of correlation for the environmental variables. Four environmental variables exhibited significant correlations with the NMDS axes, signifying their crucial influences in shaping vegetation composition: elevation (r² = 0.889; p < 0.001), mean annual temperature (temp; r² = 0.858; p < 0.001), slope (r² = 0.227; p < 0.001) and northness (r² = 0.082; p < 0.01). The two other environmental variables didn’t show a significant correlation (p > 0.3): annual precipitation (precip; r² = 0.012) and eastness (r² = 0.007). Axis 1 distinctly delineates the supra-, montane- and oro-Mediterranean levels. Elevation and temperature run almost parallel to the first axis, indicating a direct relationship between these two environmental variables. Therefore, Axis 1 can be interpreted as an altitudinal gradient where temperature increases in the opposite direction. Axis 2 separates clusters within each altitudinal level, with slope being the variable most correlated to this axis. The fourth significant variable, northness, displays a diagonal orientation in the ordination space, suggesting a combined correlation with variations along both NMDS axes.

Figure 5. 

NMDS Ordination of 169 plots sampled in western Mount Hermon in Lebanon. Colors indicate the 10 groups obtained from the UPGMA clustering. Arrows indicate the environmental variables. temp = mean annual temperature; precip = annual precipitation.

Characterisation of habitat types

The habitat types are described based on the grouping obtained from the classification, the correlation with environmental variables, as displayed in the NMDS ordination, and their constant and diagnostic plant species. The code of habitat type according to the Lebanese national typology is also provided. The habitats are presented in the same order of grouping of the classification.

Oro-Mediterranean belt (2,000–2,800 m a.s.l.)

In Mount Hermon, the oro-Mediterranean belt features a complex succession of low thorny-cushion shrublands, rocky grasslands, limestone rock pavements, cliffs and screes. The environment of this altitudinal belt is characterised by bare mineral substrate with limited soil presence and scarce vegetation cover owing to harsh climatic conditions, including abundant precipitation, snow cover in winter, prolonged summer drought, and the intense wind.

1. Oro-Mediterranean hedgehog-heaths of Astragalus echinus and Noaea mucronata in thalweg and slopes – S759_LB1

This habitat is composed of low thorny-cushion shrublands known as hedgehog-heaths (Figure 6). This vegetation is characterised by the diagnostic species Prunus prostrata, Noaea mucronata, Astragalus echinus, Onobrychis cornuta, Cousinia hermonis, Eryngium heldreichii, Astragalus coluteoides, Acantholimon libanoticum, Verbascum cedreti, and Veronica polifolia in decreasing diagnostic value. This type of hedgehog-heaths is found in slopes and thalweg where the soil is relatively deeper. Other taxa are commonly found without being diagnostic or constant, namely Astragalus hermoneus, A. nummularius subsp. trichopterus, Campanula stricta var. libanotica, Cruciata taurica, Dichoropetalum alpinum and Teucrium orientale. Twelve of the twenty-six endemic taxa, such as Astragalus hermoneus, Lophiolepis lappacea subsp. hermonis, Marrubium globosum subsp. libanoticum, and Phlomis brevilabris, were observed in this habitat type.

Figure 6. 

Oro-Mediterranean hedgehog-heaths of Astragalus echinus and Noaea mucronata in thalweg and slopes in the western slopes of Mount Hermon in Lebanon in 2020 A. Dominated by Astragalus hermoneus and A. coluteoides; B. Dominated by Onobrychis cornuta and Prunus prostrata; C. Dominated by A. echinus and Noaea mucronata; D. Dominated by Eryngium heldreichii.

2. Oro-Mediterranean hedgehog-heaths of Tanacetum densum and Astragalus cruentiflorus – S759_LB1

This habitat represents another type of hedgehog-heath, distinguished by the diagnostic taxa Tanacetum densum, Bromus tomentellus, Alkanna orientalis, Acantholimon ulicinum, Astragalus cruentiflorus, Sabulina juniperina, and Odontarrhena condensata subsp. flexibilis in decreasing diagnostic value (Figure 7). The surveyed sites exhibited pronounced xerophytic conditions, attributed to their exposure to wind and sun, along with the scarcity of soil, typical of sites located on the hilltops or south-oriented slopes. Three endemic taxa, namely Cousinia hermonis, Draba antilibanotica, and D. vesicaria, were observed in this habitat type.

Figure 7. 

Oro-Mediterranean hedgehog-heaths of Tanacetum densum and Astragalus cruentiflorus in the western slopes of Mount Hermon in Lebanon in 2022.

3. Oro-Mediterranean and montane limestone cliffs of Rosularia sempervivum subsp. libanotica – U38

Limestone cliffs are present in both montane and oro-Mediterranean levels (Figure 8). The diagnostic taxa are Rosularia sempervivum subsp. libanotica, Ferulago trachycarpa, Arabis caucasica, Silene odontopetala, Scrophularia libanotica var. australis, Rhamnus libanotica, Lamium garganicum subsp. striatum, Arenaria deflexa, and Campanula cymbalaria in decreasing diagnostic value. Some diagnostic species are always located at the bottom of the cliffs, namely Nepeta cilicica, Heracleum humile, and Brunnera orientalis. Four endemic taxa, namely Bellevalia hermonis, Centaurea drabifolia subsp. libanotica, Scutellaria utriculata, and Silene makmeliana, were observed in this habitat type.

Figure 8. 

Limestone cliffs at the oro-Mediterranean level in the western slopes of Mount Hermon in Lebanon in 2022; B. Rosularia sempervivum subsp. libanotica in flower.

Supra-Mediterranean belt (1,000–1,600 m a.s.l.)

The supra-Mediterranean altitudinal level is characterised by the presence of sclerophyllous forests of Quercus coccifera and deciduous broadleaf woodland of Q. infectoria, primarily found in the hills at the base of Mount Hermon. This altitudinal belt exhibits a predilection for human activities. The general landscape consisted of rocky grasslands and shrublands used as pastures, interspersed with patches of woodlands, agricultural fields, and human settlements.

4. Supra-Mediterranean grasslands with Eryngium glomeratum – R1E_LB1

These grasslands are the results of the degradation of the supra-Mediterranean evergreen woodlands of Quercus coccifera and evergreen-deciduous woodlands of Q. infectoria (Figure 9). They are maintained as grasslands by intensive grazing which prevent the recolonisation by trees. The diagnostic taxa are Avena sterilis, Eryngium glomeratum, Echinops spinosissimus subsp. macrolepis, Hordeum bulbosum, Trifolium boissieri, Picnomon acarna, Rhamnus punctata, Bituminaria bituminosa, Carthamus glaucus, and Ononis spinosa subsp. leiosperma in decreasing diagnostic value. Sometimes very sparse shrubs or low trees of Q. coccifera, dwarfed by overgrazing, occur in the rocky grasslands. Two endemic taxa, Astragalus oleifolius and Centaurea iberica subsp. hermonis, were observed in this habitat type.

5. Upper supra-Mediterranean deciduous woodlands of Quercus infectoria, Q. coccifera, and thickets of Crataegus azarolus – T1953_LB1

This group includes two distinct types of vegetation that share a similar floristic composition. One of them, the thickets of Crataegus azarolus, represents the degraded form of the deciduous woodlands of Q. infectoria, and Quercus coccifera (Figure 10). The prevalence of Quercus infectoria, either dominantly or in conjunction with Quercus coccifera, is a defining characteristic of this woodland type, referred to as sub-Mediterranean thermophilous deciduous oak forest. The diagnostic species are Crataegus azarolus, Salvia rubifolia, Trifolium clusii, Chrysojasminum fruticans, Q. coccifera and Q. infectoria. Thickets are dense stands of shrubs or low trees, typically not exceeding 2 m in height. The thickets of C. azarolus are widespread at the upper supra-Mediterranean level, resulting from the degradation of woodlands. These formations act as pioneer habitats, colonising abandoned agricultural terraces and paving the way for the subsequent establishment of oak forests. The thickets are scattered intermittently, alternating with extensive patches of grasslands, where Phlomis chrysophylla, Hordeum bulbosum, Rubia tenuifolia, and Erysimum verrucosum are also considered diagnostic species, albeit with a low fidelity coefficient as they are also found in other vegetation groups. Erysimum verrucosum is endemic to Mount Hermon, and Salvia rubifolia is endemic to Mount Lebanon, Anti-Lebanon ranges, and Mount Hermon. In total, four endemic taxa were observed in these habitat types.

The dominance of the evergreen Q. coccifera and deciduous Q. infectoria in woodlands alternates according to geomomorphology and soil depth. Small hills with rocky and shallow soils host woodlands of Q. coccifera. In contrast, thalwegs, characterised by the presence of soil, greater protection from the wind, and lower sun exposition, support woodlands with tall trees, particularly Q. infectoria. However, in the past, some parts of Mount Hermon’s thalwegs were preferred for wood exploitation and agriculture, leading to tree removal. Areas with soils are distributed between the rocky hills, giving them this linear shape. Presently, some of these thalwegs face intensive grazing, hindering the regeneration of the deciduous woodlands. This has resulted in a composite landscape featuring alternating successions of linear thalwegs grasslands with tall Q. infectoria trees inserted among rocky hills with evergreen shorter Q. coccifera trees, creating a characteristic mosaic of vegetation types.

Figure 9. 

Supra-Mediterranean grasslands in the western slopes of Mount Hermon in Lebanon in 2020 A. above Qsar el-Jabal, next to Ain Hircha in summer; B. above Berket el-Yebse in spring.

Figure 10. 

Upper supra-Mediterranean woodlands of Quercus infectoria, Q. coccifera with thickets of Crataegus azarolus in 2020 A. In Hima Hassan, in July; B. A thalweg with grassland within the woodland above Kfar Qouq in May; C. Contrast between preserved woodlands above Hima el-Kadarin on the left and degraded woodland with thickets of C. azarolus characterized by their reddish color in August; D. Degraded slope with remaining Q. infectoria in July.

6. Lower supra-Mediterranean evergreen Quercus coccifera woodlands – T213_LB2

Climactic forests of the lower supra-Mediterranean in western Mount Hermon are characterised by the dominance of Quercus coccifera, reaching up to 70% of the proportion of trees in some sites (Figure 11). Stands are dense but often low, not exceeding 4 m in height, and display a coppice physiognomy believed to be a result of a long history of exploitation and forest fires. Some of the most significant diagnostic tree species of this habitat are Q. coccifera, Q. infectoria and Pistacia terebinthus. Some of the most significant diagnostic herbaceous taxa are Dactylis glomerata, Trifolium stellatum, T. purpureum, T. grandiflorum, Euphorbia hierosolymitana var. hierosolymitana, Galium libanoticum, Lathyrus aphaca, Bromus sterilis, Crepis reuteriana, Lagoecia cuminoides, Eryngium creticum, Rhagadiolus edulis, Poa bulbosa, P. pratensis, Campanula rapunculus, Lolium perenne, Achnatherum bromoides and Klasea cerinthifolia. Four endemic taxa, namely Centaurea iberica subsp. hermonis, Galium libanoticum, Salvia rubifolia, and Silene makmeliana, were observed in these habitat types.

Figure 11. 

Quercus coccifera woodlands. A. In Wadi el-Feqaa; B. In Hima el-Kadarin.

Montane-Mediterranean belt (1,600–2,000 m a.s.l.)

Mount Hermon is distinctive for hosting sparse deciduous woodlands, primarily composed of the endemic thermophilous oak Quercus look. The specific topography, slope aspect, and steepness play crucial roles in determining the type of woodlands present. Additionally, historical tree exploitation has significantly impacted the physiognomy of these woodlands, often resulting in a succession of sparse woodlands and degraded habitats, such as rocky shrublands, rocky grasslands, and screes.

7. Scree deciduous montane woodlands of Prunus korshinskyi and Lonicera nummulariifolia – T19B9_LB1

Steep slopes of unstable limestone screes in the montane-Mediterranean belt are predominantly covered with sparse woodlands, primarily dominated by Prunus korshinskyi or Lonicera nummulariifolia (Figure 12). These formations, rarely exceeding 3 m in height, feature P. korshinskyi and L. nummulariifolia along with diagnostic herbaceous species well-adapted to screes and unstable rocky substrate, such as Galium incanum, Cephalaria stellipilis and Pseudoroegneria libanotica. Other taxa commonly found on rocky outcrops and scress were frequently observed, though not considered diagnostic, including the sub-shrub Atraphaxis billardierei and other forbs, such as Allium libani, Arrhenatherum kotschyi, Cousinia hermonis, Hypericum libanoticum, Nepeta glomerata and Scrophularia peyronii. Six endemic taxa, namely Allium libani, Cousinia hermonis, Erysimum verrucosum, Euphorbia erinacea, Hypericum libanoticum and Scutellaria utriculata were observed in these habitat types.

Figure 12. 

Montane deciduous woodlands on screes on the western slopes of Mount Hermon in Lebanon in 2020: A. with Prunus korshinskyi; B. with Lonicera nummulariifolia; C. Fruit of Prunus korshinskyi; D. Flowers of Lonicera nummulariifolia.

8. Montane woodlands of deciduous Prunus korshinskyi and evergreen Quercus coccifera – T19B9_LB1 X T213_LB2

These montane woodlands typically develop on stable rocky slopes, often at hilltops or on west-oriented slopes (Figure 13). They form sparse woodlands alternating with low garrigues of Phlomis. The diagnostic species include deciduous Prunus korshinskyi, Phlomis chrysophylla, Phleum montanum, and Euphorbia erinacea. While evergreen Quercus coccifera was a constant species, it was not considered diagnostic. In some areas with more intense grazing, the trees are reduced to low, compact and rounded shrubs, transforming the appearance of woodlands into garrigues. Q. coccifera is particularly susceptible to these effects. During the surveys in these woodlands, at around 1,600 m elevation, few isolated individuals of Juniperus excelsa were found. Apart from Quercus look and Euphorbia erinacea, two other endemic taxa were observed in this habitat type, namely Hypericum libanoticum and Scutellaria utriculata.

Figure 13. 

A. Montane woodlands of deciduous Prunus korshinskyi and evergreen Quercus coccifera on the western slopes of Mount Hermon in Lebanon in 2020; B. With garrigues of Phlomis chrysophylla.

9. Montane-Mediterranean thalweg deciduous thickets of Astragalus gummifer –S3575_LB3

This type comprises montane deciduous thickets in thalwegs formed by Astragalus gummifer, a spiny shrub that can reach up to 1 m height and occupy up to 90% of the surface, restricting the establishment of other species (Figure 14). In areas with scree substrate, Lonicera nummulariifolia woodlands were observed in the contact zone around the thickets. Other taxa were commonly found without being diagnostic, such as the herbaceous Eryngium billardierei, Erysimum verrucosum, Ferula biverticellata, Hordeum bulbosum, Marrubium globosum subsp. libanoticum, and Phlomis brevilabris. Five endemic taxa were observed in this habitat type, namely Centaurea iberica subsp. hermonis, Erysimum verrucosum, Galium libanoticum, Marrubium globosum subsp. libanoticum, and Phlomis brevilabris.

Figure 14. 

Astragalus gummifer thickets in thalwegs in the western slopes of Mount Hermon in Lebanon. A. View from a distance; B. Close-up.

10. Montane-Mediterranean deciduous Quercus look and Acer monspessulanum subsp. microphyllum woodlands – T1953_LB3

Quercus look, a thermophilous deciduous oak, forms small relictual clumps of trees or sparse woodlands in the montane-Mediterranean belt (Figure 15). The substrate is predominantly rocky, with soil poorly developed on limestone bedrock. Individuals of Q. look can reach up to 5 m height but are sometimes reduced to branched shrubs due to logging and overgrazing. Other diagnostic taxa were Rubia tenuifolia, Cephalaria stellipilis, Acer monspessulanum subsp. microphyllum, and Eryngium billardierei in decreasing diagnostic value. A. monspessulanum subsp. microphyllum is a tree often not exceeding 3 m height and, at times, reduced to low, compact, and rounded shrubs. The vine Rubia tenuifolia was prevalent in this habitat, observed growing without support, creeping at ground level, although it is typically widespread in Levantine woodlands at lower altitudes. Although not considered diagnostic, other taxa were relatively common in this habitat, such as the grass species Dactylis glomerata, Elymus panormitanus, and Hordeum bulbosum, as well as some forbs, such as Anthemis pauciloba, Bupleurum exaltatum, Erysimum verrucosum, Nepeta cilicica, N. italica, Salvia rubifolia, Scutellaria brevibracteata, and Silene libanotica. Ten endemic taxa were present in this habitat type, namely Astragalus nummularius subsp. trichopterus, Centaurea iberica subsp. hermonis, Erysimum verrucosum, Euphorbia erinacea, Galium libanoticum, Hypericum libanoticum, Q. look, Salvia rubifolia, Scutellaria utriculata, and Silene libanotica.

Figure 15. 

Montane woodlands of Quercus look and Acer monspessulanum subsp. microphyllum in the western slopes of Mount Hermon in Lebanon in 2020: A. B. Patches of woodlands; C. Leaves of Q. look; D. Immature acorns.

Habitat mapping

The mapping of habitat types of the western slopes of Mount Hermon encompassed areas characterised by their floristic composition (Figure 16), as well as agricultural fields, urbanised areas and planted forests.

Phryganas of Sarcopoterium spinosum (S7242) were sampled only once and grouped with supra-Mediterranean grasslands (cluster 4). However, they were mapped separately. These were confined to a small area in the foothills of Mount Hermon, typically occurring below 1,250 m of elevation. A single survey proved inadequate to fully characterise the floristic particularities of this habitat.

The last remaining population of Juniperus excelsa (T3D71) on the western slopes of Mount Hermon was located during fieldwork and consequently included in the mapping. Due to access restrictions, this habitat could not be physically reached and sampled.

Given the intricate and interwoven mosaics formed by related habitats sharing similar flora, the oro-Mediterranean habitats of hedgehog-heaths of Astragalus echinus and Noaea mucronata in thalweg and slopes, hedgehog-heaths of Tanacetum densum and Astragalus cruentiflorus, and cliffs of Rosularia sempervivum subsp. libanotica, were treated as a unified polygon. A similar approach was applied to the montane-Mediterranean woodlands of Prunus korshinskyi associated with Lonicera nummulariifolia or with Quercus coccifera in combination with Phlomis chrysophylla garrigues.

Patches of cultivated lands are dispersed at lower altitudes, encircling villages or situated on plateaus up to an elevation of 1,550 meters (Figure 17). These cultivated areas are mostly concentrated in thalwegs, where the soil is deeper and more level compared to the surrounding hilly, rocky slopes. This phenomenon accounts for the linear layout of numerous agricultural lands in the region. Rain-fed fields cultivating cereals (barley, wheat) and legumes (chickpeas, lentils) are prevalent. Terraced cultivation includes grapevines and fruit trees such as almonds, cherries and walnuts.

During the study, two planted forest sites were explored, although their flora was not surveyed. Comprising a mix of three conifers non-native to Mount Hermon, namely Cupressus sempervirens, Pinus brutia, and Cedrus libani, these forests were planted in the 1960s by the Ministry of Agriculture. The trees in these planted forests exceed 5 m height. Undoubtedly, these forests play a crucial ecological role in the area, serving as a refuge for numerous wild animals, including birds and mammals.

Finally, the surveys and the boundaries of the nature reserve were mapped to illustrate the range of the habitat types included within the 12.89 km² of protected area (Figure 18). The nature reserve of the western slopes of Mount Hermon comprehensively represents oro- and montane-Mediterranean habitats, as well as upper supra-Mediterranean woodlands and their degraded form of Crataegus azarolus thickets.

Figure 16. 

A. Contrast between the slope of phrygana on the left and the slope with evergreen oak woodlands in Wadi el-Feqaa; B. Close-up on phrygana of Sarcopoterium spinosum.

Figure 17. 

A. Agricultural lands in Berke el-Yebse ; B. Planted forest viewed from Rachaya.

Figure 18. 

Map of the habitat types identified on the western slopes of Mount Hermon in Lebanon.

Discussion

The floristic study conducted on the western slopes of Mount Hermon enabled the characterisation of ten habitat types, based on species differences and the influence of four significant environmental variables. Elevation and mean annual temperature emerged as the most influential factors shaping floristic composition. Although mean annual temperature and elevation exhibit a direct relationship, it is prudent to consider them separately due to the varied influences associated with them. Slope ranked as the third most significant environment variable, notably influencing floristic variation among the habitat groups of the oro-Mediterranean level. Northness also exerted a significant influence, particularly affecting the differentiation of the scree woodlands of Prunus korshinskyi from the other montane habitats.

Five habitat types represent novelties, constituting sub-types of broader habitat types described in the national typology (El Zein et al. 2022). They include the description of two types of hedgehog-heaths at the oro-Mediterreanean level. The group of Astragalus echinus and Noaea mucronata exhibits a preference for slopes and thalwegs where exposure to wind and sun is reduced, and the soils is deeper. The group of Tanacetum densum and Astragalus cruentiflorus thrives on hilltops and south-oriented slopes, showing a preference for xeric and rocky situations. Hedgehog-heaths are unique habitats and constitute relict vegetation poorly studied in the Levant, primarily composed of endemics or rare species belonging to the ancient Mediterranean Tertiary flora (Musarella et al. 2020). Both of these habitats are endemic to Mount Lebanon, Mount Hermon and Anti-Lebanon, indicated by the presence of taxa strictly endemic to these ranges. The first habitat includes Astragalus coluteoides, A. hermoneus, and A. nummularius subsp. trichopterus, while the second contains the diagnostic species A. cruentiflorus.

Another novelty is the description of a plant group diagnostic of the cliffs at the oro-Mediterranean level in western Mount Hermon. This group comprises twelve taxa, with Rosularia sempervivum subsp. libanotica having the highest diagnostic value, and the majority of them are endemic to the mountains of South Turkey, Mount Lebanon, Anti-Lebanon and Mount Hermon. We propose categorizing this habitat as a sub-type of “Mediterranean bare limestone inland cliffs” (U387) of the national habitat typology, specifically characterizing the oro-Mediterranean limestone cliffs of Mount Hermon in Lebanon. This contributes significantly to our understanding of plant communities on cliffs, as this habitat has remained poorly studied in the Levant, and oro-Mediterranean cliffs of this region were previously undocumented (Davis 1951; Wagensommer 2017).

The scree deciduous montane woodlands of Prunus korshinskyi and Lonicera nummulariifolia represent a habitat type that has been insufficiently characterised in the past. This unique habitat is distinguished by its specific substrate, notably the screes, and the distinctive aspect of the slope. The formations of P. korshinskyi show affinities with the Arabic Peninsula (Zohary 1973), while the presence of L. nummulariifolia characterises typical Eastern Mediterranean and Irano-Turanian habitats (Davis et al. 1965). At the Lebanese level, woodlands featuring P. korshinskyi are a rarity and are exclusively found on the eastern slopes of Mount Lebanon between Wadi Faara, Ainata and Yammoune in a limited area (Mouterde 1966, 1970). We propose this habitat as a sub-type of “Lebanese wild fruit trees steppe wood” of the national typology (T19B9_LB1). The second montane woodland including P. korshinskyi in association with Quercus coccifera constitutes a second sub-habitat.

Quercus look is a tree endemic to Mount Lebanon, Anti-Lebanon and Mount Hermon (Avishai 2017). The inhabitants of the region of Rachaya historically distinguished this oak species from others, assigning it the name “Look” in Arabic which later became its specific epithet (Avishai 2017). Deciduous montane woodlands of Q. look associated with Acer monspessulanum subsp. microphyllum constitute a unique habitat endemic to the region, specifically occurring in the region of Mount Hermon. Similar Q. look woodlands were also identified at the montane-Mediterranean level on the heights of Mount Barouk in Mount Lebanon (Chouchani et al. 1975; Abi-Saleh and Safi 1988). In this area, this tree species was known to form pure stands within the ecological zone of Cedrus libani, sometimes co-dominating with the latter. Nevertheless, these woodlands lacked precise documentation in earlier studies, and this research significantly contributes by providing additional insights into their floral composition. These contributions acknowledge the distinctiveness and rarity of the Q. look woodlands.

The floristic complexity of the supra-Mediterranean woodlands was highlighted, revealing two main types distinguished by elevation and the prevalence of either Quercus coccifera or Q. infectoria. However, the floristic composition did not effectively differentiate between degraded forms of these habitats. For instance, the thickets of Crataegus azarolus, representing the degraded form of the upper supra-Mediterranean woodlands of Q. infectoria, and the gradual transitions were all grouped together. This emphasizes the crucial role that physiognomy of the vegetation plays in interpreting results.

This brings to light an important parameter affecting the habitats of the western slopes of Mount Hermon, namely the impact of the historical anthropic activities that have gradually modified the physiognomy of the vegetation and thus the landscape of the area. The current landscapes of Mount Hermon are made of relict habitats compared to what they used to be (Abel 1933; Vaumas 1954; Mouterde 1966; Mikesell 1969). For example, the deciduous woodlands of Quercus look have almost disappeared and only scattered patches remain in Mount Hermon. The current pastoral activities are not allowing a decent regeneration as the seedlings are systematically grazed by goats.

Similar to many areas in the western slopes of Anti-Lebanon range, the woodlands of Juniperus excelsa have disappeared due to intense exploitation for firewood (Mouterde 1966). This was the only conifer species observed in our surveys and it was reported to have a wider distribution in the past. Usually, montane-Mediterranean belts are dominated by coniferous formations (Quézel and Barbero 1982) but it is not the case anymore in Mount Hermon. These woodlands have become rare due to past wood exploitation. The presence of J. excelsa woodlands on the western slopes of Anti-Lebanon, between 1,700 and 2,800 m a.s.l. (Vaumas 1954; Mouterde 1966), indicates that the relict woodlands present on the Lebanese slopes of Mount Hermon could have once extended over a larger area. Historical documents described Mount Hermon with significant forest cover during the antiquity (Abel 1933; Vaumas 1954; Mikesell 1969). Nevertheless, until now it still constitutes the southernmost edge of the distribution range of this conifer (Douaihy et al. 2011, 2013; Caudullo et al. 2017).

Human activities have significantly influenced the diversity at the landscape level, resulting in a characteristic mosaic of vegetation types. This mosaic consists of an alternating succession of grasslands in linear thalwegs with Q. infectoria interspersed among rocky hills dominated by Q. coccifera. Below 1,500 meters a.s.l., various habitat types bear traces of ancient agriculture. Old terraces, constructed with rocks, are prevalent remnants of agriculture. Often reclaimed by shrubs or trees, these terraces comprise a mix of rocky micro-habitats and grasslands. Another common trace of past agriculture is the presence of rock heaps, remnants of stones cleared for ploughing, mainly located in thalwegs.

The woodlands on Mount Hermon were previously characterised as similar to those on Mount Lebanon, distinguishing them from those of Anti-Lebanon (Abi-Saleh 1982; Abi-Saleh and Safi 1988). However, some typical Mediterranean taxa were absent in Mount Hermon. For instance, in the supra-Mediterranean belt, Acer obtusifolium, Arbutus andrachne, Cercis siliquastrum, Laurus nobilis, Phillyrea latifolia, and Pinus brutia were absent, and similarly, in the montane-Mediterranean woodlands, Acer hyrcanum subsp. tauricola, Fraxinus ornus, and Ostrya carpinifolia were also missing. Intriguingly, additional elements have offered insights into floristic affinities between the eastern slopes of Mount Barouk’s and the western slopes Mount Hermon. These indications include the absence of certain Mediterranean tree species that are typical of western Mount Lebanon, and the shared presence of the same narrow endemic taxa, namely Allium feinbergii, Bellevalia hermonis, Quercus look, and Salvia rubifolia (Mouterde 1966, 1970, 1984).

Conclusion

This study provides the first comprehensive list of plant taxa of the western slopes of Mount Hermon in Lebanon utilizing a habitat-based approach. The significance of elevation, mean annual temperature, slope and northness in shaping the distribution of taxa and habitat types has been demonstrated. The impact of anthropic activities is also an intriguing aspect that should be more frequently quantified and included in studies. However, understanding historical impacts without proper documentation and verified references can be challenging, and the landscapes stand as the last remaining means to decipher the history of the region.

Floristic affinities with Mount Barouk, located in the south part of Mount Lebanon, were highlighted. The analysis of floristic composition allowed the identification and characterisation of ten habitat types: three at the oro-Mediterranean level, hedgehog-heaths of Astragalus echinus and Noaea mucronata in thalweg and slopes, hedgehog-heaths of Tanacetum densum and Astragalus cruentiflorus, cliffs of Rosularia sempervivum subsp. libanotica, three at the supra-Mediterranean level, namely grasslands with Eryngium glomeratum, woodlands of Quercus infectoria, Q. coccifera and Crataegus azarolus, woodlands of evergreen Q. coccifera, and four at the montane level, namely scree deciduous woodlands of Prunus korshinskyi and Lonicera nummulariifolia, woodlands of deciduous P. korshinskyi and evergreen Q. coccifera, shrublands of Astragalus gummifer, deciduous woodlands of Quercus look and Acer monspessulanum subsp. microphyllum. Five novelties can be added to the national typology of habitat as sub-types.

Our research yielded 383 taxa including twenty-four endemics to Mount Hermon and Mount Lebanon, two taxa endemic to Mount Hermon and Anti-Lebanon, and one exclusively endemic to Mount Hermon. It provides a complementary list of taxa for the region. The oro-Mediterranean hedgehog-heath of Astragalus echinus and Noaea mucronata hosted the highest number of endemic plant taxa (12), followed by the montane woodlands of Quercus look and Acer monspessulanum subsp. microphyllum (10), the scree deciduous montane woodlands of Prunus korshinskyi and Lonicera nummulariifolia (6), and by montane thalweg deciduous thickets of Astragalus gummifer (5). The vulnerability of specific habitat types, particularly Q. look woodlands and Juniperus excelsa woodlands, as a result of past exploitations, should be considered. This highlights the need for protective measures aimed at conserving these woodlands and enhancing their restoration. Such initiatives could offer dual benefits by preserving the environment and also supporting pastoralist activities that require shaded areas during hot summer days.

Finally, the habitat mapping serves as a valuable tool for visualizing the distribution of habitats and surrounding human activities. It constitutes a crucial resource for the effective management and conservation of the natural heritage. The comprehensive map reveals that the nature reserve on the western slopes of Mount Hermon encompasses the majority of the identified habitats. The insights gained from this study are instrumental in informing the development of a management plan for the nature reserve, ensuring the preservation of its diversity.

Data availability

The data related to the geographic coordinates and elevation, list of observed plant species, and percentage frequency and diagnostic value for each species within each plant community are provided as Suppl. materials 14. The entire data are available upon request to the corresponding author.

Author contributions

HE, LC, SB and CK conceptualized the study, carried out the investigation and designed the visualisation of the results. HE conducted the fieldwork and data collection, curated the data, designed the methodology, carried out the analyses, wrote the original draft and took the photographies. LC, SB, CK, DC and MM supervised the study, reviewed and edited the manuscript, were responsible for acquiring funding and were part of the project administration.

Acknowledgements

The authors are grateful to the Observatoire Libano-Francais de l’Environnement (O-LIFE, presently NeoLife) for supporting this work (contribution number SA 15-2022). The authors would also like to acknowledge the “STEPping up Nature Reserves Capacity – STEP4Nature” project funded by the Ministry of Foreign Affairs and International Cooperation of the Italian Republic through the Italian Agency for Development Cooperation (AICS) and implemented by the United Nations Development Programme (UNDP) in coordination with the Ministry of Environment of Lebanon for contributing to field surveys. The authors are thankful to Mehdi Fayeq for accompanying the project team in the field and sharing invaluable information about the area. The authors would like to express their gratitude to Zdeňka Lososová, Idoia Biurrun and the two anonymous reviewers for their valuable remarks, which have contributed to enhancing the quality of this paper.

References

  • Arnold N, Baydoun S, Chalak L, Raus T (2015) A contribution to the flora and ethnobotanical knowledge of Mount Hermon, Lebanon. Flora Mediterranea 25: 13–55. https://doi.org/10.7320/FlMedit25.013
  • Avishai M (2017) Quercus look Kotschy: a distinct Mt. Hermon species. International Oaks 28: 73–88.
  • Ayalon A, Bar-Matthews M, Frumkin A, Matthews A (2013) Last Glacial warm events on Mount Hermon: the southern extension of the Alpine karst range of the east Mediterranean. Quaternary Science Reviews 59: 43–56. https://doi.org/10.1016/j.quascirev.2012.10.047
  • Barghachoun A-R (1976) Etudes écologiques, phytogéographiques et phytosociologiques, des forêts de chênes caduques du Liban. PhD thesis. Université d’Aix-Marseille, Marseille, FR.
  • Baydoun S, Arnold NA (2017) A glance at the wild flora of Mount Hermon Lebanon. Research Center for Environment and Development, Beirut Arab University, Beirut, LB, 151 pp.
  • Belbin L, McDonald C (1993) Comparing three classification strategies for use in ecology. Journal of Vegetation Science 4: 341–348. https://doi.org/10.2307/3235592
  • Bou Dagher-Kharrat M, El Zein H, Rouhan G (2018) Setting conservation priorities for Lebanese flora—Identification of important plant areas. Journal for Nature Conservation 43: 85–94. https://doi.org/10.1016/j.jnc.2017.11.004
  • Cabido M, Zeballos SR, Zak M, Carranza ML, Giorgis MA, Cantero JJ, Acosta ATR (2018) Native woody vegetation in central Argentina: Classification of Chaco and Espinal forests. Applied Vegetation Science 21: 298–311. https://doi.org/10.1111/avsc.12369
  • Cañadas EM, Fenu G, Peñas J, Lorite J, Mattana E, Bacchetta G (2014) Hotspots within hotspots: Endemic plant richness, environmental drivers, and implications for conservation. Biological Conservation 170: 282–291. https://doi.org/10.1016/j.biocon.2013.12.007
  • Cheng Z, Aakala T, Larjavaara M (2023) Elevation, aspect, and slope influence woody vegetation structure and composition but not species richness in a human-influenced landscape in northwestern Yunnan, China. Frontiers in Forests and Global Change 6: e1187724. https://doi.org/10.3389/ffgc.2023.1187724
  • Clarke KR, Warwick RM (1994) Similarity-based testing for community pattern: the two-way layout with no replication. Marine Biology 118: 167–176. https://doi.org/10.1007/BF00699231
  • Csárdi G, Hester J, Wickham H, Chang W, Morgan M, Tenenbaum D, Posit Software PBC (2021) remotes: R Package Installation from Remote Repositories, Including “GitHub.”. Version 2.5.0. https://CRAN.R-project.org/package=remotes [accessed 7 May 2023]
  • Dar S (1993) Settlements and cult sites on Mount Hermon, Israel: Ituraean Culture in the Hellenistic and Roman Periods. [British Archaeological Reports 589]. Oxford University Press, Oxford, UK, 380 pp.
  • Davies CE, Moss D, Hill MO (2004) EUNIS Habitats Classification. European Environment Agency, Copenhagen, DK.
  • Davis PH, Cullen J, Coode MJE (1965) Flora of Turkey and the East Aegean Islands. Edinburgh University Press, Edinburgh, UK, 5836 pp.
  • Douaihy B, Vendramin GG, Boratyński A, Machon N, Bou Dagher-Kharrat M (2011) High genetic diversity with moderate differentiation in Juniperus excelsa from Lebanon and the eastern Mediterranean region. AoB PLANTS 2011: plr003. https://doi.org/10.1093/aobpla/plr003
  • Douaihy CB, Restoux G, Machon N, Bou Dagher-Kharrat M (2013) Ecological characterization of the Juniperus excelsa stands in Lebanon. Ecologia mediterranea 39: 169–180. https://doi.org/10.3406/ecmed.2013.1303
  • Dubertret L (1955) Carte géologique du Liban. Ministère des Travaux Publics de la République Libanaise, Beirut, LB.
  • El Zein H, Kahale R (2022) First comprehensive IUCN Red List assessment of 100 endemic species of the flora of Lebanon. Flora Mediterranea 32: 327–338. https://doi.org/10.7320/FlMedit32.327
  • El Zein H, Itani M, Talhouk S, Bou Dagher-Kharrat M, Yazbek M, Al-Zein SM (2018) Key biodiversity areas for plants in Lebanon. In: Valderrábano M, Gil T, Heywood V, de Montmollin B (Eds) Conserving wild plants in the south and east Mediterranean region. IUCN, Gland, CH and Málaga, ES, 78–84.
  • El Zein H, Stephan J, Khater C, S. Al-Zein M, Bou Dagher-Kharrat M (2022) Aligning terrestrial habitat typology of Lebanon with EUNIS habitat classification. Phytocoenologia 51: 233–244. https://doi.org/10.1127/phyto/2022/0390
  • El Zein H, Chalak L, Baydoun S, Plouhinec K (2023) Newly recorded for the vascular flora of Lebanon: Ferula biverticellata J. Thiébaut (Apiaceae). Check List 19: 293–299. https://doi.org/10.15560/19.3.293
  • Farr TG, Rosen PA, Caro E, Crippen R, Duren R, Hensley S, Kobrick M, Paller M, Rodriguez E, … Alsdorf D (2007) The Shuttle Radar Topography Mission. Reviews of Geophysics 45. https://doi.org/10.1029/2005RG000183
  • Farra Haddad N (2021) Mount Hermon (Jabal El Sheikh) in Lebanon, A Sacred Biblical Mountain: Pilgrimages, traditions and rituals. International Journal of Religious Tourism and Pilgrimage 9: e4. https://doi.org/10.21427/HNYV-8F38
  • Gillison AN (2001) Vegetation survey and habitat assessment of the Tesso Nilo Forest Complex. WWF, Queensland, AU, 76 pp.
  • Hao M, Zhao W, Qin L, Mao P, Qiu X, Xu L, Xiong YJ, Ran Y, Qiu G (2021) A methodology to determine the optimal quadrat size for desert vegetation surveying based on unmanned aerial vehicle (UAV) RGB photography. International Journal of Remote Sensing 42: 84–105. https://doi.org/10.1080/01431161.2020.1800123
  • Horikoshi M, Tang Y, Dickey A, Grenié M, Thompson R, Selzer L, Strbenac D, Voronin K, Pulatov D (2023) ggfortify: Data Visualization Tools for Statistical Analysis Results. Version 0.4.16. https://CRAN.R-project.org/package=ggfortify [accessed 7 May 2023]
  • Hunter JT, Hunter VH (2020) Montane mire vegetation of the New England Tablelands Bioregion of Eastern Australia. Vegetation Classification and Survey 1: 37–51. https://doi.org/10.3897/VCS/2020/48765
  • IPNI (2023) International Plant Names Index. International Plant Names Index. https://www.ipni.org [accessed 25 Jan 2023]
  • Joly D, Gillet F (2017) Interpolation of temperatures under forest cover on a regional scale in the French Jura Mountains. International Journal of Climatology 37: 659–670. https://doi.org/10.1002/joc.5029
  • Karger DN, Conrad O, Böhner J, Kawohl T, Kreft H, Soria-Auza RW, Zimmermann NE, Linder HP, Kessler M (2017) Climatologies at high resolution for the earth’s land surface areas. Scientific Data 4: e170122. https://doi.org/10.1038/sdata.2017.122
  • Krupp F, Khalaf M, Malek M, Streit B, Al-Jumaily M (2009) The Middle Eastern Biodiversity Network: Generating and sharing knowledge for ecosystem management and conservation. In: Neubert E, Zuhair A, Stefano T, Burçin AG (Eds) Animal Biodiversity in the Middle East. Proceedings of the First Middle Eastern Biodiversity Congress. Zoo Keys, Aqaba, JO, 3–15. https://doi.org/10.3897/zookeys.31.371 [accessed 7 May 2023]
  • Legendre P, Legendre L (2012) Numerical Ecology. 3rd ed. Elsevier, Amsterdam, NL, 1008 pp.
  • Médail F, Quézel P (1997) Hot-Spots analysis for conservation of plant biodiversity in the Mediterranean Basin. Annals of the Missouri Botanical Garden 84: 112–127. https://doi.org/10.2307/2399957
  • Montenegro-Hoyos A, Vega N, Linares-Palomino R (2022) Plant diversity and structure in desert communities of the Andean piedmont in Ica, Peru. Vegetation Classification and Survey 3: 53–66. https://doi.org/10.3897/VCS.68006
  • Morris C, Badik KJ, Morris LR, Weltz MA (2016) Integrating precipitation, grazing, past effects and interactions in long-term vegetation change. Journal of Arid Environments 124: 111–117. https://doi.org/10.1016/j.jaridenv.2015.08.005
  • Mouterde P (1966) Nouvelle flore du Liban et de la Syrie. Vol. 1. Dar El-Machreq, Beirut, LB, 642 pp.
  • Mouterde P (1970) Nouvelle flore du Liban et de la Syrie. Vol. 2. Dar El-Machreq, Beirut, LB, 720 pp.
  • Mouterde P (1984) Nouvelle flore du Liban et de la Syrie. Vol. 3. Dar El-Machreq, Beirut, LB, 579 pp.
  • Musarella CM, Brullo S, del Galdo GG (2020) Contribution to the orophilous cushion-like vegetation of Central-Southern and Insular Greece. Plants 9: e1678. https://doi.org/10.3390/plants9121678
  • Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB, Kent J (2000) Biodiversity hotspots for conservation priorities. Nature 403: 853–858. https://doi.org/10.1038/35002501
  • Oksanen J, Simpson GL, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’Hara RB, Solymos P, Stevens MHH, … Weedon J (2022) vegan: Community Ecology Package. R Package version 2.3-2. http://cran.r-project.org/package=vegan [accessed 7 May 2023]
  • Post EG, Dinsmore JE (1932) Flora of Syria, Palestine, and Sinai. 2nd ed. American University Press, Beirut, LB, 920 pp.
  • QGIS Development Team (2023) QGIS Geographic Information System. http://qgis.osgeo.org/ [accessed 25 Jan 2023]
  • Sarkar D, Andrews F, Wright K, Klepeis N, Larsson J, Wen Z, Murrell P, Eng S, Zeileis A, Courtiol A (2023) lattice: Trellis Graphics for R. Version 0.22-6. https://CRAN.R-project.org/package=lattice [accessed 7 May 2023]
  • Sokal RR, Michener CD (1958) A statistical method for evaluating systematic relationships. University of Kansas Science Bulletin 28: 1409–1438.
  • Somerfield PJ (2008) Identification of the Bray-Curtis similarity index: Comment on Yoshioka (2008). Marine Ecology Progress Series 372: 303–306. https://doi.org/10.3354/meps07841
  • Stephan J, Barhoun J, Maatouk E, Foucault B, Safi S (2019) Syntaxonomy of open oak forests on basalt formations in north Lebanon. Lebanese Science Journal 20: 391–412. https://doi.org/10.22453/LSJ-020.3.391-412
  • Takhtajan A (1986) Floristic Regions of the World. University of California Press, Berkeley, CA, US, 522 pp.
  • van der Maarel E (1979) Transformation of cover-abundance values in phytosociology and its effects on community similarity. Vegetatio 39: 97–114. https://doi.org/10.1007/BF00052021
  • Vaumas E de (1954) Le Liban (Montagne libanaise, Bekaa, Anti-Liban, Hermon, Haute Galilée libanaise). Etude de géographie physique. Firmin-Didot, Paris, FR, 367 pp.
  • Wagensommer RP (2017) Phytosociological investigation on the thermo-chasmophilous vegetation of the Eastern Mediterranean territories. PhD thesis. Università di Catania, Catania, IT. http://dspace.unict.it:8080/handle/10761/3718 [accessed 23 Feb 2024]
  • Wickham H, François R, Henry L, Müller K, Vaughan D, Posit Software PBC (2023b) dplyr: A Grammar of Data Manipulation. Version 1.1.4. https://CRAN.R-project.org/package=dplyr [accessed 7 May 2023]
  • Wickham H, Chang W, Henry L, Pedersen TL, Takahashi K, Wilke C, Woo K, Yutani H, Dunnington D, … Posit PBC (2023c) ggplot2: Create Elegant Data Visualisations Using the Grammar of Graphics. Version 3.5.0. https://CRAN.R-project.org/package=ggplot2 [accessed 7 May 2023]
  • YMCA Lebanon (2005) Enviromental impact assessment report wastewater treatment plant in Rashaya - Rashaya Caza Lebanon. YMCA Lebanon, Beirut, LB, 95 pp.
  • Zeballos SR, Acosta ATR, Agüero WD, Ahumada RJ, Almirón MG, Argibay DS, Arroyo DN, Blanco LJ, Biurrun FN, … Cabido MR (2023) Vegetation types of the Arid Chaco in Central-Western Argentina. Vegetation Classification and Survey 4: 167–188. https://doi.org/10.3897/VCS.100532
  • Zohary M (1973) Geobotanical Foundations of the Middle East. Vol. 1. Gustav Fischer Verlag, Amsterdam, NL, 340 pp.

E-mail and ORCID

Supplementary materials

Supplementary material 1 

Geographic coordinates, elevation and cluster group of the plots (xlsx).

Download file (17.34 kb)
Supplementary material 2 

List of the species observed (pdf).

Download file (253.96 kb)
Supplementary material 3 

Full synoptic table with percentage frequency and diagnostic value for each species (pdf).

Download file (858.24 kb)
Supplementary material 4 

Full synoptic table with percentage frequency and diagnostic value for each species (*.xlsx).

Download file (50.96 kb)
login to comment