Climate changes since the last glacial maximum in Morocco and predicted impact on the Mediterranean ecosystems
by
Cheddadi, Rachid
Institut des Sciences de l'Evolution, UMR CNRS 5554, UM II, case postale 61, 34095 Montpellier, France
Coauthors: Taieb, M., Damnati, B., Ortu, E., Guiot, J., Lamb, H.

Introduction Morocco is the easternmost country in the Mediterranean basin, extending from 28°N to 36° N latitude and from 2°W to 12° W longitude. The climate of the country lies between the Mediterranean climatic belt and the Saharan subtropical desert, with high mountains ranges introducing an altitudinal climatic gradient that overlies the latitudinal gradient. The country falls under Oceanic, Mediterranean, and Saharan influences. The hot and dry summers are dependent upon the seasonal shift of the subtropical high-pressure belt, while winter rainfall is determined by the southward displacement of the northwesterly cyclonic activity. Two gradients characterize the climate of Morocco: a north-south gradient and a west-east gradient; along them precipitation decreases and temperature increases. The climate and vegetation changes during the Holocene in Morocco is known in a more or less continuous way through studies of diatoms and lake levels (Gasse and Fontes, 1992), lake plankton (El Hamouti et al., 1991) and pollen analysis (Reille, 1976, 1977; Lamb et al., 1989, Lamb et al., 1995). The aim of the present talk is two fold: 1. To discuss some climate reconstructions obtained from two pollen records located in the Middle Atlas, Morocco 2. To show some simulations of vegetation changes in the Mediterranean basin using a few future climate scenarios. Palaeodata The climate reconstructions presented here are based on palynological data obtained from lake Tigalmamine (Cheddadi et al., 1998) and lake Ifrah (Cheddadi et al., in prep). Both lakes (figure 1) are located in the Middle Atlas, Morocco. Fig. 1: Vegetation map and sites location. Lake Ifrah has been recently cored and it yielded a record, which is still not dated. However, the palynological data provide a good hint of the chronological frame. Thus, the record from lake Ifrah seems to cover the last glacial period and the beginning of the Holocene. One should state that the climate changes during the last glacial period in North Africa, and particularly in Morocco, are poorly known. Earlier studies have stated that the arid conditions prevailed over northern Africa during the Last glacial period (Gasse et al., 1990; Street and Grove, 1979). The pollen record from lake Ifrah allowed performing a quantitative climate reconstruction, which shows that the winter temperatures during the last glacial maximum have been lower than the present by about 10 to 12°C. The annual amount of rainfall during the same period did not exceed 400 mm/year. Nowadays the area of lake Ifrah receives between 800 and 900 mm/year. The early Holocene, between 10.5 and 9 kyr BP, from both lake Tigalmamine and lake Ifrah show similar climatic response. It is a period drier and slightly warmer than the present, by about 2 to 3°C, during both winter and summer (figure 2). Such dry and warm period seems to be quite synchronous with an arid phase identified in Sebkha Mellala, Algeria (Gasse et al., 1990). Such low precipitation period may be correlated with the Preboreal in Western Europe, where the extension of the forest (Pinus, Betula, Quercus) indicates rapid warming conditions but with still relatively low precipitation (Guiot, 1987). Fig. 2: Climate reconstruction from lake Tigalmamine. The Holocene period between approximately 9 kyr BP and the present is well detailed from lake Tigalmamine (figure 2) but it seems to be missing from the record of lake Ifrah. The Tigalmamine record shows that both winter and summer temperature values obtained for the period between ca 9 and 8.5 kyr BP are ca 4 °C higher than the present. Such an important warming lead to an increased evaporation over the Atlantic Ocean. A strengthening of the westerlies may explain the subsequent increase of precipitation over the Atlas Mountains. A gradual increase of the annual amount of precipitation is observed between 7 and 5.8 kyr BP which could be related to changes in the thermohaline circulation of the Atlantic ocean as shown by Street-Perrott and Perrott (1990). After 6 kyr BP, the moisture made available by the global warming and the strengthening of the westerlies is recorded in the Middle Atlas mountains in terms of increased annual amount of rainfall. Kutzback and Perrott (1985), using an atmospheric general-circulation model have shown that at 9 and 6 kyr BP the wind direction was south-westerly and the transport of moisture from the oceans onto the northern hemisphere land masses is increased. A climatic changes is recorded after 4.5 kyr BP inducing a drop in the annual precipitation and a slight increase in both winter and summer temperatures. Once more, one may make a correlation with a lowering in the sea surface temperature as shown in the core SU81-18 from the Atlantic Ocean (Bard et al., 1987). A significant cooling and an increase in the Pann follow after 4 kyr BP. In eastern Africa, a drop of about 2°C in the annual temperature at the same period is observed (Bonnefille et al., 1990). After ca 2 kyr BP, the pollen data indicate that the human activity have increased substantially. In the pollen diagram one can observe that Cedrus percents decrease slightly and are replaced by Gramineae. This grass spread might not be related to cultures since no cereal pollen have been identified but to a degradation of the forest by cattle grazing which opens the forest (Finckh & Deil, 1989) and allows grasses to occupy the area. Future simulations For the next century, GCM simulations predict an increase of the annual temperature between 1 and 3°C during the next half-century (IPCC, 1995) around the Mediterranean. The annual amount of rainfall is expected to decrease by 10 to 20% in the same area. What will be the impact of such abrupt climatic changes on the vegetation? Cheddadi et al. (2001) have used a vegetation model, BIOME3, to simulate the impact of climate on the ecosystems (figure 3). The scenarios used to run BIOME3 are all characterized by an atmospheric concentration of CO2 ([CO2]) of 500 ppmv and by an annual temperature 2°C higher than the present. Two types of experiments have been performed: 1. In the first experiment (figure 3A), the precipitation was maintained at its present-day level. Using such scenario, BIOME3 indicates a substantial southward shift of the Mediterranean vegetation and a spread of the evergreen and conifer forests in the northern Mediterranean. Xerophytic woodlands become restricted to a narrow strip. In the western basin, Mediterranean xerophytic woodlands are restricted to southern Spain and southern Italy and they no longer occur in southern France. In the Eastern Mediterranean, conifers occupy the central part of Turkey. In northwestern Africa, Mediterranean xerophytic vegetation occupies a more extensive territory than today and the arid steppe/desert boundary shifts southward. 2. in the second experiment (figures 3B and 3C) we have decreased the amount of annual rainfall by 10, 15, 20, 25 and 30% of the present day values. This set of model simulations show that a decrease of 30 % of the present-day precipitation induces the development of steppe-like vegetation in the northern Mediterranean and the extension of the desert in northwestern Africa. Fig. 3 (A, B and C): Future climate impact on the Mediterranean ecosystems. These two sets of simulations show that, firstly, an increase of the [CO2], jointly with an increase of ca 2°C of the annual temperature, would not necessarily lead to desertification on any part of the Mediterranean unless the annual amount of precipitation decreases by 20 to 30%. The IPCC simulations predict such a strong decrease. These results are not intended to minimize the effect and/or the impact of the increase of the [CO2] during the next few decades, in terms of agriculture and economy. They only emphasize the fact that the Mediterranean vegetation would not necessarily be replaced by a more arid vegetation if the rainfall amount is not decreased by the amount that is predicted by the IPCC. Hennessy et al. (1997) indicate that a doubling of the [CO2] would lead to an average increase of about 5 to 10% annual precipitation at the Mediterranean latitude. The latter hypothesis would signify that, if the vegetation model predictions can be considered as reliable, there would be more evergreen trees than today in the Mediterranean. Finally, one should state that the vegetation model we have used does not take into account the human activity such as grazing, cultivation, deforestation ... that has a serious impact on the evolution of the ecosystems and their feedback on the climate. The human impact remains a major factor in the shaping of forest ecosystems and it constrains the natural impact of the climate. Bonnefille R, Roeland JC Guiot J., 1990. Temperature and rainfall estimates for the past 40,000 years in equatorial Africa. Nature, 346: 347-349. Cheddadi, R., Lamb, H., Guiot, J., van der Kaars, S., 1998: Reconstruction of the Holocene climatic events using a pollen record from Tigalmamine lake, Morocco: Relationships to global climate change. Climate Dynamics, 14 (12): 883-890. Cheddadi, R., Guiot, J., Jolly, D., 2001. The Mediterranean vegetation: what if the atmospheric CO2 increased? Landscape Ecology, 16 (7): 667-675. El Hamouti N, Lamb HF, Fontes J Ch, Gasse F., 1991. Changements hydrologiques abrupts dans le Moyen Atlas marocain depuis le dernier maximum glaciaire. Comptes Rendus de l'Académie des Sciences, Séries II, 313: 259-265. Finckh, M., & Deil, U., 1989. Approche de l'étude de la dégradation des forêts du Jbel Bougriy (Moyen Atlas Central) Revue Géographique du Maroc, 13: 54-69. Gasse F, Fontes J. Ch., 1992. Climatic changes in northwest Africa during the last deglaciation (16-7 ka BP). NATO ASI Series (12): 295-322 Gasse, F., Téhet, R., Durand, A., Gibert, E. and Fontes, J. Ch., 1990. The arid-humid transition in the Sahara and the Sahel during the last deglaciation. Nature (346): 141-146. Guiot, J., 1987. Late quaternary climatic change in France estimated from multivariate pollen time series. Quaternary Research, 28: 100-118. Hennessy, K.J., Gregory, J.M., & Mitchell, J.F.B., 1997. Changes in daily precipitation under enhanced greenhouse conditions. Climate Dynamics, 13: 667-680. Kutzbach, J., & Street-Perrott, F.A., 1985. Milankovitch forcing of fluctuations in the level of tropical lakes from 18 to 0 kyr BP Nature, 317: 130-134. Lamb, H.F., Eicher, U. and Switsur, V.R., 1989. An 18,000 year record of vegetation, lake-level, and climatic change from Tigalmamine, Middle Atlas, Morocco. Journal of Biogeography, 16: 65-74. Lamb, H.F., Gasse, F., Benkadour, A., El Hamouti, N., van der Kaars, S., Perkins, W.T., Pearce, N.J. and Roberts, C.N., 1995. Relation between century-scale Holocene arid intervals in tropical and temperate zones. Nature, 373: 134-137. Reille, M., 1976. Analyse pollinique de sédiments postglaciaires dans le Moyen Atlas et le Haut Atlas marocains: premiers résultats. Ecol. Medit., 2: 153-170. Reille, M., 1977. Contribution pollenanalytique a l'histoire Holocène de la végétation des montagnes du Rif (Maroc septentrional). INQUA: 53-76. Street F.A, Grove A.T., 1979. Global maps of lake-level fluctuations since 30 000 BP. Quaternary Research, 12: 83-118. Street-Perrott, F.A., and Perrott, R.A., 1990. Abrupt climate fluctuations in the tropics: the influence of Atlantic ocean circulation. Nature, 343: 607-612.

Date received: November 25, 2003