The flora of the Ru'us al-Jibal--the mountains of the Musandam Peninsula: an annotated checklist and selected observations.
Ghazanfar (1999) identified "the northern mountains of Oman (including the Musandam mountains) [as] one of the three local centres of endemism in the country" and her Figure 3.3 depicts the high Musandam as an "LCE" separate from the Hajar Mountains generally. On current evidence, however, the "Musandam mountains" (the Ru'us al-Jibal) should probably be excluded from the most rigorous interpretation of that statement. Only four UAE/Oman endemics are found in the Ru'us alJibal and no species has yet been identified that is strictly endemic to the Ru'us al-Jibal.
For the northern Oman mountains as a whole, Ghazanfar (1998b, 1999, 2003) has estimated that approximately 25 plant species are nationally or regionally endemic. That figure would include the mountain regions of the UAE, which has no nationally endemic species. Most of those endemics, according to Ghazanfar, are uncommon and restricted to one or two locations in the mountains, although she includes only five northern mountain endemics in her Red List of the Flora of Oman (Ghazanfar 1999, Table 3.3). She also lists six "common species" from among those endemics, of which only one is found in the Ru'us al-Jibal--Pteropyrum scoparium, which is however acknowledged to be possibly conspecific with P. aucheri of Iran, Afghanistan and Pakistan (Miller & Cope 1996; Ghazanfar 2003). Only four other northern Oman endemics are found within the Ru'us al-Jibal: Desmidorchis arabica, a cactus-like milkweed that is widespread but not common (and is apparently included in Ghazanfar's Red List as Caralluma aucheri); Echinops erinaceus, a spiny thistle, locally common on scree and rubble; Pulicaria edmondsonii, a sloped-welling dwarf shrub that is widespread and locally common; and Stipa mandavillei, a grass found at high elevations that is evidently rare outside the Jebel Akhdar.
Nevertheless, local endemism is high in the Ru'us al-Jibal in the sense that some 75 species, representing approximately 22% of the above-mentioned total of 338 for the Ru'us al-Jibal, are confined, within Eastern Arabia, either exclusively (62 spp.) or very nearly so (13 spp.) to the Ru'us al-Jibal, being absent (or very nearly so) in the Hajar Mountains to the south. A complete list of these locally endemic species is given in Table 3 (see Checklist for additional details).
Among the most common local endemics are a diverse array of distinctive perennials including Artemisia sieberi, Astragalus fasciculifolius, Centaurea wendelboi, Dianthus crinitus, Gladiolus italicus, Ixiolirion tataricum, Jurenia berardioides, Lactuca orientalis, Leopoldia longipes, Moraea sisyrinchium and Prunus arabica, along with a few annuals such as Helianthemum salicifolium and Hippocrepis unisiliquosa. All of the foregoing are higher elevation species. In addition, Convolvulus acanthocladus, one of the three most abundant dwarf shrubs at high elevations in the Ru'us al-Jibal, is a "near" local endemic, having only a very limited distribution in the mountains to the south.
Approximately 80% of the Table 3 species can be found at medium or high elevations (above 700 m) and approximately 75% are found only at those elevations. These numbers emphasise the contribution of higher elevation species to both the diversity and distinctiveness of the flora of the Ru'us al-Jibal.
Most of the local endemics (like most of the other high elevaton species of the Ru'us al-Jibal) are species that have their principal ranges in more temperate regions, particularly Iran and neighbouring areas of Central Asia (see also Section 5.2 below), and it should be emphasised that, although the plants listed in Table 3 are limited in Eastern Arabia to the mountains of the Ru'us al-Jibal, they are not necessarily restricted to mountain environments elsewhere. A number of the more common Table 3 species (as well as other common species found only in mountain environments in the UAE) are also common or locally common at low elevations on the sand and gravel plains of Qatar, northern Saudi Arabia and Kuwait, where winter temperatures are lower and winter rainfall is more reliable than in the UAE (Mandaville 1990; G.M. Brown, pers. comm.). Such species include many annuals, e.g., Althaea ludwigii, Anagallis arvensis, Arnebia decumbens, Calendula arvensis, Campanula erinus, Helianthemum salicifolium, Plantago amplexicaulis, Reichardia tingitana, Rumex vesicarius, Scabiosa olivieri and Trigonella stellata, as well as a variety of perennials such as Artemisia sieberi, Farsetia aegyptia, Helianthemum lippii, Ixiolirion tataricum, Moraea sisyrhinchium, Piptatherum holciforme and Prunus arabica (see, e.g., Shuaib 1995).
[FIGURE 3.1.6 OMITTED]
A number of the high elevation local endemics found in the Ru'us al-Jibal are ruderal species also discussed separately in Section 7 below. Some of these have probably been introduced initially with agricultural activities.
The presence of some very rare local endemics may be attributable to chance establishment and/or preservation. In the case of the trees Pistacia khinjuk and Cordia sp. aff. quercifolia, for example, both species are represented by fewer than two dozen specimens located along limited stretches of single wadis. Both are also restricted to cliff or ledge sites protected from browsing.
[FIGURE 3.1.7 OMITTED]
[FIGURE 3.1.8 OMITTED]
[FIGURE 3.1.9 OMITTED]
[FIGURE 3.1.10 OMITTED]
[FIGURE 3.1.11 OMITTED]
[FIGURE 3.1.12 OMITTED]
[FIGURE 3.1.13 OMITTED]
[FIGURE 3.1.14 OMITTED]
[FIGURE 3.1.15 OMITTED]
[FIGURE 3.1.16 OMITTED]
[FIGURE 3.1.17 OMITTED]
[FIGURE 3.1.18 OMITTED]
[FIGURE 3.1.19 OMITTED]
[FIGURE 3.1.20 OMITTED]
[FIGURE 3.1.21 OMITTED]
[FIGURE 3.1.22 OMITTED]
[FIGURE 3.1.23 OMITTED]
[FIGURE 3.1.24 OMITTED]
[FIGURE 3.1.25 OMITTED]
[FIGURE 3.1.26 OMITTED]
[FIGURE 3.2.1 OMITTED]
[FIGURE 3.2.2 OMITTED]
[FIGURE 3.2.3 OMITTED]
[FIGURE 3.2.4 OMITTED]
[FIGURE 3.2.5 OMITTED]
[FIGURE 3.2.6 OMITTED]
3. Vegetation zones.
The vegetation of the Ru'us al-Jibal can be represented by three broad zones characterised by habitat and elevation: Zone 1: a mountain wadi zone (c.100-600 m); Zone 2: a low and medium elevation montane zone (c.200-1100 m); and Zone 3: a high elevation montane zone (above c.1100 m). These zones are described in more detail below (see also Table 7).
Zone 1: Mountain wadi zone. This zone includes wadi beds, wadi banks and associated gravel fans and terraces. In the Ru'us al-Jibal, this environment is common at elevations from c.100-600 m. At higher elevations, wadis tend to be narrower and rockier and the distinction between wadi vegetation and slope vegetation (Zone 2) is less evident.
The most conspicuous species are the legume Tephrosia apollinea and the trees Acacia tortilis, Ficus cordata salicifolia and Ziziphus spina-christi. Locally common small shrubs include Gaillonia aucheri, Pulicaria nobilis and Ochradenus aucheri. Other commonly occurring species include many annuals, e.g., Anagallis arvensis, Erodium spp., Erucaria hispanica, Euphorbia granulata, Geranium mascatense, Parietaria alsinifolia, Rumex vesicarius, Scabiosa olivieri and Sisymbrium erysimoides.
Zone 2: Low and medium elevation montane zone. This zone encompasses all terrain at elevations from c.100-1100 m, other than the mountain wadi zone (Zone 1), and therefore includes slopes, cliffs, plateaux and basins, upper wadis and gulleys and terraced fields.
The predominant species are Acacia tortilis, Euphorbia larica and Cymbopogon spp. Other characteristic species include the perennials Astragalus fasciculifolius, Gaillonia aucheri, Gymnocarpos decandrus, Launaea bornmuelleri, Lavandula subnuda, Leucas inflata, Moringa peregrina, Periploca aphylla, Pulicaria edmondsonii and Vernonia arabica and the annuals Hippocrepis unisiliquosa, Pentanema divaricatum, Plantago spp., Reichardia tingitana, Stipa capensis, Trigonella stellata, Viola cinerea and Zoegea purpurea.
The upper boundary for this zone is a natural one, chosen on the basis of the disappearance of E. larica and associates such as Gaillonia aucheri, Lavandula subnuda, Pulicaria edmondsonii and Vernonia arabica, coupled with the first significant appearances of dwarf shrub species more characteristic of the high elevation zone, such as Artemisia sieberi, Centaurea wendelboi and Ephedra pachyclada, as well as Helianthemum salicifolium and other high elevation annuals.
The upper half of this zone, from c.600-1100 m, is nevertheless transitional and could be further distinguished on that basis. Within it, some species characteristic of the zone typically disappear at elevations lower than 1100 m, whereas other species more characteristic of the high elevation zone (Zone 3) first appear at somewhat lower elevations. For example:
--The Arabian almond Prunus arabica first appears and effectively replaces the morphologically similar (but ecologically distinct) Moringa peregrina above c.600 m.
--Ficus cordata salicifolia gives way to F. johannis very reliably at c.500 m, except in a few sheltered gorges.
--The dwarf shrubs Astragalus fasciculifolius, Fagonia schimperi, Farsetia aegyptia, Gymnocarpos decandrus, Helianthemum lippii, Lactuca orientalis, Launaea bornmuelleri, Periploca aphylla and Teucrium stocksianum first appear at c.400-700 m and can be locally common within the upper part of the low and medium elevation montane zone as well as in the high elevation zone.
--Convolvulus acanthocladus and the less common C. ulicinus, both characteristic of high elevations, make their first appearances at c.700-800 m (C. acanthocladus beginning somewhat lower in the north).
--Acacia tortilis is uncommon above c.800 m, except in association with cultivation.
In this context it should be noted that the physiography of the Ru'us al-Jibal is such that the lower half of Zone 2 often consists of very steep and rocky terrain, with gentler slopes predominating only above 500-600 m or more (see Fig. 1.4.2). This may have the effect of reducing somewhat both the number and diversity of slope species observed below c.500-600 m.
Zone 3: High elevation zone. This zone encompasses all terrain from c.1100 m to the summit plateaux and peaks at 1500-2000 m. The most common and characteristic species in this zone are Convolvulus acanthocladus, Artemisia sieberi and Ephedra pachyclada (all dwarf shrubs), Cymbopogon jwarancusa (a perennial tussock grass), the larger and more conspicuous almond tree Prunus arabica (stunted where it grows exposed at higher elevations) and the erect shrub Dodonaea viscosa. This association has been called "Artemisia steppe" (Mandaville 1985; Ghazanfar 1999, 2003), following the usage of Zohary (1963, 1973), who found associations of the same species and/or genera to be characteristic of many areas of the plateau of central Iran.
Also common or locally common at high elevations are the perennials Astragalus fasciculifolius, Centaurea wendelboi, Dianthus crinitus, Diplotaxis harra, Fagonia schimperi, Farsetia aegyptia, Gymnocarpos decandrus, Helianthemum lippii, Jurenia berardioides, Launaea bornmuelleri, Phagnalon schweinfurthii and Teucrium stocksianum. High elevation species seen most often in agricultural or peri-agricultural contexts include Gladiolus italicus, Ixiolirion tataricum, Moraea sisyrhinchium, Papaver sp. and Roemeria sp. (see also Section 7 below). Notably absent at high elevations is Euphorbia larica, one of the most common and characteristic shrubs of slopes and plateaux at lower elevations, which declines in abundance above 1000 m and is absent by 1200 m.
A number of species, particularly annuals, can be found at all elevations in the Ru'us al-Jibal, as recorded in the Checklist. Common perennials in that category include the lily Asphodelus tenuifolius and the fern Onychium divaricatum. The following annuals can regularly be seen at higher elevations: Anagallis arvensis, Anthemis odontostephana, Anticharis arabicus, Asteriscus hierochunticus, Erucaria hispanica, Filago desertorum, Galium decaisnei, Geranium spp., Helianthemum salicifolium, Salvia aegyptiaca, Senecio glacus, Scabiosa olivieri, Thymelaea mesopotamica, Trigonella stellata, Viola cinerea and Zoegea purpurea.
4. Recent environmental stresses: overgrazing and drought
It is important to emphasise that the Ru'us al-Jibal, as wild and remote as it may at first seem, is not a pristine wilderness. Rather, it has been the site of seasonal habitation and localised but widespread terraced agriculture and probably livestock (goats and sheep) herding for at least several hundred years (see box, this page). In recent decades in particular, the vegetation of the Ru'us al-Jibal has been subject to heavy grazing pressure by feral and domestic goats, a phenomenon which has been accelerated by improved access and economic well-being, allowing resident farmers to support larger herds through the use of "imported" food and water when necessary.
This applies not least to the very highest area of the Ru'us al-Jibal, where the main graded road ascends from the extensive cultivation at As-Sayh (a silt plain formed behind a landslide dam at c.1150 m) (Fig. 1.6.6) to the stony Sahasa plateau at 1450 m (Figs. 1.3.4 and 3.1.9), which lies between the two highest summits, Jebel Harim (c.2008 m) and Jebel Jais (c.1935 m). Herds totalling c.100 goats have been observed near the summit of Jebel Jais and sheep are kept at, and graze from, the highest settlement in the Ru'us al-Jibal, at 1600 m, a 3% hour walk from the road.
Therefore, in contrast with Mandaville's (1977) assessment for the high Jebel Akhdar, the vegetation in many areas at high elevation in the Ru'us al-Jibal is significantly affected by grazing (see Figs. 3.1.1 and 3.1.2). Munton visited the Ru'us al-Jibal in the late 1970s to investigate the presence of Arabian tahr, at or just before the advent of vehicular access to As-Sayh, and commented that "the whole area is used by herds of sheep and goats at all altitudes" (Munton 1985). (Munton also speculated, from the lack of diversity of tree and shrub species in the Ru'us al-Jibal, that cutting of live wood has taken place in the past (see also Section 8.3 below). It is evident, at least, that cultivated sidr trees (Ziziphus spina-christi) have often been coppiced (Fig. 5.5.13).
It is worth reiterating that it is domestic and feral goats that are primarily responsible for overgrazing in the Ru'us al-Jibal, not donkeys (which are nevertheless sometimes unfairly blamed, perhaps because they do not have an economic constituency). The number of domestic or feral donkeys in most of the Ru'us al-Jibal today is negligible, although the number of domestic donkeys in the vicinity of As-Sayh (perhaps two to three dozen) may be increasing slightly.
One effect of overgrazing is that palatable species are reduced in number or eliminated (e.g., many Lamiaceae or mint family species, and many annuals), whereas spiny or unpalatable shrubs are left to dominate, although even the latter may be subject to grazing of tender new growth. This can result over time in the creation of a distinctive landscape of dense, spiny cushions or "tragacanthic steppe" (Zohary 1963, 1973). Among the more common Ru'us al-Jibal shrubs which appear to respond in this manner are Convolvulus acanthocladus, Convolvulus ulicinus, Farsetia aegyptia, Lactuca orientalis and Launaea spinosa. Gymnocarpos decandrus, another typically cushion-shaped dwarf shrub, is more tomentose than spiny, and does not appear to be grazed. Factors other than grazing may also be at work, however, since the cushion shape minimises the ratio of surface area to volume and could therefore be an adaptation to minimise insolation and water loss during prolonged summer drought (see also Section 8.4 below). The mountains of southern Spain feature a comparable plant community known as the "Hedgehog Zone", characterised by distinctive species such as Erinacea anthyllis, Genista umbellata, Bupleurum spinosum and (in the Balearics) Astragulus balearicus (R.J. Hornby, pers. comm., 2010).
During the course of this study the Ru'us al-Jibal also suffered (along with the UAE and Northern Oman generally) from an exceptional drought. UAE rainfall records show that the period from mid-1999 through mid-2004 was by far the driest 5-year period in the 70 years for which reliable records are available, with annual rainfall in each of those years below the long-term mean and aggregate rainfall less than half that of any other 5-year period (Feulner 2006).
The combination of extreme drought and accelerated grazing pressure resulted in floral changes in the high Ru'us al-Jibal that were evident even on the time scale of this study (from the early 1990s through 2010), at least in more accessible areas. A number of examples are cited in the following paragraphs.
Throughout the 1990s, the three most common small shrubs at high elevation in the Ru'us al-Jibal were Artemisia sieberi (the locally endemic wormwood, a species of sagebrush), Cymbopogon jwarancusa (a tussock grass and a source of citronella) and the robust, spiny Convolvulus acanthocladus. As the turn-of-the century drought progressed, both A. sieberi and C. jwarancusa appeared to diminish substantially in abundance, or at least in visual predominance, with the result that the spiny cushions of Convolvulus acanthocladus came to appear correspondingly more dominant. A regrettable aesthetic consequence was the absence for many years of the pervasive thyme-scented fragrance of A. sieberi that had been such a distinctive and pleasant accompaniment to the author's early excursions on the high plateaux. The dramatic declines of A. sieberi and C. jwarancusa are best attributed to climatic factors alone, since they appear seldom, if ever, to be the targets of grazing goats. Moreover, from about 2008 onwards, following several winters of rainfall at or above the long-term mean, both species had begun to recover their former predominance in the high elevation plant community.
A number of common but edible perennials such as Astragalus fasciculifolius, Centaurea wendelboi, Farsetia aegyptia, Launaea bornmuelleri and Teucrium stocksianum also appeared to be substantially reduced in numbers during the drought period. In these instances, overgrazing must be suspected as a contributing factor, although both F. aegyptia and T. stocksianum have appeared to be somewhat refreshed following successive abundant winter rains.
A few cases seem less equivocal. Overgrazing in the area of the high summits has reduced Phagnalon schweinfurthii to stunted specimens growing in cracks between rocks, although its normal habit is that of an erect dwarf shrub. An even more extreme example is the aromatic mint Salvia mirzayanii, which Mandaville (1985) reported from the west slopes of Jebel Harim, remarking on its "rather showy blue-violet flowers". For more than a decade and a half, however, it has been recorded only from a single remote ridgetop to the east of Jebel Harim, where no more than two dozen specimens have been observed. That same remote area is also the sole known Ru'us al-Jibal locality for two other species, Malcolmia africana and Astragalus sp. aff. schimperi, both probably edible species.
The spiny but edible leguminous shrub Astragalus fasciculifolius suffered from grazing in a somewhat different way during the turn-of-the-century drought. Beginning in 2000, the author began regularly to encounter specimens from which goats had detached the spiny crown of the plant and stripped the bark from the central stem, exposing the pale yellow-orange interior (Fig. 5.5.3). Presumably this was done for either dietary or medicinal reasons, to take advantage of some substance concentrated in or under the bark of the plant. One result is that in most areas A. fasciculifolius still does not appear to be as common, nor are its shrubs as large, as they were formerly.
5. Biogeographical affinities
The similarity of the flora of the Hajar Mountains to that of southern Iran and the Makran region (encompassing south-eastern Iran and the southern part of the province of Baluchistan in south-western Pakistan) has been recognised by many authors (e.g., Mandaville 1977; Kurschner 1986; Ghazanfar 1999, 2003), but successive efforts to refine the nature and boundaries of plant geographic zones in Arabia, Iran and adjacent areas have produced a daunting array of overlapping terminology.
Section 5.1 below is a brief attempt to introduce the non-specialist reader to the nature of the concepts introduced by plant geographers in the region and the general conclusions that have emerged. Section 5.2 presents a summary comparison of the flora of the mountains of Eastern Arabia (the Ru'us al-Jibal and Hajar Mountains) with that of Iran (to the north and east) and the rest of the Arabian Peninsula (to the west and south). In Section 6, the flora of the Ru'us al-Jibal is contrasted in detail with that of the Hajar Mountains.
To the extent that an accurate characterisation and delineation of plant geographic units permits useful conclusions to be drawn about biological and historical processes and events, the contribution of the present paper resides primarily in the content of the Checklist, which may permit subsequent researchers to make a better informed assessment of the proper categorisation of the flora of the Ru'us al-Jibal.
5.1. Phytogeographical categorisation and nomenclature in the Middle East. Eig (1931-1932) and Zohary (1973) attempted to generalise about the plant geography of the arid regions of North Africa, the Middle East and Iran and elaborated the concept of more or less latitudinal zones characterised by distinctive floral associations. The different zones are considered to be distinguished primarily by their temperature regime, including summer and winter maxima and minima. As a result, the generally latitudinal character of the zones may be greatly modified by features such as maritime influence and elevation. The concept of latitudinal zonation works relatively well in North Africa; it works much less well in the region of Arabia, where detours must be made for the Red Sea and the Arabian Gulf, as well as the mountain ranges that fringe much of Arabia and Iran.
Zohary (1963, 1973) recognised two major phytogeographical zones in Central and Eastern Arabia--a northerly one (the Saharo-Arabian region) and a southerly one (the Sudanian region) (see Map 5). The Sudanian region also extends eastward across southern Iran and Baluchistan (the Makran). Within the Sudanian region, Zohary found the flora of Eastern Arabia, southern Iran and southern Pakistan sufficiently distinctive to distinguish those areas as the Nubo-Sindian province.
Zohary nevertheless found it difficult to recognise the extension of the Saharo-Arabian region into Iran (Zohary 1963, 1973). Instead, in south-central Iran the Nubo-Sindian flora of the Sudanian Zone grades northwards more or less directly into the next phytogeographical region to the north, the Irano-Turanian region, which occupies most of central and northern Iran. The Irano-Turanian region also encompasses the arid areas of Central Asia to the north and east, and, like the Sudanian and Saharo-Arabian zones (but more northerly), the Irano-Turanian region can be traced westward, into the Anatolian plateau, through northern Iraq and the Syrian desert, and across North Africa to the Atlantic (Zohary 1963, 1973). Within Iran, the Irano-Turanian is characterised by species today best represented, or having their origin, in Iran or Central Asia.
The Sudanian region is considered tropical, and one of the principal differences between the Nubo-Sindian and Irano-Turanian florae in Iran is the tolerance (or requirement) of the former for high summer temperatures and/or an intolerance of low winter temperatures (Zohary 1963, 1973).
In Eastern Arabia, Zohary placed the boundary between the Sudanian and Saharo-Arabian regions along the southern edge of the Empty Quarter but included the mountains and coastal regions of Northern Oman and the UAE within the Sudanian region (see Map 4). Mandaville (1990), on more extensive floristic evidence, considered that most of Central Arabia should be considered within the Sudanian region sensu Zohary, and that the boundary should be drawn much further to the north, passing through central Qatar (which would correspond more closely to the regular influence of the westerly cyclonic belts). Mandaville (1985) nevertheless made an exception for the higher elevation flora of the Jebel Akhdar and the Ru'us al-Jibal, which he considered to constitute Irano-Turanian enclaves.
Kurschner (1986) took more direct account of the floral affinities between Northern Oman, southern Iran and Baluchistan by recognising an "Omano-Makranian subprovince". Leonard (1989), studying the Iranian flora, similarly identified an "Omano-Sindian area".
Ghazanfar (1992b, 1999, 2003) and Miller & Cope (1996) have adopted the subsequent nomenclature of Leonard (1989) and White & Leonard (1991) for Arabia, which accepts Zohary's major boundaries, but introduces new terminology: (1) the Somali-Masai regional centre of endemism in the southwest (including only southern Yemen and Dhofar) and (2) the Saharo-Sindian regional zone to the north. The latter is divided into three subzones in Arabia: (i) the Saharan regional sub-zone, encompassing the Red Sea coast of Arabia and a belt along the south coast; (ii) the Arabian regional sub-zone, encompassing most of central Arabia; and (iii) the Nubo-Sindian local centre of endemism, encompassing much of eastern Arabia and extending to the north-east across southern Iran, the Makran and much of Pakistan.
Ghazanfar (1999, 2003) delineates in greater detail the boundaries of the three Saharo-Sindian sub-units in eastern Arabia (see Map 6). The Arabian regional subzone extends roughly to the southern and eastern border of the sands of the Empty Quarter (including the inland sand desert regions of Dubai and Abu Dhabi as well as the southern portion of the Eastern Province of Saudi Arabia); the Saharan sub-zone extends in a belt along the south of the Empty Quarter, running north of the Hadramaut and Dhofar and across the central plains of Oman, Huqf and the Jiddat al Harasis to the Wahiba Sands; and the Nubo-Sindian local centre of endemism encompasses the Hajar Mountains and its outliers and alluvial plains, as well as the UAE coastline. Ghazanfar nevertheless acknowledges that it is often difficult to delimit the Nubo-Sindian local centre of endemism from the adjacent Arabian regional sub-zone.
Within the aforementioned zones and sub-zones of eastern Arabia, Ghazanfar also recognises three discrete and subsidiary local centres of endemism ("LCEs"): (a) the Dhofar Mountains; (b) the central limestone plateau area (Haushi-Huqf-Hayma-Harmoul); and (c) the Hajar Mountains (including the Musandam mountains or Ru'us al-Jibal) (Ghazanfar 1999, 2003). The inclusion of the Ru'u's al-Jibal as a separate enclave within the Hajar Mountain LCE should be reassessed in light of the information presented in the Checklist. Specifically, as noted above in Section 2, it now appears that no plant species are endemic to the Ru'us-al-Jibal and only five Eastern Arabian endemic species are found there: Desmidorchis arabica, Echinops erinaceus, Pulicaria edmondsonii, Pteropyrum scoparium (possibly synonymous with the Iranian P. aucheri) and Stipa mandavillei.
[FIGURE 3.3.1 OMITTED]
[FIGURE 3.3.2 OMITTED]
[FIGURE 3.3.3 OMITTED]
[FIGURE 3.3.4 OMITTED]
[FIGURE 3.3.5 OMITTED]
[FIGURE 3.3.6 OMITTED]
[FIGURE 3.3.7 OMITTED]
[FIGURE 3.3.8 OMITTED]
[FIGURE 3.3.9 OMITTED]
5.2. Iranian/Eastern affinities of the Ru'us al-Jibal flora. Putting aside the broader phytogeographical classifications, there has been been agreement on the affinity between the flora of the Hajar Mountains generally (including the Ru'us al-Jibal) and that of southern Iran and Baluchistan. Zohary's personal acquaintance with Eastern Arabia seems to have been limited (for example, his 1973 maps do not identify the Ru'us al-Jibal as an area of distinctively high elevation) but he referred generally to "the Bandar Abbas-Jebel Akhdar connection" (Zohary 1973). Subsequent authors have also emphasised this relationship (e.g., Mandaville 1977; Kurschner 1986; Leonard 1989; Ghazanfar 1999, 2003). Kurschner (1986) designated the combined area as the Omano-Makranian subprovince of Zohary's Nubo-Sindian province of the Sudanian region, and Leonard (1989) recognised an Omano-Sindian area. It adds weight to this botanical generalisation that various elements of the fauna of the Hajar Mountains and the Ru'us al-Jibal have a similar biogeographical distribution encompassing Northern Oman, the Musandam and southern Iran and Pakistan. These include, inter alia, land snails (e.g., Granaria persica, restricted in Eastern Arabia to the high Ru'us al-Jibal), reptiles (e.g., the false horned viper Pseudocerastes persicus), and even mammals (e.g., Blanford's fox Vulpes cana).
This floristic affinity is also seen in the fact that, although most of the perennial species of the Hajar Mountains probably have a Nubo-Sindian, Sudanian or even pan-Eremic distribution, several of the most common species of the Ru'us al-Jibal (and of the Hajar Mountains generally) are more common in Iran than they are elsewhere in Arabia, to the north-west, west or south. The best example is Euphorbia larica, which is among the most common and characteristic mountain species of the UAE and Northern Oman, and which is common in southern Iran (Zohary 1963). In Arabia, E. larica is effectively limited to Northern Oman. It extends south-west to the dry parts of north-eastern Dhofar, but away from the Hajar Mountains it is always rare. Other examples include Physorrhynchus chamaerapistrum and Plocama aucheri (formerly Gaillonia aucheri or Jaubertia aucheri), both of which have the western limit of their range in the Hajar Mountains.
Zohary (1963) discusses in various contexts the characteristic species of southern and south-western Iran. He classifies the flora from the coasts of the Arabian Gulf and the Gulf of Oman inland to elevations of c.1000 m as primarily Nubo-Sindian, and the overall vegetation as an impoverished savannah or tropical desert. His description of the species encountered there (at pp. 42-43) makes familiar reading for anyone acquainted with the wadis, lower slopes, foothills and alluvial plains of the Hajar Mountains. A majority (59 of 111) of the southern Iranian species mentioned by Zohary at pp.63-64 (and a greater majority of the genera) are common to those Hajar Mountain environments as well. The list of shared species is shown in Table 4. Zohary considered most of these species to be of "African" origin, but cited a few as belonging to Asian tropical stock (e.g., Euphorbia larica, Nannorrhops ritchieana and Prosopis cineraria).
Some obvious reasons for the observed floral similarity include physical proximity and physiographic and environmental similarity (rocky mountains, foothills and alluvial plains, versus the sand and gravel flats of central Arabia), as well as similar rainfall amount and patterns, including the absence of summer rains.
Beyond these generalisations there exist certain subsidiary patterns that have the potential to repay closer attention. For example, of the 43 southern Iranian species listed in Table 4 that are primarily mountain species in Eastern Arabia, 13 (30%) are among the Hajar Mountain species that are absent or rare in the Ru'us al-Jibal and 10 of these are among the common such species listed in Table 5 (see Section 6 below and compare Table 4 and Table 5). In contrast, only 3 of the Table 4 Iranian species are essentially limited to the Ru'us al-Jibal (Astragalus fasciculifolius, Convolvulus acanthocladus and Teucrium oliverianum).
Among the high elevation species of the Ru'us alJibal, the affinity with Iran is particularly striking. The majority of the most common species have the centres of their range (and/or have their closest congeners) in Iran or neighbouring areas of Central Asia, e.g.: Artemisia sieberi, Astragalus fasciculifolius, Centaurea wendelboi, Convolvulus acanthocladus, Cymbopogon jwarancusa, Ephedra pachyclada, Jurenia berardioides, Prunus arabica and Teucrium stocksianum (Zohary 1963, Mandaville 1985, Ghazanfar 1998b). The same can be said of other, less common, high elevation species such as Aegilops kotschyi, Dianthus crinitus, Leontice leontopetalum and Salvia mirzayanii (Zohary 1963, Mandaville 1985, Ghazanfar 1998b).
Of the foregoing species, Ephedra pachyclada is the only one also found in the mountains of western Arabia (Boulos FoE; Collenette 1985), although Artemisia sieberi otherwise ranges from the eastern Mediterranean to south-west Pakistan and Central Asia. Cymbopogon jwarancusa is characterised by Ghazanfar (1998b) as a mesic Asian relict. Prunus arabica and Artemisia sieberi both exhibit what Mandaville (1985) called a "two-pronged" entry into the Arabian Peninsula from Iran: a northern route via the Syrian desert and a southern route via the Ru'us al Jibal.
Of the other most common species of the high Ru'us al-Jibal, Dodonaea viscosa is nearly cosmopolitan in tropical and subtropical regions; Gymnocarpos decandrus and Helianthemum lippii have a pan-Eremic distribution from the Canary Islands to Pakistan (Boulos 1999, 2000).
Within Iran, most of the high elevation Ru'us al-Jibal species have a more northerly or higher elevation distribution than the southerly and lower elevation Nubo-Sindian species. Instead, they belong to Zohary's Irano-Turanian region. Zohary's descriptions of the vegetation of the Central Iranian plateau frequently refer to Artemisia steppe, including open forests of Amygdalus (Prunus) spp. and Pistacia spp., and he identifies genera (or sections thereof) such as Aegilops, Amygdalus (Prunus), Anthemis, Artemisia, Astragalus, Heliotropium, Pistacia, Salvia and Trigonella as being mainly Irano-Turanian. Some common Irano-Turanian species are not present in the Ru'us al-Jibal, but are common in the high Jebel Akhdar, mostly at elevations greater than those reached in the Ru'us al-Jibal. This group includes, e.g., Dionysia mira and Ebenus stellata.
Overall, it is possible to generalise as follows: The flora and vegetation of the Hajar Mountains generally resembles that of southern Iran, the Makran and Baluchistan. The flora and vegetation of the high Ru'us al-Jibal, however, more closely resembles that of the central Iranian plateau. And, to anticipate the discussion below (see Section 8.4) the flora and vegetation of the high Jebel Akhdar show greater similarities with mountain areas to the north and northeast in Afghanistan and Pakistan. The main parameters that give rise to these relationships are probably temperature (especially annual minima and maxima) and rainfall amount and pattern. The foregoing generalisations do not, however, readily account for the consistent floral differences between the Ru'us al-Jibal at low and medium elevations and the Hajar Mountains generally (see Section 6 below).
[It is a curious historical parallel that the original usage of the term "Makran" (by Herodotus among others) referred collectively to most of the geographical areas under discussion here for their biogeographical affinity--from the southern parts of Sindh, Baluchistan and Iran across the Strait of Hormuz to the southern Arabian Gulf and Northern Oman--that being the extent of Maka, a satrapy of the 4th to 6th century B.C.
6. Many typical Hajar Mountain species are absent in the Ru'us al-Jibal
Some three dozen species, mostly perennials but including a few annuals, that are reasonably common and characteristic within the Hajar Mountains of the UAE and Northern Oman, are nevertheless absent or very rare in the Ru'us al-Jibal. These are listed in Table 5. The perennials are mostly dwarf shrubs but also include trees, shrubs and herbs.
The reasons for such a clear and consistent floral discontinuity most likely relate to factors deriving from the substantial differences in the bedrock geology of the two areas: the carbonate rocks that comprise most of the Ru'us al-Jibal versus the ultrabasic silicate rocks of the ophiolite nappe that make up the bulk of the Hajar Mountains to the south (along with subsidiary areas of basic silicate rocks and and deep water siliceous sediments of the Hawasina group). Those derivative factors probably include not only soil and groundwater chemistry but also retention and circulation of scarce groundwater. The putative influence of these and other factors is discussed below, although in the absence of specialised study it is impossible to offer direct or definitive evidence in support of any particular hypothesis.
In some cases, however, the observed distribution may reflect broader regional biogeographic gradients.
6.1. Differential hydrology: karst versus ophiolite. In the case of the few water-loving species whose primary habitat is gravel wadi beds, particularly Saccharum griffithii, Nerium oleander, Arundo donax, Schoenus nigricans, Juncus socotranus and Doellia bovei, their absence in the Ru'us al-Jibal is probably due directly to the extreme scarcity of shallow groundwater in the wadis of the Ru'us al-Jibal, which in turn is attributable to the karst hydrology of the carbonate bedrock, involving internal solution channels. This explanation is supported by the presence of several of the foregoing species at a unique site within the Ru'us al-Jibal, at 'Ayn as-Sih, discussed below at Section 12.
In contrast, the hydrological regime of the Hajar Mountain ophiolite conforms relatively well to the generalised description by Harrison & Kruckeberg (2008): ultrabasic rocks, where they appear at the earth's surface, have usually been pervasively fractured by tectonic forces, and "readily form vast masses of sharp-edged scree that drift down hillsides ... where only the hardiest plants can survive. The same rock characteristics, however, allow underground water to move through cracks in the masses of shattered rock, instead of through pores in rock, as it does elsewhere. Where such water encounters an impermeable layer, it may emerge at a slow but steady rate," facilitating plant growth.
Thus in some cases drainage characteristics of the substrate unrelated to chemistry may be the principal determinant of distribution. Among the best examples are the two Lindenbergia species (L. arabica (Fig. 6.1.7), which is endemic to the UAE and Oman, and L. indica). Both are typically found on, or at the base of, the vertical wadi walls of coarse gravel terraces--an extremely common habitat in the Hajar Mountains but a rare one in the Ru'us al-Jibal. The gravel terraces serve, among other things, as reservoirs for the steady percolation and release of groundwater.
Another example is Prosopis cineraria, the ghaf tree. The ghaf is primarily a species of sand and gravel plains, but it is not unusual to encounter it in or adjacent to broader, flatter gravel wadis within the Hajar Mountains. Yet although it is common on the plains to the west of the southern Ru'us al-Jibal, from Ra's al-Khaimah southwards, it is absent within the wadis of the Ru'us al-Jibal itself. Edaphic (soil/substrate) and hydrologic factors are the most likely explanations for its absence. P. cineraria is a notoriously deep rooted species but it may be unable to penetrate the carbonate bedrock, or to find sufficient water under karst conditions.
6.2. Distinctive chemistry of the ophiolite environment. The weathering of ultrabasic rocks such as the ophiolite of the Hajar Mountains creates soils that generally have distinctive chemical characteristics (Harrison & Kruckeberg 2008). They are deficient in calcium and other essential nutrients, such as nitrogen and phosphorus; they are rich in magnesium, which interferes with the uptake of calcium; and they usually have high levels of heavy metals such as chromium, nickel and cobalt. The slow percolation of groundwater through ultrabasic rock in an arid climate also produces exceptionally alkaline groundwater, with pH as high as 11.9 (Clark & Fontes 1990; J. Burt, pers. comm. 2004).
Consistent with the hypothesis of control by the substrate, in the case of many of the Table 5 species that have been found in small numbers in the Ru'us al-Jibal, their occurence there is limited to the southern fringes, in particular Jebel Ayuzah and Jebel Wamm in the south-east (Table 6b), where the bedrock is not the shallow water carbonates that dominate most of the Ru'us al-Jibal. However, neither is it the ophiolite that dominates the mountains to the south, but rather terrigenous sediments--the turbidites, shales and localised conglomerates of the Hawasina group (Figs. 1.5.2 and 1.5.3)--leaving the nature of the correlation with substrate uncertain. (A plant so far known only from this geologically intermediate area is the recently recognised Chesneya parviflora, common in the Makran.)
The most compelling example of a species having a distribution favoring carbonates and avoiding ophiolite is the hanging caper Capparis cartilagenia, which is common on cliffs in the Ru'us al-Jibal and on virtually all outliers of carbonate rock to the south (e.g., Jebel Mleihah, Jebel Rawdhah, Jebel Ghaweel and Jebel Hafit) but absent in the intervening ophiolite. Two other species that may avoid the ophiolite are Koelpinia linearis (Asteraceae) and Ducrosia anethifolia (Apiaceae), but both are sufficiently rare, even in the carbonates where they are found, that the failure so far to observe them within the ophiolite cannot as confidently be taken to demonstrate absence.
In a small number of cases, Table 5 species found in the Ru'us al-Jibal are limited to the south-west Ru'us al-Jibal, where the bedrock is the same as the carbonates of the Ru'us al-Jibal generally (Table 6a). That area is distinguished, however, by the fact that it is much lower overall and offers gentler slopes at lower elevations (Fig. 1.5.1). The characteristic profile of the Ru'us al-Jibal is of steep slopes rising from base level to 500 m or more, and only then giving way to flatter terrain. In the south-western Ru'us al-Jibal, however, it is possible to ascend by ramp routes from almost base level on the adjacent plains to plateaux at 300-500 metres. Possibly this physiographic difference is sufficient to accommodate a few species and/or to facilitate their introduction.
What is arguably most remarkable, in light of the distinctive nature of the ophiolite substrate, is that the flora of the Hajar Mountains is not richer in endemic species. On a worldwide scale, ultrabasic habitats are uncommon and typically localised, but they are nevertheless widely distributed, occurring in association with most mountain belts, both ancient and modern. Elsewhere they are notorious for endemism: in California, some 12.5% of the state's 2,000 endemic species are said to be restricted to ultrabasic rock; within the tropics, both Cuba and New Caledonia, each of which has substantial areas of ultrabasic rock, have more than 900 ultrabasic endemics, a number of which are distinctive for having evolved the ability to concentrate heavy metals (Harrison & Kruckeberg 2008).
In Oman, on the other hand, despite its endowment with the world's most extensive exposure of ultrabasic rocks, the most common national or regional endemics (Ghazanfar 1999) are either not found in or not restricted to the ophiolite. In fact, it is not clear that any at all of the c.25 national or regional endemics recognised from the mountains of Northern Oman (Ghazanfar 1998b, 1999, 2003) are so restricted. Moreover, in Oman it seems that the only plant mentioned as a known hyperaccumulator of heavy metals is the fern Pteris vittata (Chen et al. 2002).
The Oman experience is seconded by a preliminary survey of two ultrabasic areas in central Iran, totalling c. 450 sq. km and located within a melange zone. Only a single species, the herb Cleome heratensis (Capparaceae), was identified as endemic to the ultrabasic rocks (Ghaderian & Baker 2007). C. heratensis is said to cover extensive areas in summer and autumn when there is no rainfall, and is able to restrict excessive heavy metal uptake.
6.3. Regional biogeographic gradients. Several Hajar Mountain species diminish or disappear in a northerly direction even before reaching the Ru'us alJibal. The large shrub Acridocarpus orientalis, the aromatic Composite Pluchea arabica and the tree Maerua crassifolia*, for example, are all but absent north of the Al-Ain/Buraimi area and Wadi Jizzi in Oman (approximately 24[degrees]15' N latitude) and are not found north of Mahdhah, although all are very common in the mountains to the south. The spiny shrub Zilla spinosa and the rectilinear mountain front form of Convolvulus acanthocladus extend only slightly north of Mahdhah. The tomentose Heliotropium rariflorum (tentatively identified by Tim Harrison, pers. comm.) is locally common on the gravel outwash plains south of Mahdhah but has never been recorded further north. [*Note: A small number of Musandam area records exist for Maerua crassifolia. The author has been able to obtain reasonable confirmation of one of these, which is within the Ru'us al-Jibal as defined here (see Checklist). The others are believed most likely to occur in anthropogenic environments.]
Other species have their northernmost occurrences within the Hajar Mountains between Hatta (Wadi Hatta) and Masafi (Wadi Ham). These include the toxic Composite Iphiona aucheri, the dwarf palm Nannorrhops ritchieana, the wild olive Olea europaea and the elusive Rumex limoniastrum, collected from Eastern Arabia by Aucher-Eloy but not recorded again until the mid-1990s (Curtis 1999). Still other species become increasingly rare northwards within the Hajar Mountains, e.g., the erect shrub Hibiscus micranthus and the semi-prostrate Schweinfurthia papilionacea.
The reason for these intra-regional boundaries is speculative at present. They could be an indicator of the geographical advance or retreat of the species in one direction or the other, but they could also reflect factors such as lower annual rainfall, or lower or less reliable summer rainfall, along a northerly gradient. The latter, in turn, could be a result of regional atmospheric conditions (e.g., the northern limit of the influence of the Indian Ocean monsoon) or the lower elevation of the mountains between Hatta and the Ru'us al-Jibal (with maximum summit elevations of 1100+ m, versus 1400+m to the south) and a correspondingly reduced orographic effect. These factors are discussed at greater length in Section 9 below in connection with the case of the wild olive O. europaea.
The distributions do not, in any case, seem to represent a simple gradient in response to factors directly correlated with latitude, such as differences in daylight hours or lower temperatures in a northerly direction. As noted above, a number of the Hajar Mountain species that are absent or rare in the Ru'us alJibal are nevertheless common and characteristic in Iran and/or the Makran (see Section 5.2 above and compare Table 4 and Table 5). For example, Iphiona aucheri is more common in southern Iran than in the Hajar Mountains and was once thought to be an Iranian endemic (Zohary 1963, as Grantia aucheri); Nannorrhops ritchieana is present in south-central Iran and the Makran, as well as northern India, and is considered to be an Omano-Makranian species (Deil & al Gifri 1998) of Asian origin (Zohary 1973), although its Arabian range extends to southern Yemen and it is much more common on the plains of central Oman than in the Hajar Mountains; and O. europaea can also be found in south-central Iran and the Makran, as well as the mountains of Afghanistan and northern Pakistan (see Section 9 below).
Whatever the reasons for these observed boundaries or gradients, their existence should serve as a caution against casual predictions that species found in Northern Oman can be expected to occur in the UAE.
6.4. The wadi bottom habitat. Isolated records of several typical Hajar Mountain species (Cometes surattensis, Lotononis platycarpa, Reseda sp.) are attributable to a spring visit (in April 2005) to wadi bed and low wadi bank habitats along the middle reaches of Wadi Bih, specifically a wide spot in the already broad gravel bed of Wadi Bih, in an area of metre high terraces formed at the junction of Wadi Bih with a minor tributary wadi descending from the cliffs above, near Sabtan. These observations leave open the possibility that a more dedicated investigation of wadi bottom environments could reveal additional records of Hajar Mountain species not otherwise represented in the Ru'us al-Jibal.
The same locality (which is subject to modest grazing pressure by goats) was visited for comparison in early May 2009, when the only unexpected record was a single small Haplophyllum tuberculatum. Abundant on both occasions were Ammi majus, Erucaria hispanica and Pentanema divaricatum and, in 2009, Aristida abnormis and Scabiosa olivieri. Nearby, also in May 2009, Ochradenus aucheri was uncharacteristically dominant on rubble slopes adjacent to the main wadi, where Plocama aucheri was also common but grazed.
Rhazya stricta is most common in the UAE and Northern Oman as a species of gravel plains, but it is not unusual to encounter it in Hajar Mountain wadis, especially in or alongside the beds of broader, flatter wadis. In all cases the profusion of R. stricta is an indicator of overgrazing. It is absent, however, within the wadis of the Ru'us al-Jibal, with a single exception for which overgrazing also appears to be the presumptive explanation. The middle reaches of Wadi Khabb in the southern Ru'us al-Jibal, in the vicinity of its Wadi Shakh tributary, is the only area within the Ru'us al-Jibal that is extensively cultivated far above the wadi mouth. The cultivation there is associated with deep water sediments of the Hawasina group, having a different chemical compositon from the typical Musandam carbonates and also a different hydrology, with springs that are tapped to supply water. The presence of people and agriculture is in turn associated with large numbers of domestic goats and consequent heavy overgrazing, which seems to give R. stricta the same comparative advantage in Wadi Khabb that it confers elsewhere.
7. Many higher elevation ruderals of the Ru'us alJibal are absent in the Hajar Mountains, including the Jebel Akhdar
The Ru'us al-Jibal features a number of higher elevation ruderal species that are not found in the Hajar Mountains to the south. These include most prominently the monocotyledons Moraea sisyrinchium, Leopoldia longipes and Ixiolirion tataricum and the eudicotyledon Heliotropium bacciferum. In the same category, but somewhat more localised and peri-anthropic in their distribution, are the annuals Asperugo procumbens, Cardaria draba, Galium spp. (G. aparine, G. ceratopodum and/or G. tricornutum), Matricaria aurea, Papaver sp., Plantago notata, Roemeria sp. and Scrophularia arguta and the perennials Gladiolus italicus, Leontice leontopetalum and Teucrium oliverianum. A number of these peri-anthropic species are subject to grazing pressure and are reliably found only where they are physically protected, whether by fencing or natural obstacles.
One obvious explanation for the restriction of these species to the Ru'us al-Jibal is that similar habitats are not available in the mountains immediately to the south. Within the Ru'us al-Jibal, scattered terraced cultivation is widespread, essentially all of it at elevations of 4501500 m. In contrast, between the Ru'us al-Jibal and the Jebel Akhdar, a distance of some 250 km, there are virtually no cultivated fields (active or historical) at elevations above c.700-800 m (which in some areas is the base level at the mountain front). This is primarily due to the different physiography of the intervening Hajar Mountains, which are somewhat lower and which tend to weather to hillsides of steep but fractured bedrock, unsuitable for terracing, without the high plateaux that characterise the Ru'us al-Jibal. However, all of the characteristic ruderals listed above are apparently also absent from the Jebel Akhdar (with the exception of a single peri-anthropic occurrence of Asperugo procumbens), where habitats seemingly comparable to those of the Ru'us al-Jibal are common.
Another possibility, or perhaps a complementary one, is that some of these high elevation ruderals have been introduced through agricultural activities, whether in historical times or more recently. One species that is reckoned to have escaped from cultivation and perianthropic confines and is now established in the mountain environment more generally, both in the Ru'us al-Jibal and in the Hajar Mountains, is Linum corymbulosum (Ghazanfar 2007).
It should also be highlighted that most of these high elevation ruderals have their principal geographic ranges within more temperate latitudes, from Central Asia westward across Mesopotamia, the Levant and Turkey to the circum-Mediterranean (Boulos, 19992005) and the Atlantic (in the case of Asperugo procumbens, Cardaria draba, Papaver sp. and at least two of the Galium spp.) (R.J. Hornby, pers. comm.).
8. Botanical comparison of the Ru'us al Jibal with the Jebel Akhdar
The Ru'us al-Jibal and the Jebel Akhdar are recognised as similar in their geology, physiography and history of human habitation, and the Jebel Akhdar has been relatively well studied botanically, by the standards of the region, so it is potentially enlightening to compare the two.
8.1. Comparison of the vegetation zones of the Ru'us al-Jibal and the Jebel Akhdar. Mandaville (1977) and Ghazanfar (1991a, 2003) have published broadly similar accounts of the zonation of the Jebel Akhdar vegetation. These are compared in Table 7 with the author's zonation for the Ru'us al-Jibal. Both Mandaville and Ghazanfar recognise four main zones in the Jebel Akhdar:
(1) a mountain wadi zone dominated by Acacia tortilis et al.
(2) a lower slopes zone dominated by Euphorbia larica et al.
(3) an upper slopes zone dominated by Sideroxylon mascatense (syns. Reptonia mascatensis, Monotheca buxifolia) and Olea europaea.
(4) a summit zone dominated by Juniperus excelsa polycarpos et al.
Zones 1 and 2 described above for the Ru'us al-Jibal are substantially similar to the two lowest zones recognised by Mandaville and Ghazanfar for the Jebel Akhdar. The most significant difference is that each of the zones is shifted downward in elevation by c.300-500 m in the Ru'us al-Jibal. This may reflect in part the fact that the Ru'us al-Jibal is situated some 250 km or 2.5 degrees latitude further north, so that a similar temperature regime prevails at lower elevation. However, an equally or more significant contributing factor may be that the base elevation at the mountain front is some 400-500 m higher in the Jebel Akhdar than in the Ru'us al-Jibal, so that similar physiographic environments and their associated habitats are correspondingly higher in the Jebel Akhdar. For example, the plains to the south of the Jebel Akhdar at Bahla and Nizwa, and to the north at Rustaq and Awabi, have an elevation of c.550 m, whereas the plains to the south-west of the Ru'us alJibal at Khatt, or to the south-east at Dibba, have an elevation of less than 100 m, and to the north and east of the Ru'us al-Jibal the base level is effectively sea level.
There are nevertheless a few noteworthy floristic differences in these lower zones, between the Ru'us alJibal and the Jebel Akhdar. As characteristic species of the mountain wadi zone, Mandaville includes the large shrub Acridocarpus orientalis and Ghazanfar includes the tree Maerua crassifolia. Although A. orientalis is very common in the Jebel Akhdar and in the Hajar Mountains immediately to the northwest of the Jebel Akhdar, both of those species are absent to the north of latitude 24[degrees]22'00", only some 15 km north of Al-Ain, Buraimi and Wadi Jizzi, and still some 150 km south of the Ru'us al-Jibal. Similarly, as smaller shrubs characteristic of the mountain wadi zone, Ghazanfar designates Rhazya stricta, Fagonia indica and Pteropyrum scoparium, but R. stricta is rare and P. scoparium is uncommon within the Ru'us al-Jibal. See also Section 6.3 above.
At higher elevations the situation is quite different. Above Zones 1 and 2 (wadis and low to medium elevation slopes) there is only limited botanical similarity between the two regions. Zone 4 of the Jebel Akhdar is absent in the Ru'us al-Jibal, probably for the obvious reason that it occupies elevations above those reached in the Ru'us al-Jibal, even allowing for a 300-500 m adjustment.
Zone 3 differs significantly between the two regions in terms of both species and vegetation structure. In the Jebel Akhdar, Zone 3 consists of an open woodland on moderate slopes and plateaux, with trees and large shrubs, and it can be forested in wadis and on protected slopes (see Figs. 4.1.5 to 4.1.12). In the Ru'us al-Jibal, Zone 3 is characterised by what has been called Artemisia steppe, consisting of low shrubs (including prominently Artemisia sieberi), many of them more or less tragacanthic (growing as spiny cushions), and scattered almond trees, stunted where exposed (see Section 3 above and Figs. 1.3.6, 3.1.1 to 3.1.19).
Accordingly, to the extent that the two regions appear visually similar, this may be due more to their similar geology and physiography, and to the constraints imposed on plant growth forms, than to actual similarities in their respective florae. Details of floristic differences, and of the tree flora in particular, are discussed immediately below.
8.2. Floristically, the high Ru'us al-Jibal is not closely similar to the high Jebel Akhdar. At elevations above Zone 2, i.e., above the low and medium elevation montane zone of the Ru'us al-Jibal and the lower slopes zone of the Jebel Akhdar (effectively, the Euphorbia larica zone in each), only a few species are common in both ranges: Cymbopogon jwarancusa, Dodonaea viscosa, Ephedra pachyclada, Launaea bornmuelleri and Phagnalon schweinfurthii. Three of the four most common perennials of the high Ru'us al-Jibal are effectively absent in the high Jebel Akhdar: Artemisia sieberi, Convolvulus acanthocladus (extremely rare), and Prunus arabica. In addition, several of the next most common dwarf shrubs in the high Ru'us al-Jibal are either rare (Centaurea wendelboi, Farsetia aegyptia) or absent (Astragalus fasciculifolius, Gymnocarpos decandrus) in the Jebel Akhdar. Teucrium is common at higher elevations in both areas, but the species found in the Ru'us al-Jibal is T. stocksianum whereas in the Jebel Akhdar it is T. muscatense (Ghazanfar, in press, synonymises these two Teucrium taxa, but see the Checklist and compare Figs. 4.1.15 and 5.4.24).
Conversely, with the exception of Dodonaea viscosa and Ephedra pachyclada, none of the species cited by Mandaville as dominants or associates in the Sideroxylon-Olea zone (Mandaville's "Reptonia-Olea Woodland") are found in the Ru'us al-Jibal: Sideroxylon mascatense, Olea europaea, Sageretia thea (syn. S. spiciflora), Ziziphus hajarensis, Juniperus excelsa, Clematis orientalis, Ebenus stellata and Acacia gerrardi.
8.3. The diverse tree flora of the Jebel Akhdar is absent in the Ru'us al-Jibal. The number of species of trees and large shrubs found in the Ru'us al-Jibal is very small: Acacia tortilis, Dodonaea viscosa, Ficus cordata salicifolia, Ficus johannis, Moringa peregrina, Prunus arabica and Ziziphus spina-christi.* Of these, A. tortilis, F.c. salicifolia and M. peregrina are found only at low and medium elevations, generally up to a maximum of 600-1000 m. Ficus johannis and Ziziphus spina-cristi survive in small numbers to c.1500-1600 m, but only Dodonaea viscosa and Prunus arabica are present above that. [*The foregoing list excludes the extremely rare Acacia ehrenbergiana, Cordia sp. aff. quercifolia, Ehretia obtusifolia, Grewia tenax, Grewia villosa and Pistacia khinjuk, each found only at a single locality at low elevations.]
In contrast, the Jebel Akhdar above 1000 m is characterised by many trees and large shrubs, including Acacia gerrardii, Acridocarpus orientalis, Berberis baluchistanica, Daphne mucronata, Dodonaea viscosa, Juniperus excelsa polycarpos, Lonicera aucheri, Olea europaea, Sageretia thea (syn. S. spiciflora), Sideroxylon mascatense (syns. Reptonia mascatensis, Monotheca buxifolia) and Ziziphus hajarensis. Prunus arabica is absent from the Jebel Akhdar, although it is present in the Jebel Bani Jabr to the south-east. Only at about 2400 m and above does the number of Jebel Akhdar tree species start to decline (see Figs. 4.1.1 to 4.1.5). At the highest elevations, open plateaux are more or less limited to juniper and D. viscosa, although other tree species (B. baluchistanica, D. mucronata, L. aucheri, O. europaea, S. mascatense) remain present in ravines, sometimes almost to the 3000 m summits. Ficus cordata salicifolia, Moringa peregrina and Ziziphus spina-christi are common at lower and moderate elevations in the Jebel Akhdar, as they are in the Ru'us al-Jibal.
Munton (1985), studying in detail the habitat of the Arabian tahr, also remarked that the Ru'us al-Jibal was "notable for its comparative lack of diversity of shrub and tree species" and speculated that it was "possible that, unlike in areas further south [the Jebel Akhdar and Eastern Hajar], cutting of live wood has taken place in the past." That possibility may not be exceptional. Deil & al Gifri (1998) have commented more generally that the majority of Arabian grasslands are secondary in nature, having replaced woodlands which were destroyed by felling and burning, and Miller & Cope (1996) have lamented that: "The greatest problem in describing the vegetation of the Arabian Peninsula is the effects man and his livestock have had for thousands of years. The potential vegetation in most areas can only be guessed at from small vestiges of apparently natural vegetation which have managed to survive in inaccessible spots."
8.4. Rainfall amount and patterns: the probable explanation for botanical differences. What accounts for the manifest differences in the high elevation flora and vegetation of the Ru'us al Jibal as compared with the Jebel Akhdar? Differences in the amount and pattern of annual rainfall are almost certainly the principal factors.
Zohary (1973) considered that the arboreal assemblage of the Jebel Akhdar suggested an annual rainfall of about 200 mm, whereas Mandaville's (1977) estimate from vegetation was somewhat higher at 250-375 mm. More recent but still limited data from the Saiq Plateau indicate that average annual rainfall in the Jebel Akhdar may be from 307 mm (Ghazanfar 1992b) to 350 mm (Fisher et al. 1999, referencing Fisher & Membery 1998). Miller & Cope (1996) estimate a figure of 300-350 mm.
For the Ru'us al-Jibal, Ghazanfar's (1992b) average annual figure of 188 mm/yr for Khasab is consistent with the limited measured data (for 1975-1977) reported in Mandaville (1985). This is significantly higher than Mandaville's interpolation of 100-150 mm/yr from low elevation coastal locations at Sharjah, UAE and Bandar Abbas, Iran (Mandaville 1985), and suggests a modest to significant orographic effect. Data from the coastal site of Sha'am, at the western edge of the Ru'us al-Jibal, record a 25-year average (1967-1992) of 165 mm/year (Boer 1997, citing UAE Ministry of Agriculture and Fisheries 1993). Fisher et al. (1999), referencing Fisher & Membery (1998), show average annual rainfall of 198 mm at Khasab; this somewhat higher figure may be influenced by the unusually wet years of the mid-1990s. A map of isohyets (contours of equal rainfall) showing mean annual UAE rainfall for the 18-year period 1971/1972 through 1988/1989 indicates a figure for the western Ru'us al-Jibal in excess of 160 mm, but apparently not in excess of 180 mm (UAE University 1993, Plate 50).
Generally consistent with the foregoing are annual rainfall data published by Emirates Wildlife SocietyWWF (2006) for multiple stations in the Shimaliyah range of the Hajar Mountains, on the East Coast of the UAE between Khor Fakkan and Masafi, some 10-40 km south of the Ru'us Al-Jibal. These indicate a 30-year (1975-2004) regional average of c.160 mm/year, with the composite annual average ranging from 43 mm to 323 mm. Peaks in Shimaliyah are substantially lower than in the Ru'us al-Jibal, with the few very highest ranging from 900-1100 m. The Shimaliyah data are also consistent with the historical periodicity of other UAE rainfall records (Feulner 2006), including the exceptional rainfall of the mid-1990s (1995-1998) and the exceptional drought of 1999-2003.
The foregoing figures indicate that average annual rainfall is some 50-75% higher in the Jebel Akhdar than in the Ru'us al-Jibal. Of equal significance, however, is the annual rainfall distribution. The data show that the Jebel Akhdar receives a substantial and reliable dose of summer rainfall (July and August) (Ghazanfar 1991a, 1992b), which the Ru'us al-Jibal today does not (Ghazanfar 1992b, UAE University 1993 and personal observation).
An explanation of floral and vegetation differences in terms of rainfall amounts and patterns is supported by the similarity of the flora of the high Ru'us al-Jibal, and particularly the development of Artemisia steppe, to that of the central plateau of Iran, where total annual rainfall in areas of comparable vegetation ranges from c.100-135 mm and where summer rain is also lacking (Zohary 1963). Mandaville (1977) contributes a further supportive comparison, noting that the wild olive species present in the mountains of Afghanistan and Pakistan (which he called Olea ferruginea) is associated there with Sideroxylon mascatense (syns. Reptonia mascatensis, Monotheca buxifolia), which is common and closely associated with O. europaea in the Jebel Akhdar. Mandaville notes that although total rainfall in the mountains of Afghanistan and Pakistan is higher than in the Jebel Akhdar, both regions are characterised by relatively abundant summer rainfall.
The summer rainfall in the Jebel Akhdar is probably the result of a combination of the influence of the Indian Ocean monsoon and the orographic effect of the Jebel Akhdar massif, with extensive plateaux above 1800 m and summit ridges to just over 3,000 m. Limited data from the city of Nizwa, on the plains just south of the Jebel Akhdar, indicate that precipitation is low (<100 mm/yr) and falls mainly in the summer (Ghazanfar 1991a), probably reflecting monsoon related phenomena. The Jebel Akhdar may be a barrier to northward penetration of the monsoon, since both Muscat (Boer 1997) and Seeb (Ghazanfar 1992b) on the Gulf of Oman coast north of the Jebel Akhdar exhibit the pattern of winter rain and scarce summer precipitation that seems to characterise the rest of the Hajar Mountains. However, the easterly city of Sur, on the Gulf of Oman coast near Ra's al-Hadd, also shows the typical winter rainfall pattern (Ghazanfar 1992b), although it is not much protected from the south.
8.5. Further comparison with the Eastern Hajar Mountains. The distinctiveness of the Jebel Akhdar flora and vegetation is emphasised by comparison with high elevation areas in the Eastern Hajar, including Jebel Aswad (to c.1900 m) (described in Mandaville 1977), the Selma Plateau area (c.1400-1800 m) and the Jebel Bani Jabr (to c.2000 m). All of these areas are carbonate massifs like the Ru'us al-Jibal and Jebel Akhdar. Most of the Eastern Hajar remains remote, especially at higher elevations, and botanical information is limited. The Eastern Hajar is evidently somewhat drier and lacks the abundance and diversity of trees and large shrubs seen in the Jebel Akhdar, although in the most accessible locations, e.g., the Selma Plateau, overgrazing contributes greatly to the impoverished flora.
Floristically, the Eastern Hajar seems to be somewhat intermediate between the Ru'us al-Jibal and the Jebel Akhdar. It has the characteristic Jebel Akhdar tree assemblage of Sideroxylon mascatense and Olea europaea (although Mandaville's photographs suggest this is limited to wadi environments), as well as common Jebel Akhdar plants such as Ebenus stellata, Euryops arabicus, Teucrium mascatense (Mandaville 1977) and Ziziphus hajarensis. However, several characteristic Ru'us al-Jibal species are present in the Eastern Hajar, although they are absent in the Jebel Akhdar (Mandaville 1977) and also absent or rare in the intervening Western Hajar. These include the Arabian almond Prunus arabica, which is common in the Eastern Hajar at a distance of 430 km from its nearest representatives in the Ru'us al-Jibal, and the distinctive spiny Astragalus fasciculifolius, which makes a similar disjunctive appearance in the Eastern Hajar. Also found in the Ru'us al-Jibal and Eastern Hajar, but rare, if not absent, in the Jebel Akhdar are Anthemis odontostephana, Convolvulus acanthocladus and Zoegea purpurea (Mandaville 1977).
What might this imply, if anything, in terms of the biogeographical history of the region? Since it is not intermediate geographically, is the high Eastern Hajar averaging climatic factors? The Eastern Hajar cannot fail to receive the south-west winds of the Indian Ocean monsoon. Why is the result not the same as in the Jebel Akhdar? Is it only the exceptional height of the Jebel Akhdar that accounts for its greater annual rainfall, significant summer precipitation, and as a result, its relatively diverse and abundant flora? Mandaville (1977) has inferred this. If so, might "Almond steppe" (it cannot be called Artemisia steppe because in the Eastern Hajar there is no Artemisia) be simply the "default" mountain vegetation in circumstances of lower rainfall and/or lack of summer precipitation? Examination of rainfall data for high elevation sites in the Eastern Hajar (not to mention additional botanical fieldwork) would help to test these hypotheses.
[FIGURE 3.4.1 OMITTED]
[FIGURE 3.4.2 OMITTED]
[FIGURE 3.4.3 OMITTED]
[FIGURE 3.4.4 OMITTED]
[FIGURE 4.1.1 OMITTED]
[FIGURE 4.1.2 OMITTED]
[FIGURE 4.1.3 OMITTED]
[FIGURE 4.1.4 OMITTED]
[FIGURE 4.1.5 OMITTED]
[FIGURE 4.1.6 OMITTED]
[FIGURE 4.1.7 OMITTED]
[FIGURE 4.1.8 OMITTED]
[FIGURE 4.1.9 OMITTED]
[FIGURE 4.1.10 OMITTED]
[FIGURE 4.1.11 OMITTED]
[FIGURE 4.1.12 OMITTED]
[FIGURE 4.1.13 OMITTED]
[FIGURE 4.1.14 OMITTED]
[FIGURE 4.1.15 OMITTED]
9. The absence of the wild olive (Olea europaea)
One of the most intriguing absentee species in the Ru'us al-Jibal is the wild olive tree, Olea europaea, also discussed in the regional literature as O. africana, O. aucheri, O. chrysophylla and O. ferruginea. Boulos (2000) reasons from the conclusions of Green & Wickens (1989) that the correct name for Asian and African populations should be O. europaea ssp. cuspidata (Wall. ex G. Don) Ciferri. Wild olives are found throughout the mountains of south Arabia, from the 'Asir region of Saudi Arabia through Yemen, the Dhofar region of Oman and northwards into the Jebel Akhdar and the Hajar Mountains of Oman and the UAE. In the Jebel Akhdar, wild olives are a significant element of the vegetation from at least 1100-2500 m and can be found to c.2900 m (see Section 8.3 above and Figs. 4.1.5 to 4.1.12). The wild olive also ranges to the north and east, into southern Iran (Zohary 1963) and Baluchistan (Stewart 1959, cited in Ghazanfar 1991), across Afghanistan and into the mountains of northern Pakistan (Zohary 1963, Mandaville 1977, Boulos 2000, pers. obs.).
Within the ophiolite terrain of the Western Hajar, to the north-west of the Jebel Akhdar, higher elevations are generally very difficult to access and are therefore poorly known, but ascents to a number of summit areas ranging from 900-1800 m between 'Ibri, Oman and Fujairah, UAE confirm not only that wild olives are present in this region, generally beginning at c.800 m, but also that they are the dominant tree species at elevations above c.1000-1200 m and continuing to the local summits. At their northernmost frontier, hundreds of olive trees can be found along the 900-1000 m gabbro ridges of Jebel Qitab and Jebel Sfai, south-west of Fujairah, some 65 km south of the Ru'us al-Jibal, although in that area they are modest in size and are generally restricted to north-facing slopes and ravines (Figs. 5.5.10 and 5.5.11).
The absence of the wild olive in the modern Ru'us alJibal is all the more puzzling in light of the elevation shift observed in vegetation zones between the Jebel Akhdar and the Ru'us al-Jibal, discussed above. If the wild olive zone in the Jebel Akhdar (c.1100-2500 m) were likewise shifted downwards by 300-500 m in the Ru'us al-Jibal, then wild olives would be expected there at elevations above c.700 m and they should be common above c.1000 m. If O. europaea is present in significant numbers in mountains on both sides of the Strait of Hormuz, why is it not also found in the intervening territory of the Ru'us al-Jibal?
The question is obvious; the answer is less so. Given the substantial populations to both the north and south, it is not reasonable to suppose that the wild olive somehow bypassed the Ru'us al-Jibal in the course of its dispersal between Africa and Asia. Therefore two basic possibilities exist, which are briefly discussed below but cannot be definitively resolved by the present study: (1) that wild olives cannot thrive naturally in the Ru'us al-Jibal, at least under current environmental conditions, and that, if once present, they have become extinct locally due to natural causes; or (2) that wild olives were formerly present in the Ru'us al-Jibal but have disappeared due to exploitation by man and his domestic animals. It is also possible that environmental factors and anthropogenic influences have acted in combination to eliminate the wild olive.
9.1. Environmental factors. In comparison with the wild olive stronghold in the Jebel Akhdar, the failure of the wild olive to thrive in the physically and geologically similar environment of the Ru'us al-Jibal can be plausibly explained in terms of differential rainfall, which in the Ru'us al-Jibal is significantly lower and more seasonal (see Section 8.4 above), and probably also somewhat less reliable.
An explanation in terms of differential rainfall patterns is less easy to accept, however, in view of the presence of wild olives in proximity to the Ru'us al-Jibal in the Hajar Mountains south-west of Fujairah. That area is sufficiently close to the Ru'us al-Jibal (c.65 km) that it seems at first unlikely that the local rainfall regime would differ greatly between the two areas, notwithstanding that any peripheral influence of the Indian Ocean monsoon would attenuate in the northerly direction. It is, nevertheless, the impression of a number of UAE-based naturalists that convective summer rainfall is greater in the mountains from Hatta southwards (having regular maximum summit elevations of 1400-1600+ m) than from Hatta to the north (with rare maximum summit elevations of 1100+ m). There is certainly more permanent surface water in the mountains to the south of Hatta.
Alternatively, or in addition to possible rainfall differences, hydrological or edaphic differences may play a determinative role, although it is apparent from the known distribution that the distinction between carbonate and ophiolite substrates is not, per se, a controlling factor for O. europaea. This hypothesis proposes that the ophiolite bedrock is sufficiently more advantageous, in terms of retaining groundwater in the near surface, that under conditions of lower and more seasonal rainfall it can support wild olive growth, even at higher elevations and among steep slopes and gulleys, whereas the carbonate bedrock of the Ru'us alJibal cannot.
In this context it is intriguing to note that an early division of the UAE into bioclimatic zones by Satchell 1978 (reproduced in Boer 1997 at Fig. 6) considered the Hajar Mountains of the UAE to be more "mesic" than the Ru'us al-Jibal to the north, although this classification does not match the reported pattern of isohyets in the mountain regions (Boer 1997, Fig. 2). Using FAO data, Satchell grouped the East Coast with the adjacent Hajar Mountains as "sub-humid", whereas the Ru'us al-Jibal was grouped with the gravel plains to the west of the Hajar Mountains and the Arabian Gulf coast of the Northern Emirates as "arid". (The majority of the UAE, consisting of sand desert, was classified as "hyperarid", while the high peaks southwards from Hatta to Wadi Jizzi were "semi-arid".) Such a classification seems to imply the influence of the Indian Ocean and Gulf of Oman on the climate of the Hajar Mountains and the Gulf of Oman coast, exclusive of the Ru'us al-Jibal.
The current distribution does not exclude the possibility that the wild olive was once present in the Ru'us al-Jibal, and perhaps even widespread, under more favourable conditions, and that it waned and ultimately disappeared entirely in response to deteriorating climate. That is arguably implied by the generally accepted hypothesis that the climate of Eastern Arabia has been more mesic from time to time over the past few thousand to few million years (see Section 1.3 above). There is, however, as yet no direct evidence of the former presence of the wild olive in the Ru'us al-Jibal.
9.2. Anthropogenic influences. Circumstantial evidence exists in local culture for the former presence of wild olives in the Ru'us al-Jibal. The wild olive tree is known to present-day Shihuh tribal residents of the Ru'us Al-Jibal by the name 'itm (or 'utm), the same name used in the Jebel Akhdar. According to Shihuh tradition, the wood of the 'itm protects the bearer against the jinn (evil spirits). This is recited even by younger individuals who admit they have never seen a wild olive tree, and the wood of the 'itm is considered, at least in theory, the material most preferred for making the handle of the trademark Shihuh axe, the jirz (Y. Al-Shihi, pers. comm.). However, despite general statements affirming the presence of olive trees, inquiries to older Shihuh residents of the Ru'us Al-Jibal have not succeeded in identifying specific locations where the wild olive can be found today; instead it is speculated that "only one or two" now exist (Y. Al-Shihi, pers. comm.). Written accounts uniformly refer to the wild almond (Prunus arabica, locally called mizi) as the wood traditionally used for the handle of the jirz (see, e.g., Vincent (1991)).
The existence of such an oral tradition permits the speculation that selective cropping by humans, and/or browsing by their domesticated animals, could have put fatal pressure on the wild olive. The Shihuh, who have traditionally maintained only seasonal settlements in the Ru'us al-Jibal, seem generally to have been adept at conserving scarce resources, at least until the modern era. Among other things, they have selectively preserved and cultivated the edible fig, Ficus johannis, and the sidr tree, Ziziphus spina-christi, valuable both for its edible berries and as a source of lumber through coppicing. Thus it seems at least somewhat unlikely that they would have permitted the elimination of the wild olive, but the talismanic value of the olive wood might conceivably have been allowed to overcome more worldly considerations and the olive, a notoriously slow-growing tree, may have been particularly susceptible to over-exploitation. In this context it should be recalled that Munton, who visited the Ru'us al-Jibal in the late 1970s to investigate the presence of Arabian tahr, speculated from the lack of diversity of tree and shrub species that cutting of live wood has taken place in the past (Munton 1985).
Browsing by feral and domesticated animals does not seem to threaten the present day survival of the wild olive in the Jebel Akhdar, but the problem of feral goats is reckoned to be more pervasive in the Ru'us al-Jibal than in the Jebel Akhdar, due to different customs of husbandry; at least, this is true today (Richard Wood, pers. comm.). Within the wild olive populations nearest the Ru'us al-Jibal, in the UAE and Northern Oman, browsing of accessible branches and shoots is evident and the absence of any very young olive trees is conspicuous.
If humans or their livestock are primarily responsible for the demise of the wild olive in the Ru'us al-Jibal, this must have occurred at a very rapid pace or there must have been very few olive trees present to begin with (or both), since seasonal habitation and cultivation of the Ru'us al-Jibal is estimated from archaeological evidence to date from no more than c.700 years ago (Derek Kennet, pers. comm., 2002). In any case the process must have been singularly effective.
Alternatively, it is possible that the Shihuh belief about olive wood is an example of a tradition that has travelled farther than its cultural and environmental referents, since the Shihuh and other inhabitants of the Northern Oman mountains are generally considered to be historical migrants from the highlands of Yemen, via Dhofar.
10. No inroads by exotic species. No obviously exotic species, invasive or otherwise, were encountered at wild sites within the Ru'us al-Jibal, notwithstanding the large scale introduction of dry-adapted exotic plants for landscaping in the UAE as a whole over the past 30-40 years. This may be an oblique tribute to the rigors of the local environment. Even as cultivated plants within terraced settlements, exotic species (e.g., the neem tree, Melia azederach, and Cordia spp.) are very rare. The most commonly cultivated tree, after the date palm, is the native mountain fig, Ficus johannis.
The introduced mesquite tree, Prosopis juliflora, has spread extensively on the gravel plains south and west of the Ru'us al-Jibal and at Dibba to the south-east. It is also found in the vicinity of some larger coastal settlements along the northwest of the Musandam region and at Khasab in the north. P. juliflora is now considered a very serious invasive species in many countries where it has been introduced, including the Dhofar region of Oman, but eradication efforts have generally not proven successful. It is therefore comforting to be able to state that only a single specimen of P. juliflora is presently known within the territory of the Ru'us al-Jibal, as defined for the purposes of this study (see Checklist).
There is reason to be concerned for the future, however. Since about 2001, possibly in response to the the final agreement of borders between the UAE and Oman within the Ru'us al-Jibal, a number of high terraced settlements have been the site of modest but significant infrastructural improvements such as new stone dwellings, new concrete cisterns, PVC water tanks, improved foot trails, etc., for use by expatriate labourers and recreational visits by local owners. By early 2007, at least one remote high homestead was surrounded by a cement block wall adorned by electric light globes, to be powered by a portable generator. A falcon training camp is said to have been established recently at another isolated location. Resumption of traditional agriculture does not seem to be a significant motive for these developments, although resident labourers sometimes maintain small plots. The current phase of activity commenced and proceeded initially during a period of drought when agriculture was not feasible, and the properties in question had been largely disused and uninhabited during most of the preceding, relatively wet, decade of the 1990s. Since 2005, in addition, vehicle tracks of varying quality have been driven into a few high areas.
In perhaps the greatest potential threat to the high elevation environment of the Ru'us al-Jibal, a major project was proposed for the construction of a decentralised resort community (there was persistent talk of skiing!) in the very highest area of the UAE, consisting of several square kilometres of land rising to c.1800+ m on the south-western slopes of Jebel Jais (whose 1900+ m summit is in Oman) and representing the UAE's only significant area of habitat above 1500 m (see Section 1.5, Section 3 (Zone 3) and Figs. 1.3.7 and 1.3.8). Apart from the direct damage due to habitat destruction, experience suggests that the introduction of exotic flora as landscaping amenities is more likely than not in connection with such a project. That could pose a serious threat to the distinctive high elevation flora, not only in the UAE but in Oman as well, since the summit of Jebel Jais is contiguous with other Omani peaks and plateaux directly to the north and north-east, which represent the very highest area of the Ru'us al-Jibal.
The project in question has become one of many victims of the global economic crisis that commenced in mid-2008, but significant environmental damage has already been done by construction of a major but unfinished access road carved into the eastern slope of the ridge of Jebel Rahabah (1563 m), the UAE's highest peak (Fig. 1.3.9). The road was apparently designed to accommodate the anticipated demand for heavy vehicles for the resort project: it is often impressively wide and is built at an extremely low gradient, so that it snakes back and forth along almost the full length of the ridge in numerous successive shallow switchbacks. Some of these characteristics may also have been recommended from an engineering standpoint, since the east flank of Jebel Rahabah is a dip slope subject to landslides. The overall result, however, is to maximise the environmental impact of the road, which now effectively prevents access by foot over much of the mountainside.
11. The Wadi Khabb Shamsi narrows--a refuge for rare species
Wadi Khabb Shamsi drains the south-eastern Ru'us al-Jibal. It flows for some 25 km from Jebel Qi'wi and Aqabat Oso and debouches at Dibba. About 13 km upstream from the mountain front the wadi narrows and makes a right angle bend in an area where for about half a kilometre the wadi is no more than c.30-40 m wide, and at the bend it constricts to only c.10-12 m (Fig. 1.4.12). Flotsam indicates that, when the wadi is in flood, water depth at the bend can reach 15 m or more. In this area the cliffs rise steeply on both sides (to 1200 m on the south-west and 600 m on the north-east), although they are cut by fractures and gulleys.
In the narrows and its immediate vicinity are found specimens of a number of plant species not known from any other locations within the Ru'us al-Jibal. Some are also rare within the Hajar Mountains generally. Included are the herbs and small shrubs Abutilon fruticosum, Boerhavia diffusa, Commelina albescens, Crotalaria aegyptiaca, Dalechampia scandens, Forskaolea viridis, Pulicaria glutinosa and perhaps Torilis leptophylla, and the large shrubs Ehretia obtusifolia, Grewia tenax and Grewia villosa. Specimens of A. fruticosum, D. scandens, E. obtusifolia, G. tenax (Fig. 5.5.8) and G. villosa are restricted to more or less inaccessible cliff or ledge sites. Pistacia khinjuk is restricted to a single wadi about 2.5 km upstream from the narrows, where it is represented by about two dozen small trees on cliffs along some 2 km of wadi.
It is reasonable to suspect that the presence of these species is related to the relatively sheltered nature of the narrows area, whose physiography provides a certain amount of protection from heat and dessication in addition to grazing pressure. If this is true, then many of the listed species could be relicts of more extensive populations that existed in earlier, more equable times.
12. 'Ayn as-Sih--a unique oasis for hygrophilous species
The spring known as 'Ayn as-Sih is located deep within the gorge of Wadi al-'Ayn, which flows north-east from Jebel Harim to Khasab (Figs. 3.4.1 and 3.4.2). Access to the spring is by foot only and is not straightforward. The site is at c.470 m and the cliffs immediately above rise steeply to more than 800 m. Water seeps from the south-east wall at various points over a length of c.200 m and from a height of c.50-75 m above the wadi bottom, nourishing the coarse rubble of the wadi bank below (Figs. 3.4.3 and 3.4.4).
This is the only natural site known in the Ru'us alJibal for a number of hygrophilous species, including Arundo donax, Epipactis veratrifolia, Nerium oleander, Pteris vittata and Schoenus nigricans, and the only Ru'us al-Jibal site known for self-seeded (feral) date palms Phoenix dactylifera (Figs. 3.4.1 and 3.4.3). It is one of only two Ru'us al-Jibal sites for Portulaca oleracea. Also collected here, by Michael Gallagher in 1982, was the holotype of Echinops atrox, a rare species otherwise known in the region only by a few records from Dhofar (N. Kilian, pers. comm.).
A small number of other springs are found in the Ru'us al Jibal, several in association with 'dry' waterfalls and some modestly improved by man, but none are on the scale of 'Ayn as-Sih.
13. Exceptional records from Jazirat al-Ghanem.
The records of the Mandaville expedition from the island of Jazirat al-Ghanem, off the extreme north-west tip of the Musandam (Mandaville 1985), are both sufficiently numerous and sufficiently exceptional that they have been excluded from the Checklist and, consequently, from the statistics and generalisations based upon it.
The island lies less than 1 km offshore and is elongated parallel to the coast, 5 km long by 1.5 km wide. It reaches a maximum elevation of c.160 m near its south end, so the entire island is lower than the vast majority of the area considered within the present study. Nonetheless, the Jazirat al-Ghanem records deserve brief discussion as a footnote to the observations presented here.
Traditionally the island had been used intermittently by local tribesmen for grazing of goats and sheep. Mandaville (1985) wrote that at the time of his study there were few signs of recent grazing pressure on the vegetation, but expedition member Torben Larsen, writing a number of years later, had a somewhat different recollection: "Until recently this island had been teeming with goats, which had devastated the landscape by eating down all but the hardiest bushes and trees. In the few days available we found little difference in species composition in our various specialties, but I am sure that the quantitative mix must have been different from the mainland" (Larsen 2002).
13.1. Jazirat al-Ghanem records not included in the Checklist. Even after discounting obvious coastal species and low elevation ruderals, the inclusion of the Jazirat al-Ghanem records would have added nine species to the Checklist for which Jazirat al-Ghanem was the only site: Commicarpus squarrosus, Digitaria nodosa, Echiochilon thesigeri, Erodium laciniatum, Melhania muricata, Ochradenus baccatus, Seddera latifolia, Sedum affnanum and Teucrium polium.
All but one of those species are also either absent or very rare in the UAE and Northern Oman. The exception is Erodium laciniatum (Geraniaceae), which, based on its regional distribution, can reasonably be expected to occur at lower elevations in the Ru'us alJibal and has possibly been overlooked. Two of the Jazirat al-Ghanem species, Digitaria nodosa (Poaceae) and Echiochilon thesigeri (Boraginaceae), are known from a small number of Hajar Mountain records and could possibly occur in the Ru'us al-Jibal as well. O. baccatus and T. polium are species better known to the north, in Saudi Arabia, Qatar and Iran, but each has a very similar congener in the Ru'us al-Jibal (O. arabicus and T. stocksianum) and the possibility exists that the determinations in Mandaville (1985) might today be revised. Ghazanfar (2003) does not record O. baccatus from Northern Oman.
C. squarrosus and S. affnanum are otherwise unknown in the UAE and Northern Oman (the Mandaville record of C. squarrosus is apparently distinct from the species now considered as C. mistus by Ghazanfar (2003)). Melhania muricata is otherwise known only from Jebel Qitab, a high ridge south-west of Fujairah, where it is found at elevations above 800 m, and from a few records in the Jebel Akhdar. E. thesigeri is otherwise known only from Jebel Qitab, from Jebel Hafit near Al-Ain, and from further south in Northern Oman. Seddera latifolia is rare in Northern Oman and has only recently been recorded in the UAE, from the base of cliffs along the coastal plain north of Fujairah; the site has recently been destroyed by construction work.
The Mandaville team also collected on Jazirat alGhanem three species for which only a small number of Ru'us al-Jibal records exist: Abutilon fruticosum, Convolvulus virgatus and Hibiscus micranthus. A. fruticosum is known in the Ru'us al-Jibal only from inaccessible specimens in the gorge of Wadi Khabb Shamsi (see Section 11 above) and in the UAE only from a few plants on the hillsides of a single wadi southwest of Fujairah. Convolvulus virgatus is rare in the Ru'us al-Jibal and is found only at scattered locations up to c.1200 m, but grazing is probably a factor since occasional large, healthy plants have been found at inaccessible cliff sites. It is common in the Hajar Mountains to the south, where it is often heavily grazed. H. micranthus is otherwise absent in the Ru'us al-Jibal, except at low elevation in the area of the isthmus (described immediately below and in Section 13.2), but is widespread (although not common) in the Hajar Mountains.
The presence of at least a few of these anomalous species may be related in one way or another to the fact that Jazirat Al-Ghanem had, until the 1970s, supported a British naval facility. The Mandaville team also found Abutilon fruticosum and Hibiscus micranthus at the tiny island site of Jazirat al-Maqlab, an abandoned 19th century telegraph relay station in Khor ash-Shamm. Like Jazirat al-Ghanem, the Jazirat al-Maqlab site was said by Mandaville to appear free of recent grazing by goats, although it still reflected the disturbance of construction and occupation a century before.
13.2. Ru'us al-Jibal species at anomalously low elevations. On Jazirat al-Ghanem the Mandaville expedition also recorded a number of characteristic Ru'us al-Jibal species at unusually low elevations. Convolvulus acanthocladus was collected there at 100 m and all of the following mountain species were collected at 30 m or less: Callipeltis cucullaris, Convolvulus ulicinus, Ephedra pachyclada (as E. intermedia), Grewia erythraea, and Vernonia arabica. Several of these (C. acanthocladus, C. ulicinus and E. pachyclada) are normally found only at medium to high elevations in the Ru'us al-Jibal. The author encountered the same phenomenon on the east coast of the Musandam, not far beyond the isthmus, at a site called Sufayrat, near Ra's Da'aliq, north of the mouth of Khor Habalayn. There, not far from an established homestead and a herd of c.20 goats, Echiochilon persicum, Grewia erythraea, Hibiscus micranthus, Lavandula subnuda, Leucas inflata, Teucrium stocksianum and Vernonia arabica were all found, most of them heavily grazed, at elevations below c.75 m.
Gary M. Brown (pers. comm., 2009) mentions that this phenomenon--the presence of "mountain" species at low elevation in maritime contexts--is not limited to the Musandam peninsula but is a conspicuous feature as well on certain of the UAE's offshore islands (most of them to the west, off the coast of Abu Dhabi), which host a number of species from the Hajar Mountains.
One possible explanation in the isthmian region is that the intermittent temporary introduction and grazing of free ranging livestock from mountain areas has facilitated the introduction of the seeds of certain mountain species. On most of Abu Dhabi's offshore islands, however, there is no historical tradition of temporary grazing of livestock from mountain areas, so it may be that the more important factor in permitting mountain species to become established below their customary elevations is simply the moderating influence of the sea on peak summer temperatures and evaporation of moisture, perhaps coupled with lack of competition from low-elevation species that never happened to reach these remote areas. A clearer case of maritime influence alone may be the abundant presence of Ficus cordata salicifolia, normally a wadi bank species, on slopes overlooking the east coast of the Ru'us al-Jibal, from Dibba northwards to Limah and beyond (Fig. 5.4.13).
14. Field relationships with possible taxonomic implications
In several instances, field observations of distinctive growth forms associated with particular habitats and/or geographically separated populations suggest the possible occurrence of unrecognised species or subspecies. Three of the most persuasive examples are briefly discussed below, in the hope of encouraging definitive expert attention.
Convolvulus acanthocladus. C. acanthocladus is one of the most common plants at higher elevations in the Ru'us al-Jibal, where it is a spiny, tomentose dwarf shrub (Fig. 5.4.6). The same plant occurs sparsely on the highest ridges (900-1000 m) some 65 km south of the Ru'us al-Jibal, but otherwise is absent for ca.135 km to the south.
According to expert determinations, C. acanthocladus reappears on the west flank of the Hajar Mountains in the Mahdhah area, but there it is a shrub of lower slopes and gravel terraces along the mountain front (at c.600 m) and has a decidedly more rectilinear habit (Fig. 5.5.4). The latter form can be seen southwards to the Ibri area, and also in the upper Wadi Hawasina watershed, which drains the east flank of the Hajar Mountains, north-east of Ibri.
The tomentose form is also found still further south, but only as a very rare plant, in the vicinity of Jebel Kawr, near Jebel Shams in the Jebel Akhdar, at c.1200 m. Ghazanfar has collected C. acanthocladus from Jebel Shams itself (Ghazanfar 1992a), apparently at an elevation above 1800 m but otherwise without description. It is evidently rare in that area as well, as the author has never encountered it there (and has recorded only one, heavily grazed, specimen of the similar C. ulicinus).
Plants having an intermediate morphology have not been observed. Possibly it is significant is that, from the above records, the two different growth forms appear to be correlated with differences in both elevation and geology: (i) high elevation and carbonate rocks in the Ru'us al-Jibal, Jebel Kawr and Jebel Shams versus (ii) medium elevation and ophiolite rocks in the intervening Hajar Mountain foothills.
Diplotaxis harra. In the Ru'us al-Jibal, the plant recognised as D. harra is a rock and cliff-dwelling perennial dwarf shrub found at medium to high elevations, where it grows to a height of up to 50 cm (Fig. 5.4.8). The same form is found in the high Jebel Akhdar, to at least 2300 m. In most of the intervening Hajar Mountains, however, D. harra is a much smaller, generally single-stemmed annual plant growing abundantly on gravel outwash plains or on gravel terraces within the mountains (Fig. 5.5.6). As in the case of Convolvulus acanthocladus, the different growth forms are correlated with differences in both elevation and geology. Miller and Cope (1996) state that plants from the high mountains of Oman (3000 m) (which indicates specimens from the Jebel Akhdar) are glabrous and approach D. kohlaanensis. The plants in the Ru'us al-Jibal are, however, hirsute.
Dodonaea viscosa. A number of observers have commented on the apparently bimodal habitat distribution of D. viscosa in the UAE and Oman, although casual field observations have not so far recognised any consistent morphological differences. At lower elevations it is found along wadi banks, but at higher elevations it is locally common on bedrock plateaux, often in small silty depressions (see, e.g., Figs. 3.1.3, 3.1.8 and 5.4.9). It occurs at elevations up to 3000 m in the Jebel Akhdar, where above c.2400 m it is one of the more common large plants, along with the juniper tree Juniperus excelsa polycarpos (see, e.g., Figs. 4.1.1 and 4.1.4). The taxonomy of Dodonaea is problematic (Leenhouts 1983; Curtis 1999) and these field relationships should be taken into account in any definitive resolution.
This paper is the product of investigations undertaken over many years and there are correspondingly many thanks and acknowledgements to be conveyed. A.R. (Rob) Western's 1989 book, The Flora of the United Arab Emirates: An Introduction, first made it practical for non-professional observers to participate in the study of the UAE's plants and their geographic distribution, a study which the author took up enthusiastically in connection with more general exploration of the UAE and its diverse environments. Rob was also a patient correspondent in the author's early years of botanical observation, when Rob himself was still engaged in field work in the UAE.
Marijcke Jongbloed, a physician by profession, is the author of books on many aspects of UAE natural history, and an early and ground-breaking UAE conservationist as well, but her first love in the natural world was plants, culminating in her 2003 book, The Comprehensive Guide to the Wild Flowers of the United Arab Emirates. Conversation and correspondence with Marijcke over the years enlarged the author's knowledge of UAE plants generally, as did many hours spent with the herbarium that she established at the Sharjah Natural History Museum. Consultation with her in connection with the aforementioned volume, in particular in relation to mountain plants, sharpened the author's awareness of the distinctive character of the flora of the Ru'us al-Jibal.
Prof. Loutfy Boulos was introduced to the author through Dr. Jongbloed and was diligent and enthusiastic over many years in providing, to Dr. Jongbloed and later to the author, expert identification of specimens, a crucial step in transforming field information into knowledge.
Dr. Norbert Kilian reviewed and commented on the taxonomy of the Asteraceae included in the Checklist, and was generous in providing information and encouragement. Prof. Hildemar Scholz graciously assisted at a late stage of this study to improve the coverage of grass species. Dr. Shahina Ghazanfar provided numerous reprints and was patient in responding to inquiries over several years. Her draft Red Data List for the UAE (Ghazanfar, in prep.) was the primary basis for updating the scientific nomenclature used in Jongbloed (2003).
One reason for the dearth of attention given to the flora of the Ru'us al-Jibal is its general inaccessibility other than by foot. The author has overcome this obstacle by means of regular hiking visits over more than 19 years in company with a number of indulgent, engaging and insightful field companions who merit recognition and thanks here. They include (in rough chronological order of their participation): Charles S. Laubach, the late Martin Parker, Ian Lapraik, Angela and Stephen Manthorpe, Cheryl Jones, Ian Robson, Yusuf Al-Mazroui Al-Aqabi Al-Shihi, Neil Curtis, John Martin, Barbara Couldrey, Geoff Cosson, Peter L. Cunningham, David Palmer, Richard Morris, Narayan Karki (who came to know the plants of the Ru'us al-Jibal almost as well as the author) and Ramesh Bhandari.
John Martin tracked down and provided copies of the works of Michael Zohary cited in this paper.
Gary M. Brown, Richard J. Hornby and Marijcke Jongbloed paid the author's efforts the compliment of reading with care a late draft of the entire manuscript and providing constructive comments and additional information.
Finally, thanks are due to the publisher, editors and sponsors of Tribulus, who have, steadfastly and with great foresight and energy, encouraged serious amateur natural history research in the UAE and neighbouring Oman and provided a forum for its dissemination and discussion.
A note about grass species numbers
The figure of 45+ given in the main text for grass species (Poaceae) found in the Ru'us al-Jibal is undoubtedly somewhat understated due to lack of comprehensive and authoritative attention. It is reasonable to expect that the actual number of grasses could exceed fifty (50).
Nine (9) grass species that have been recorded from Musandam sites at low elevation (Mandaville 1985; Ghazanfar 1992a), outside the Ru'us al-Jibal as defined here, are mentioned in the Checklist but are disregarded for statistical purposes. However, some of those could possibly be present within the mountain environment as well, particularly in the lower reaches of Wadi Bih or in cultivated fields. The effort was made in winter and spring 2008-09, and again in spring and summer 2010, to devote particular attention to grasses in order to enhance the completeness of the Checklist. Hildemar Scholz kindly provided timely expert determinations. That effort seconded a number of Ru'us alJibal records by others, re-confirmed several single records by the author, and confirmed the presence of several UAE/Oman grass species that had not yet been reported from the Ru'us al-Jibal but were expected to occur there (e.g., Hyparrhenia hirta and Tricholaena teneriffae). It also added five identified grass species not previously expected, including three new to the region. Several grasses known only from photographic records could not be positively identified, but are evidently not among the species mentioned in Jongbloed (2003). From this experience it is certain that a number of additional grasses must remain to be recognised.
A note about photosynthesis, palaeobotany and palaeoclimatology
The terms C3 and C4 refer to two different metabolic pathways by which plants photosynthesise. C3 plants are the most common and include most trees, shrubs and herbs, and some grasses. C4 plants are mostly grasses and are associated with warm, dry, tropical and subtropical environments with higher light levels. C4 plants have mechanisms for water conservation and use carbon dioxide more effectively. In sedimentary deposits, the remains of C3 and C4 vegetation can be distinguished by their different and non-overlapping carbon isotope ratios. Phytoliths (microscopic biogenic siliceous structures found in the tissues of many plants, including most grasses) can also be recovered from sediment and can be used to distinguish the types of grasses formerly present. Different groups can be recognised that are indicative of different climatic conditions, e.g., C3 Pooidae (cool wet), C3-C4 Panicoideae (warm humid) and C4 Chloridoideae (warm dry). See Parker et al. (2004).
A note about the terraced settlements of the Ru'us al-Jibal
Some hundred or more modest to small-sized areas of terraced agriculture are scattered throughout the Ru'us al-Jibal, at elevations from c.450-1500 m. The very largest are c.0.25 sq. km. in area and may contain c.80-100 traditional stone dwellings of various ages and in various states of repair. They were established principally for the cultivation of winter wheat but also supported crops such as date palms, watermelon and the fruit of the native sidrtree, Ziziphus spina-christi, which is also coppiced for lumber. With few exceptions, only date palms have continued to be cultivated into the modern era; the largest settlements may have as many as c.100 palm trees. However, one large settlement was recalled rhapsodically to the author by an elderly resident of another settlement, nearly a day's walk distant: "Ah, Sharmilah . . . two hundred sidr trees!"
Evidence from pottery remains indicates that the earliest terraced settlements were established 600-700 years ago (Derek Kennet, pers. comm. 2002) and a 14th century Persian source makes reference to wheat from the Musandam region (William Lancaster, pers. comm., 2000). Historically, this corresponded with the ascendency of the kingdom of Hormuz, whose dominion included the coastal city of Julfar, just north of present-day Ra's al-Khaimah, the first urbanisation along the coast of what is today the UAE. The impetus for the development of the mountain hinterland was evidently the economic and demographic expansion of the local region at that time (Derek Kennet, pers. comm. 2002).
However, the construction of the terraces and the conduct of extensive seasonal agriculture also seem to reflect a climatic regime somewhat wetter than the present one, which may have been a necessary condition for successful agricultural development. This emphasises the possible importance of an additional correspondence--a climatic one--with the so-called Little Ice Age (c.1300-1850), which has been recognised on both sides of the North Atlantic and elsewhere and was responsible for, among other things, the abandonment of Viking settlements in Greenland, heavy winter snows in colonial Virginia and early Victorian England, and frozen canals in the Netherlands. Such a correspondence might account as well for the apparent decline of many Ru'us al-Jibal settlements even before the modern era, although other and later factors have also been blamed, including specifically the advent of cheap wheat on the world market during and after WWII, which made traditional wheat cultivation in the Ru'us al-Jibal unprofitable (William Lancaster, pers. comm., 2000).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||p. 49-96|
|Author:||Feulner, Gary R.|
|Date:||Jan 1, 2011|
|Previous Article:||The flora of the Ru'us al-Jibal--the mountains of the Musandam Peninsula: an annotated checklist and selected observations.|
|Next Article:||The flora of the Ru'us al-Jibal--the mountains of the Musandam Peninsula: an annotated checklist and selected observations.|