F. Fromard and T. Gauquelin
François Fromard is Research Officer at the Institut de la Carte Internationale de la Végétation, Université Paul Sabatier.
Thierry Gauquelin is a lecturer at the Laboratoire de Botanique et Biogéographie, Universite Paul Sabatier, Toulouse, France.
The thuriferous juniper (Juniperus thurifera) stands in the highlands of the Mediterranean basin are heavily degraded as a result of intensive wood removal and livestock activity in these densely populated areas, This longstanding situation, which will soon become irreversible unless remedial measures are urgently taken, has produced impoverished soils and hillside instability while contributing to desertification. Focusing on the Atlas Mountains of Morocco, this article examines existing knowledge of the thuriferous juniper ecosystem and suggests a number of research orientations and conservation measures.
FIGURE. Geographic distribution of Juniperus thurifera L
Thuriferous juniper is only found in isolated parts of the western Mediterranean (see Map): France and Spain in Europe and Algeria and Morocco in North Africa, each with two distinct subspecies which have recently been described (Gauquelin and Dagnac, 1988).
A botanical rarity in France, thuriferous juniper is found only in limited numbers in three specific areas: the Corsican highlands, where soliu (its local name) has always been known to shepherds of the rare valleys in which it grows (Asco, Niolu); the Alps (Saint Crépin, Mercantour, etc.) where the stands are more extensive and not degraded; and the Pyrenees where only one location has been identified so far (Montagne de Rié, Haute-Garonne).
There are larger and more varied stands in Spain (between the altitudes 350 m and 1300 m), but there are indications that the present sabina vera (thuriferous juniper) area is smaller than it used to be. In the Ebro basin, for example, remaining stands are fragmentary, are missing the other plant species and soil characteristics normally found in thuriferous stands, and still bear logging scars. They have disappeared near urban areas (Zaragoza) where they were abundant until the eighteenth century (Braun-Blanquet and de Bolos, 1957).
In Algeria the tree is strictly limited to the Aurès Mountains (Djbel Chélia, 2300 m) with a number of scattered and often very large specimens in the cedar plantations, which are probably the remnants of more extensive stands.
Finally, in Morocco there are still extensive, although extremely degraded, thuriferous juniper stands throughout the Atlas Mountains, except in the Seskaoua area, the western foothills and the Rif.
Thuriferous juniper, associated with semi-arid and subhumid cold winter bioclimates, is the hardiest forest species under the very severe climate of the high Moroccan mountain areas. It is found at the tree line far more often than the cedar, and isolated specimens can be found up to 3000 m (the lowest-level stands are at 1700 m in the western High Atlas).
Regardless of the physical and chemical composition of the soil, the species grows in low-density open woodland (average cover: 25 percent with 40 to 50 trees/ha), with fairly old trees (probably several hundred years of age) which are often scarred and sometimes measure up to 5 m in diameter. The stands show considerable structural and floral diversity and, in most cases, the thuriferous juniper is the only tree species present. Prickly juniper (J. oxycedrus) and Phoenician juniper (J. phoenicea) may be present in isolation, as may round-leaved oak (Quercus rotundifolia) and cedar (Cedrus atlantica). The shrub strata comprise species of the genus Fabaceae or thorny xerophyte clumps in the more degraded and higher slopes.
Barbero et al. (1990) estimate the thuriferous juniper climax area to be 327000 ha and the real surface area to be only 20000 ha. Boudy (1958) estimated that thuriferous juniper covered about 30000 ha in Morocco (Middle Atlas: 16000 ha, Higher Atlas: 15000 ha) and, in 1938, Emberger referred to the terrible degradation of the stands which, according to his estimates, then covered about 50000 ha - far less than the area lost. Thus, with a shrinkage of some 90 percent, the thuriferous juniper would appear to be the forest species that has declined the most in Morocco.
The massive degradation of the species is mainly a result of the limited availability of wood resources in the High Atlas valleys, the sizeable human population in these areas where there is no alternative source of fuel, the multiple uses of thuriferous juniper wood and the slow growth of the species. In 1979 Dresch stressed that the High Atlas was a very unusual North African example of a densely populated mountain area (Dresch, 1979). In 1975 Despois and Raynal were surprised at the population densities (which sometimes exceeded the Moroccan average), given the difficult environment (Despois and Raynal, 1975). They also indicated that population saturation in the valleys was less likely to be offset by emigration than, for example, in the Anti-Atlas.
There are still large numbers of villages in the high valleys of the Azzaden, N'Fis, Ourika and Ait Bou Gmez, often at altitudes of more than 2000 m. The population in these areas is estimated to have increased by 30 percent over a period of 20 years, while the local economy has remained relatively unchanged and livestock husbandry has intensified (Dresch, 1979). Ouhammou (1986) calculated that the Ourika Valley population increased by more than 30 percent between 1971 and 1982.
A multiple-use species
There is considerable historical evidence of the use of thuriferous juniper in the Moroccan Atlas Mountains. As early as 1938, in a number of his works on Moroccan vegetation, Emberger underlined the social role played by this tree in the high mountain areas, indicating that its wood served for heating and cooking while its leaves fed the livestock. In 1958 Metro noted that thuriferous juniper stands were often lopped by shepherds for their herds in times of snow, making their pared silhouette a characteristic feature of the Moroccan Atlas Mountains.
The trees are still lopped for fodder today, and cut and abandoned branches are often to be seen on the ground after the passage of a herd. Apart from its foliage, the species is also sought for its very hard wood which is used in construction for components such as window and door lintels and roof edges.
Another traditional use is the production of a type of tar, obtained through the dry distillation of large quantities of wood and used, as is cede oil from the prickly juniper, in veterinary medicine (serving as a cicatrizant and antiseptic). Thuriferous juniper has been used as one of the main ingredients of veterinary medicine, and distillation points can still be seen today in the thuriferous juniper areas of Zaouia Ahansal.
Impact on sand intrusion south of the Atlas Mountains
Intensive exploitation by increasing numbers of inhabitants with larger herds inevitably lead to stand degradation and valley soil and slope erosion, particularly because of the slow growth and difficult recovery of thuriferous juniper. Apart from obvious local effects, the erosion process may also have a downstream impact, particularly to the south of the Atlas Mountains. A short brochure, published by FAO (Dembner, 1987), notes that the heavier mineral particles responsible for sand intrusion in these areas do not come from the Sahara as previously thought but mainly from the deforested Atlas slopes. These coarse sands are swept downstream where they are taken up by the winds and mixed with the finer Saharan particles.
Coude-Gaussen's work (1990) on the dust-producing regions of the Sahara led to the classification of the potential of substrata to produce fine particles that are movable by erosive agents. The maps depicting the classification indicate that the North African mountain areas are unlikely to produce "dust", as the thicker plant cover provides effective protection. However, as vegetation disappears, the fine particles, liberated by frost and slope dynamics, clearly become more mobile and flow downstream in water runoff. They then form part of the "river deposit and distribution category" the fine alluvium of which is prone to wind removal and is, therefore, a primary source of dust.
Pursuing previous works on the vegetation of the High Atlas (Peyre, 1983; Alifriqui, 1986; Haloui, 1986; Ouhammou, 1986), research on the thuriferous juniper has been undertaken at the University of Cadi Ayyad in Marrakech (Gauquelin, 1988; Badri, 1990). The emphasis is on the hitherto poorly understood ecological requirements of thuriferous juniper and research activities are based on an extensive study of this plant species in Morocco and on protected plots (Oukaimeden, at 2450 m in the High Atlas Mountains of Marrakech, and the Azzaden valleys, for instance).
Influence of thuriferous juniper on soil microstructure
An important aspect explored by Gauquelin and Dagnac (1988) is that of the soil microstructure which forms under juniper cover. The authors have shown, with soil profiles of various High Atlas stands, that the areas under thuriferous juniper are relatively rich in organic carbon (300 percent in the 5 to 10 cm soil horizons, for example) and minerals (with double the normal content of calcium and magnesium in the 0 to 10 cm layer). The clay content is about 25 percent lower away from the tree cover and the percentage of fine soil (less than 20 mm deep) almost 30 percent lower.
These figures show the dual role of thuriferous juniper, i.e. the development of a mineral-rich humus layer and the protection of fine soil particles against removal by rainfall. This enriched and stabilized substratum, in which the morphological, pedological and genetic processes are much slower than in areas without juniper cover, also appears to be the only area where (rare) natural thuriferous juniper regeneration occurs. Finally, as the phenomenon of soil microstructure is closely linked to stand vitality, a soil survey can serve as a good indicator of stand structure and dynamism.
The water cycle
With regard to the surface soil horizons, through its cover and humus the thuriferous juniper acts as a moderator of water loss by increasing infiltration, reducing runoff and evaporation and thereby postponing or shortening the dry period in summer. At greater depths, reactions are more complex and vary according to the intensity and time of rainfall, with re-wetting often being delayed under tree cover (Gauquelin and Savoie, in press). These features are linked to the interception of part of the rainfall by
the tree leaves, dripping and runoff over an extended period of time. Re-wetting varied according to season and rain intensity but averaged 44 percent on plots with a density of 190 trees/ha (48 percent cover), which is admittedly relatively high compared with the thuriferous juniper average. Nonetheless, this result, which is similar to those found in temperate forests, is indicative of the ecosystem's potential role in the water cycle and of the role that the Moroccan thuriferous juniper must generally have played before its widespread degradation.
Mineral balances
Mineral balances were examined in the various sections of the vegetative cycle: living plant matter, the mineral composition of which varies with phenology; leaf litter, from which soil inputs follow a seasonal cycle; rain and leached particles (analysed in strictly quantitative terms). Annual leaf litter production averaged 11 tonnes per covered hectare, that is 5.3 t/ha/year if considered as actual stand surface area (with a plot coverage of 48 percent). These values are comparable to those found in other types of productive forest ecosystems studied such as Scotch pine (Aussenac and Boulangeat, 1980) and Mediterranean oak (Selmi, 1985).
Nutrients returned to the soil amounted to 330 kg/covered ha/yr (in practice, 160 kg/ha/yr) of nutrients (mainly calcium and nitrogen) and 17.7 kg/covered ha/yr (8.5 kg/ha/yr) of trace elements (iron and manganese). Rain and rain-leaching also contributed to the enrichment of the substratum but to a lesser extent.
Therefore, at a low level of tree mutilation and stand degradation, and assuming equal coverage, the thuriferous juniper ecosystem is totally comparable with, and at times more productive than, that of other Mediterranean or temperate Atlantic forest ecosystems.
The preceding discussion underlines the importance of the thuriferous juniper in Moroccan mountain areas; the role that the species could play in the forest dynamics of the high valleys (not only in Morocco but throughout the Mediterranean basin) if it were not subjected to human-caused stress; and, finally, the ecological and economic damage that would result should the species disappear.
The challenge is to find a way to avoid this local extinction and to maintain and rehabilitate existing stands, primarily by limiting wood removal and regulating grazing at sustainable levels without, however, negatively affecting the day-today life of the inhabitants.
Research needs
Our knowledge of the area under thuriferous juniper in Morocco is still limited, as is our knowledge of thuriferous typology, structure and degradation. High-definition satellite imagery, combined with existing regional maps, aerial photographs and ground truthing, could be of great help in producing a comprehensive, updated map of thuriferous vegetation. Although there are methodological problems, particularly in using satellite imagery for mountainous areas with a limited plant cover, work undertaken in comparable areas in of structure (Lacaze and Joffre, 1990) and satellite imagery experience in the Pyrenees indicates that this approach is feasible.
The wood productivity potential of thuriferous juniper is poorly understood. Trial plots are therefore needed to acquire these growth and yield data in a systematic manner, with subsequent extrapolation to the various types of thuriferous juniper stands identified. It would also be useful to assess the volume of wood cut by the high valley inhabitants. Field surveys are also necessary to help determine the ratio between actual wood consumption and wood that is cut but left lying on the ground.
One important, unresolved problem concerns the extreme scarcity of natural regeneration in thuriferous juniper stands and the great difficulty in germinating seed under trial conditions. Studies already initiated at the University of Marrakech should be expanded, both in the laboratory (to solve physiological problems related to seed dormancy) and in the field (the role of grazing in the elimination of seedlings, the influence of microclimatic conditions on their development and maintenance).
Protected research plots are indispensable for these studies and would enable us to enhance our understanding of the ecological requirements and physiology of the species; of the bio-geochemical and water cycles within the stands; and of the dynamics of the ecosystem, when protected from livestock. A comparative study of erosion in protected and unprotected areas should also be undertaken to pursue the findings reported by Dembner (1987) and Coude-Gaussen (1990).
Finally, we also need to consider the possibility of disturbance to thuriferous juniper forests by regional weather changes, whether brief climatic events or long-term global changes. The fact that Morocco (and even more so, the Moroccan Atlas Mountains) straddles two very different climatic zones, one humid and one arid, may reinforce the importance of such changes (Naciri, 1990). A dendrological study of thuriferous juniper forests is needed to identify the climatic fluctuations in the high Moroccan mountain area and to help determine the age of current stands for which available data are incomplete and approximate.
It would also be useful to coordinate this research with work currently being conducted on J. procera (Blot, 1991), an East African species which is found from Djibouti in Ethiopia to as far as the United Republic of Tanzania. J. procera is botanically similar to J. thurifera and also faces similar problems of degradation.
Conservation and use of thuriferous juniper
The scientific studies proposed above are only conceivable if accompanied by measures to conserve these ecosystems and sensitize the local population to this need. As early as 1958, Boudy argued that juniper stands should be treated exclusively as protection forest; that recovery would depend on the use of protected areas set aside especially for this purpose over a considerable period of time; and that trees for pollarding and lopping should lie outside these areas. He also called for a ban on all forms of wood harvesting, maintaining that the pole and fuelwood needs of mountain populations should be met from other sources.
All protection activities should be carried out in collaboration with local mountain populations. However, this will only be possible if there are wood gathering and grazing possibilities outside the protected areas. The spectacular development of grass cover after only one year of protection on small trial plots should make a strong impression on local populations.
There is no doubt that the inhabitants of the Atlas valleys recognize the value of the thuriferous juniper tree and are also aware of its degradation, although they are still driven by economic necessity to exert considerable pressure on the resource. It is also clear that a total ban on thuriferous juniper wood removal is not possible. A selective management system that includes wood harvesting and grazing, similar to that in place for holm oak in certain valleys (e.g. the Valley of Ait Bou Gmez) introduced by the inhabitants themselves could enhance the sustainable management of thuriferous juniper stands. Such a protection system is, in fact, already being implemented in some thuriferous stands, including that of d'Imelhas (Valley of Ait Bou Gmez) where the local authorities only permit wood removal once a year. It is noteworthy that this protected and managed stand is one of the few that present a relatively balanced vegetation structure where natural regeneration can be observed (Gauquelin, 1988). In their work on the habitat of these High Atlas valleys, Huet and Lamazou (1988) mentioned that harvesting was severely regulated in an area where wood was becoming scarce and where the forest was constantly receding. Each village had access to two distinct reserved forest sections: one managed for fuelwood and the other treated as a "no grazing" agoudal area where the trees were managed and selectively harvested for the production of construction timber. Juniper, arborvitae and holm oak produced the poles needed for floors and terraces.
In summary, a forest management system that enables the stands not only to survive but also to regenerate while guaranteeing the traditional user rights of the local population is the ideal solution. Field surveys are urgently needed on fuelwood and timber needs, the present role of thuriferous juniper trees in meeting these needs, pastoral practices and the role of the species in veterinary medicine. At present there is no quantitative information on these aspects of the resource.
Finally, although the technique is often criticized for its excessive use of inappropriate exotic species, reforestation should be considered on the less arid slopes of the current thuriferous juniper area; for example, using the cedar which is a very well-adapted, alternative local species.
Alifriqui, M. 1986. Contribution d l'étude du milieu et de la végétation dans le. Haut Atlas occidental d'Amizmiz (Massif de l'Erdouz, Igdat). Thesis. Marrakech, Université Cadi Ayyad.
Aussenac, G. & Boulangeat, C. 1980. Interception des précipitations et évapotranspiration réelle dans des peuplements de feuillus (Fagus sylvatica L.) et résineux (Pseudotsuga menziesii (Mirb.) Franco). Ann. Sci. Forest., 37(2): 91-107.
Badri, W. 1990. Cycle hydrologique et biogéochimique et influence du couvert sur la strate herbacée dans un peuplement à genévrier thurifère du Haut Atlas de Marrakech (Maroc). Thesis. Marrakech, Université Cadi Ayyad.
Barbero, M., Quezel, P. & Loisel, R. 1990. Les apports de la phytoécologie dans l'interprétation des changements et perturbations induits par l'homme sur les écosystèmes forestiers méditerranéens. Forêts méditer., XII: 193-215.
Benabid, A. 1987. Les grands écosystèmes terrestres et leur préservation. In M. Fennane, ed. La grande encyclopédie du Maroc, flore et végétation. Rabat, G.E.I.
Blot, J. 1991. Bilan écologique des forêts à Juniperus procera Hochst. ex Endl. de la corne orientale d'Afrique et de la péninsule arabique. Thesis. University of Bordeaux III. (Unpubl.)
Boudy, P. 1958. Economie forestière nord-africaine, I,II,III. Paris, Larose.
Braun-Blanquet, J. & de Bolos, O. 1957. Les groupements végétaux du bassin moyen de l'Ebre et leur dynamisme. An. Estación Experimental Aula Dei, 5(1-4): 266 p.
Coude-Gaussen, G. 1990. Les régions sources de poussières au Sahara. Sécheresse, Science et Changements planétaires, 1(2): 134-141.
Dembner, S.A. 1987. Stopping the sands in Morocco and beyond. FAO Forestry Department project profile. Rome, FAO.
Despois, J. & Raynal, R. 1975. Géographie de l'Afrique du Nord-Ouest. Paris, Payot.
Dresch, J. 1979. Un géographe au déclin des empires. Ivrea, Italy.
Emberger, L. 1938. Les arbres du Maroc et comment les reconnaître. Paris, Larose.
Emberger, L. 1939. Aperçu général sur la végétation du Maroc. Commentaire de la carte phytogéographique du Maroc au 500000e. Bern, Veroffentlichungen Geobotanischen Institutes Rübel.
Gauquelin, T. 1988. Dynamique de la végétation et des formations superficielles dans les montagnes du bassin occidental de la Méditerranée: étude des formations d genévrier thurifère et à xérophytes épineuses en coussinet des Atlas marocains. Thesis. Toulouse, Université Paul Sabatier.
Gauquelin, T. & Dagnac, J. 1988. Caractéristiques édaphiques des groupements a genévrier thurifère des Atlas marocains: étude des niveaux superficiels des sols sous couvert et hors couvert.. Oecol. mediterranea, IV(314): 43-56.
Gauquelin, T. & Savoie, J.M. Evolution de l'humidité du sol en relation avec le couvert arboré dans un écosystème à Juniperus thurifera L. du Haut Atlas de Marrakech (Maroc). Proc. Ecologie et Biogéographie alpines, La Thuile, September 1990. Neuchâtel, Institut de Botanique. (In press)
Haloui, B. 1986. Contribution a l'étude du milieu et de la végétation du bassin versant de l'oued Rhighaya (Haut Atlas de Marrakech). Thesis. Marrakech, Université Cadi Ayyad.
Huet, K & Lamazou, T. 1988. Sous les toits de terre. Belvisi Al Madariss, ed. Publi-Action. Casablanca, Belvisi Al Madariss.
Lacaze, B. & Joffre, R. Evaluation de la densité des arbres en zone semi-aride a partir de données SPOT panchromatiques. Paper presented at Les Journées Scientifiques de Télédétection de l'AUPELF-UREF, Toulouse 13-16 November 1990. London, John Libbey Eurotext. (In press)
Metro, A. 1958. Forêts. Atlas du Maroc. Explanatory notes, section VI, Bibliography. Rabat, Comité de Géographie du Maroc.
Naciri, M. 1990. Calamités naturelles et fatalité historique. Sécheresse, Science et Changements planétaires, 1(1): 11-16.
Ouhammou, A. 1986. Recherches sur l'étagement de la végétation dans le bassin versant de l'oued Ourika (Haut Atlas, Maroc). Thesis. Marrakech, Université Cadi Ayyad.
Peyre, C. 1983. Etagement de la végétation et gradients climatiques dans le système atlasique marocain. Le bassin de l'oued Rdat et le versant sud de l'Atlas au méridien du Tizi-n-Tichka. Bull. Fac. Sci. Marrakech, 2: 87-139.
Selmi, R. 1985. Différenciation des sols et fonctionnement des écosystèmes forestiers sur grès numidien de Kroumirie (Tunisia). Ecologie de la subéraie zéenaie. Thesis. Vandoeuvre, Université de Nancy I.
