Version classiqueVersion mobile
OpenEdition Books

The Mediterranean region under climate change

Jean-Paul Moatti
Stéphane Thiébault

Part 3. Adaptation, resilience, conservation of resources and prevention of risk

Sub-chapter 3.3.1. Grafted oleaster-olive agrosylvopastoral systems in Northern Morocco

Yildiz Aumeeruddy-Thomas, Alicia Bailly, Samuel Alleaume et Younes Hmimsa

Texte intégral

Oleaster and olive, a long history of man-forest interactions

1An average of 2.5 million tons of olive oil is produced annually at global level (COI 2015). Mediterranean countries from the EU and from southern and eastern Mediterranean regions still represent the major exporters. The amounts consumed in source areas are difficult to evaluate, although olive oil is a major element of local diets and health. Understanding the historical and cultural bonds and associated empirical knowledge developed over millennia between Mediterranean societies and oleaster (Olea europaea, var. sylvestris) and its domesticated form, olive (Olea europaea, var. europaea), is crucial to understanding the present and anticipating future scenarios.

2The olive tree and its wild form, the oleaster, are indispensable companions of Mediterranean societies both for their high symbolism (e.g. Athena’s olive tree associated with the foundation of Athens) and for material purposes for producing oil for light, food, esthetic purposes, medicine, fuel and construction wood (Kaniewsky et al. 2012). Oleasters were abundantly used in the Levant well before the advent of agriculture 19,000 years ago (Kislev et al. 1992). During the Neolithic, human groups in the Mediterranean developed swidden agriculture based on the use of fire and long forest fallows for cereal cultivation and pastoral activities around 8000 – 7000 BP. In Sicily, Tinner et al. (2009) identify an expansion of Olea trees 7300-7000 BP, with a parallel increase in Cerealia-Rumex, Chamaerops, Urtica and Ficus carica and to the expansion of agriculture. This is potentially a form of proto-culture of Olea, as was also suggested by Terral and Arnold-Simard (1996) in Spain. Although tree domestication occurred after the domestication of cereals (Zohary and Spiegel Roy 1975), incipient forms of tree domestication were probably present very early in prehistorical times as well as being an ongoing contemporaneous practice (Aumeeruddy-Thomas et al. 2014).

3Although olive domestication and cultivation was initiated in the East (Besnard et al. 2013), multiple domestication sites are found throughout the Mediterranean basin (Terral et al. 2004). The olive may have known different forms of exploitations and it has been suggested by Camps (1961), who studied Berber groups in northern Africa, that grafting oleasters was used extensively by Berbers before they were colonized by the Romans.


4In this paper we compare two distinct olive-based agroecosystems, their social-ecological characteristics and importance for food security. These are: 1) Historical grafted oleaster-olive agroecosystems found throughout the Maghreb, based on a case study conducted in northern Morocco in a sampled area in Ouezzane Province, 2) Modern olive plantations using microcuttings raised in nurseries and a range of techniques favored by new agrarian policies in the same zone. We analyze practices involved in their development, their distribution and spatial organization, diversity of food products; environmental factors including the topographic wetness index (TWI) based on a field numeric model (SRTM 2014) and the topographic position index (TPI).

5The Rif borders the Mediterranean Sea to the north and is a region crossed by a mountain range that stretches from the Tangerine peninsula (west) to the frontier of Algeria (east). It is inhabited by three socio-linguistic groups, the Jbala (arabophone), the Ghomara and Berber Rifians (berberophone).

Grafted oleaster-olive agroecosystems

6Grafting is absent today in Northern and Eastern Mediterranean countries except in some islands such as Corsica, Sardinia, and Lesbos (Greece) where there are remnants of grafted olive orchards. The Romans are known to have started large scale olive plantations and developed plant breeding in nurseries (Brun 1996), an approach which may have eliminated grafting on wild trees. In the Rif, grafting oleasters was still very important in the 1970s (Fay 2015) and our observations show that farmers still graft wild oleasters in frontier areas between the forest and agroforests. If oleasters are not available, “feral” oleasters or escapees from cultivation are saved and transplanted, until they reach the size required to be grafted (Aumeeruddy-Thomas et al. 2014).

7Farmers interviewed in this region unanimously consider that grafted oleaster trees are the most vigorous, produce more, and are less prone to be broken by strong winds or affected by dry climatic events. They are also widely known to produce oil which is much tastier. However, trees originating from micro-cuttings are seen as vulnerable, not productive, not durable and produce a less tasty oil, attributed by farmers to their lack of taproot system.

8The construction of grafted oleaster-olive agroecosystems is driven by three major factors: a) customary land tenure systems, b) use over historical times of oleaster oil still highly prized in the Rif and, c) a sophisticated knowledge of grafting techniques. Grafting oleasters originating from the clearing of natural forests enables to acquire new agrarian properties because productive trees are an important marker of collectively recognized property. Similar systems are well-known in other regions such as in South East Asia. During the French Protectorate period (1912-1955), forests were registered (“cadastré”) and access was restricted even for pastoral activities, although they represented important rangeland areas. This resulted in an accelerated rate of forest transformation and tree plantation in order to maintain access to the land (Fay 2015). Grafting oleaster thus represents a strategy for transformation of the forest into new stabilized agricultural lands.

9The farmer clears the maquis by cutting all trees at the exception of some oleasters according to the density required. He then burns, sows cereals and grafts the oleasters. These fields progressively become permanent open savannah types of agroforests (Photo 1) including pasture lands based on a rotation between cereals, pulses and fallow periods. Livestock currently graze after the cereals have been harvested. Grafting is always conducted on large diameters (> 5 cm) and is exceptionally conducted directly on roots when the oleaster is too old. The scions chosen are a set of traditional olive variety but one clone, the Picholine marocaine, is dominant, as elsewhere in Morocco (Khadari et al. 2008). The top of the grafted oleaster is coated with mud mixed with cereal straws – a mixture currently used for many constructions in the region. These trees quickly start producing relatively significant quantities (20-50 kg after three years). The productivity of the tree is attributed by the farmers to the vigor of the rootstock.

10The latter also influences the quality of the scion. In the words of some farmers, the variety changes its “passport”, the association of the two parts giving a longer life span to the scion and better oil quality.

Photo 1
Olive-Oleaster agroforests with cereal, hay production, broad bean fields in the forefront and uneven oleaster-olive tree populations throughout the landscape. © Yildiz Aumeeruddy-Thomas, Ain Dorij, Maroc, 2014

11Some oleasters are never grafted because they are known to yield very large amounts of fruits and are differentiated into two categories, big olives (berri meslal) and thin olives (berri rkek). Both are brought to the mill when significant harvests are available and produce a highly prized medicinal oil, “zit d’l berri” (Aumeeruddy-Thomas et al. 2014).

12Olive oil from oleaster – olive agroecosystems are extracted in traditional or modern electric mills. These mills are owned either by individuals or by larger enterprises. A diversity of oil types depends upon the extraction systems. They include in order of preference locally, zit d’lma, oil extracted without presses and strictly through decantation in water basins, zit chamiya, which is obtained from traditional pressing mats made from doum palm (Chamaerops humilis) and a mill using a grinding stone, and the zit makina, oil extracted from continuous systems in large industrial machines. The latter is the least appreciated because it requires hot water and high pressure but the rapidity of extraction is seen as an advantage. Makina oil is meant mainly for sales and export. Most families prefer the traditionally pressed oil which they think has the highest quality and the best taste.

13Other products from oleaster-olive agroecosystems are cereals, mostly barley and a range of other cereals that are becoming increasingly rare such as rye (Secale cereale) and old wheat varieties, pulses including local landraces of lentils, chickpea (Cicer arietinum, Kortchi, Maayzou landraces), broad bean (Vicia faba, a major food and rotational crop), pulses specifically used to feed animals with many landraces of bitter vetch (Vicia ervilia) and vetch (Lathyrus sativus). Bitter vetch has almost disappeared throughout the Mediterranean. In the Rif it is currently grown on very poor soils (El Fatehi et al. 2014). All the pulses cited are nitrogen-fixing plants and therefore enrich the soil. Other tree species such as Caroub (Ceratonia silique) used as fodder and Figs (Ficus carica) a staple fruit throughout the Maghreb, are associated with traditional oleaster-olive agroecosystems. Such agroecosystems are also rich in adventitious herbs used currently as highly nutritive steamed salads (bakoula) in spring (Clochey 2014). Finally, highly diverse animal dairy and meat products are based on the breeding of small sheep flocks, cows and other domestic animals such as chicken. Mules also graze in this agroecosystem; they are important for work and transport.

Modern olive plantations

14Modern plantations are based on an aggregated system devised by the “Plan Maroc Vert” (PMV) the present agrarian policy of the Moroccan State supported partly by a bilateral project with USA, the Millenium Challenge Account (MCA). These policies and the associated MCA project have boosted the development of new olive plantations which require large aggregated perimeters of at least 500 hectares requiring that farmers form a formally registered economic group (GIE). Farmer’s lands are generally selected in the richest arable lands where people used to cultivate cereals associated previously with relatively low density of olives and other trees. These lands, once cleared, are managed by a large state-funded enterprise that hires workers for digging holes and planting high densities of olive plants (50 – 100 trees /ha) in an orderly way (lattice-square plantations with trees 10m apart). They also water and protect the young plants against thieves. Olive plants originate from micro-cuttings raised in nurseries in large towns such as Meknes. Only one or two varieties at a time are planted. These may be Haouzia and/or Menara, a variation that probably originates from somatic mutations of the Picholine marocaine (El Bakkali et al. 2013). In areas where water is not accessible, water is transported by lorries for irrigation. Workers are very rarely hired locally. Olive trees from these projects yield only 3 – 5 kilos after 5 years. For two years, farmers are not allowed to bring animals to graze because of the high risks for small plants. Although farmers may plant some cereal and pulses, the first 2 years represent a no man’s land for farmers.

15Adventitious herbs are generally linked to cultivated cereals and pulses. The products from these new olive plantations are thus very low in quantity and do not contribute to local food security. The plantation density may in future overshadow cereals because of the high density unless farmers cut some trees after the two-year project period. Finally, oil originating from such plantations is extracted in large extraction units. In the area studied, one large production unit run by a farmer cooperative has machines that can produce 30 tons per day. At present the cooperative is experiencing the greatest difficulty because newly planted olive trees will not yield sufficient amounts for up to 10 years. Farmers therefore draw from their traditional olive plantations which in turn may decrease the individual access of each family to oil that is used directly because oil sold by cooperative aims to find larger outlets.

Landscape structure and topo-hydrological factors

16The two maps presented below (fig 1 and fig 2) show the two distinct situations that are compared. Table 1 shows associated topographic and water availability indexes that characterize them.

17The oleaster-olive agroecosystems represent a complex landscape mosaic integrating small patches of forests, open areas of annual crop productions, new olive plantations as well as mixed grafted oleaster-olive (fig 1). The latter shows that farmers have not hesitated to renovate existing oleaster-olive plantations with new plants possibly originating from projects. Through integrating traditional and modern systems, farmers’ strategies indicate their interest in renovating old grafted oleaster plantations and/or occupying land. Indeed, while doing other work in parallel, planting trees enables farmers to harvest and tend on a part-time basis or rent to a tenant and recover a portion of the harvest. Grafted oleaster coverage (23%) is similar to the 24% of annual crop coverage. The site is topographically higher in altitude and has the strongest slopes (TPI and slopes). New olive plantations and annual crop areas occupy the flattest areas. Maquis is non-appropriated land and its presence is difficult to interpret. The second site is less high in altitude and has lower values of mean slopes. Such areas are major target areas for PMV/MCA industrial plantations which represent 92% of coverage. The landscape homogeneity is much higher than in the first site. Access to soil water is not significantly different between the two sites or between categories within each site.

Figure 1
Olive-growing in the province of Ouezzane: example of a douar, the rural community of Mesmoda

Figure 2
Olive-growing in the province of Ouezzane: example of a douar, the rural community of Bni Quolla

Table 1. Measures of environmental factors in each plantation class between the two types of olive agroecosystems.

Table 1. Measures of environmental factors in each plantation class between the two types of olive agroecosystems.

Mes: Mesmouda site, represents oleaster-olive grafted agroecosystem and other classes in a complex mosaic.
Bni Quo: Bni Quolla site, represents a large PMV/MCA project plantation area with a low landscape mosaic diversity. Indexes calculated with SRTM.
TPI: Topographic Position Index, calculated in Qgis with SAGA.
TWI: Topographic Wetness Index, calculated in Qgis with SAGA.

Discussion and conclusion

18This paper shows that grafted oleaster-olive agroecosystems, while being an ancient technique, offers much hope as a model for food security within a context of climate change. Farmers’ capacity to take advantage of highly heterogeneous mountain terrain needs to be given the highest priority in future research. The vigor of the tree and the quality and taste of the oil is the main reason for farmers to maintain this technique because there is almost no forest left to transform. Grafting is scientifically acknowledged as having many positive impacts on domesticated crops (Warschefsky et al. 2016). In our case they allow the domesticated variety (mostly the Picholine marocaine) to have access to a large amount of nutrients from the rootstocks which are genetically very heterogeneous because they originate from a natural population. Rootstocks are known to be in contact with a soil microbiome made of a rich network of endophytic fungi in the case of Olea europaea and bacteria (Martins et al. 2016). These authors suggest that such associates may play an important role in production as well as protection against diseases through providing water and nutrients. Studies on other species show that specific types of genetic material may be transmitted between the rootstock and the variety (scion), a totally new research area which requires more investigation. Moreover, it is also known that the scion can also influence the biology of the rootstock because the litter is produced by the upper part of the tree which is transformed in the litter and feeds the rootstock.

19From an environmental perspective, the complex landscape mosaic of the oleaster-olive agroecosystem corresponds to the “nature matrix” that Perfecto et al. (2009) described. The high level of biotic interactions and species diversity within such complex landscapes is likely to make such systems, highly adaptive in contexts of climate change as well as regarding sanitary risks. It is interesting to note that the major olive disease, a bacteria (Xylella fastidiosa) which currently affects olive in northern Mediterranean countries has not yet affected varieties grafted on oleasters in Corsica (Rasplus, personal communication, 2016).

20The oleaster-olive agroecosystem is also by far more productive for local diets and food security. Industrial oil from new PMV/MCA plantations is likely to be in high competition with the international market and the cost of sale of this non-environmentally-based oil may fall as more and more consumers require high-quality organic olive oil. Although a general movement is starting for niche product productions for export or local tourism (produits de terroir) – possibly highly adapted to such mountainous regions – future plans should also ensure that local consumption of high quality oil is not jeopardized due to the high price of niche products.

21The major ethical question raised is whether it is morally correct to jeopardize agroecosystems that draw on long-standing knowledge that is agro-ecologically robust and important for local food security, in order to develop low-quality olive oil for export to the larger global market? Solutions to face climate risks and food security could draw on farmers’ experience and establish a system that integrates high quality niche products, areas dedicated for local consumption and other areas for intensive export-quality oil within a matrix with a high level of landscape and biological diversity.



Aumeeruddy-Thomas Y., Y. Hmimsa, M. Ater, B. Khadari (2014)
Beyond the divide between wild and domesticated: spatiality, domesticity and practices pertaining to fig (Ficus carica L.) and olive (Olea europaea L.) agroecosystems in Morocco In: Eds Alex Chevalier, Elena Marinova, Leonor Peña-Chocarro. Crops and people: choices and diversity through time. Earth EU, Brussels, OXFAM, London, p 191-197.

Besnard G., Khadari B., Navascués M., et al. (2013)
The complex history of the olive tree: from Late Quaternary diversification of Mediterranean lineages to primary domestication in the northern Levant, Proc. of the Roy. Soc. B-Biol Sc 280: 1756 DOI: 10.1098/rspb.2012.2833

Brun J.P. (1986)
L’Oléiculture antique en Provence. Les huileries du department du Var. Revue Archéologique de Narbonnaise, Supplément 15, Editions du CNRS, Paris

Camps G., (1961)
Massinissa ou les débuts de l’histoire, 1961, Imprimerie officielle, Alger.

COI (Conseil oléicole international) (2015)
Newsletter-Marché Oléicoles: Mars 2015. Madrid.

Clochey, L. (2014)
Hommes et plantes adventices de cultures céréalière en Pays Jbala (Ain Mediouna, Maroc). Mémoire de première année de master Écologie-Biodiversité (parcours Élevage des pays du Sud, Environnement, Développement), sous la direction de Y. Aumeeruddy-Thomas. Montpellier, Université Montpellier II, 53 p.

El Bakkali A, Haouane, H, Hadiddou A et al. (2013)
Genetic diversity of on-farm selected olive trees in Moroccan traditional olive orchards. Plant genetic resources characterization and utilization 11 (2):97:105 DOI: 10.1017/S1479262112000445

El Fatehi S, Béna G, Filali-Maltouf A, Ater M (2014)
Variation in yield component, phenology and morphological traits among Moroccan bitter vetch landraces. African Journal of Agricultural Research (9): 1801-1809

Fay G. (2015)
Collectivités, Territoires et maldéveloppement dans les campagnes marocaines, Préface Ismaïl Alaoui, Publication de la Faculté des Lettres et des Sciences humaines de Rabat, Rabat.

Kaniewski D, Van Campo E, Boiy T, Terral JF, Khadari B Besnard B (2012)
Primary domestication and early uses of the emblematic olive tree: palaeobotanical, historical and molecular evidence from the Middle East, Biol. Rev., pp. 000–000. doi:10.1111/j.1469-185X. 2012.00229.x

Khadari B, Charafi J, Moukhli A, Ater M (2008)
Substantial genetic diversity in cultivated Moroccan olive despite a single major cultivar: a paradoxical situation evidenced by the use of SSR loci. Tree Genet Genomes 4:213-221.

Kislev ME, Nadel D, Carmi I (1992)
Epipalaeolithic (19,000 B. P.) cereal and fruit diet at Ohalo II, Sea of Galilee, Israel. Review of Palaeobotany and Palynology 73: 161-166.

Martins F., J. Alberto Pereira, P. Bota, A. Bento, P. Baptista (2016)
Fungal endophyte communities in above-and belowground olive tree organs and the effect of season and geographic location on their structures. Fungal Ecology 20: 193-201

Perfecto I, J. Vandermeer, A. Wright (2009)
Nature’s Matrix: Linking Agriculture, Conservation and Food Sovereignty. Routledge, London.

Terral J.F., N. Alonso, R. Buxo Capdevilla, N. Chatti, L. Fabre, G.,Fiorentino, P. Marinval, G. Perez Jorda, B. Pradat, 2004
Historical biogeography of olive domestication (Olea europaea L.) as revealed by geometrical morphometry applied to biological and archaeological material. Journal of Biogeography, 31 (1): 63-77

Tinner W., J.F.N. van Leeuwen, D. Colombaroli, E. Vescovi, W.O. van des Knaap, P.D. Henne, S. Pasta, S. D’Angelo, T. LA Mantia (2009)
Holocene environmental and climatic changes at Gorgo Bosso, a coastal lake in southern Sicily, Italy, Quarternary Science Reviews 28: 1478-1510

Terral JF, Arnold-Simard G. (1996)
Beginnings of Olive Cultivation in Eastern Spain in Relation to Holocene Bioclimatic Changes, Quaternary Research 46, 176-185

Warschefsky, E.J., Klein, L.L., Frank, M.H., et al. (2016)
Rootstocks: Diversity, Domestication, and Impacts on Shoot Phenotypes. Trends in Plant Science 21(5): 418-437

Zohary D, Spiegel-Roy P. 1975
Beginnings of fruit growing in the Old World. Science 187: 319-327.

Table des illustrations

Légende Photo 1Olive-Oleaster agroforests with cereal, hay production, broad bean fields in the forefront and uneven oleaster-olive tree populations throughout the landscape. © Yildiz Aumeeruddy-Thomas, Ain Dorij, Maroc, 2014
Fichier image/, 337k
Légende Figure 1Olive-growing in the province of Ouezzane: example of a douar, the rural community of Mesmoda
Fichier image/, 364k
Légende Figure 2Olive-growing in the province of Ouezzane: example of a douar, the rural community of Bni Quolla
Fichier image/, 327k
Titre Table 1. Measures of environmental factors in each plantation class between the two types of olive agroecosystems.
Légende Mes: Mesmouda site, represents oleaster-olive grafted agroecosystem and other classes in a complex mosaic.Bni Quo: Bni Quolla site, represents a large PMV/MCA project plantation area with a low landscape mosaic diversity. Indexes calculated with SRTM.TPI: Topographic Position Index, calculated in Qgis with SAGA.TWI: Topographic Wetness Index, calculated in Qgis with SAGA.
Fichier image/, 229k


CEFE, UM, France
Master Student, University of Montpellier Centre d’Écologie Fonctionnelle et Évolutive (UMR CEFE, 5175)

Geomatic and remote sensing engineer IRSTEA, Territoires, environnement, télédétection et information spatiale (UMR TETIS) France.

Abdelmalek Essaadi University, Biodivmex Mistrals, France
Université Abdelmalek Essaadi,
Biodivmex Mistrals, France

© IRD Éditions, 2016

Conditions d’utilisation :