Note to users. If you're seeing this message, it means that your browser cannot find this page's style/presentation instructions -- or possibly that you are using a browser that does not support current Web standards. Find out more about why this message is appearing, and what you can do to make your experience of our site the best it can be.

Science 28 November 2008:
Vol. 322. no. 5906, p. 1326
DOI: 10.1126/science.1163381
Prev | Table of Contents | Next

Technical Comments

Comment on "Climate-Driven Ecosystem Succession in the Sahara: The Past 6000 Years"

Victor Brovkin1* and Martin Claussen1,2

Kröpelin et al. (Research Articles, 9 May 2008, p. 765) interpreted a sediment record from Lake Yoa in the east-central part of North Africa as support for a weak biogeophysical climate-vegetation feedback in the Sahara during the mid-Holocene. We argue that the new data do not invalidate earlier modeling results on strong land-atmosphere coupling in the Western Sahara for which the Lake Yoa record is far less representative.

1 Max Planck Institute for Meteorology, Bundesstraße 55, 20146 Hamburg, Germany.
2 Meteorological Institute, University of Hamburg, Bundesstraße 55, 20146 Hamburg, Germany.

* To whom correspondence should be addressed. E-mail: victor.brovkin{at}zmaw.de

Abrupt climate changes in the past provide invaluable insights on rapid climate dynamics in the future. The climate system is full of numerous feedbacks between different system components, such as vegetation and atmosphere, and North Africa might be one of the important tipping points in the system (1). That is why a question about the abruptness of past changes in this region requires detailed investigation.

Kröpelin et al. (2) presented fascinating details of desiccation of the east-central part of North Africa over the past 6000 years. Although we acknowledge the progress in paleoclimatic reconstruction, we disagree with some of the authors' conclusions. Kröpelin et al. state that their data imply a "disagreement with modeling results indicating abrupt mid-Holocene vegetation collapse," suggesting "that the implicated biogeophysical climate-vegetation feedback may have been relatively weak." We argue that the new data do not invalidate earlier hypotheses and modeling results (3–6).

The proposed biogeophysical feedback between vegetation cover and precipitation in the Sahara (e.g., 5–7) operates through modification of surface albedo (a fraction of solar irradiation reflected back to space) and moisture recycling in the presence of plants. In North Africa, the difference in albedo values between vegetated surfaces and desert can be particularly large, with bare ground albedo values exceeding 0.5 in some areas (8, 9). Vegetated surface traps more solar irradiation and warms the air column over land. Elevated temperature contrast between the land and adjusted ocean leads to an intensification of the regional atmospheric circulation, which results in more rainfall. As shown in modeling studies by Claussen and Gayler (4) and Renssen et al. (3), the biogeophysical feedback is most relevant for the Western Sahara (Fig. 1). The eastward increase in aridity in mid-Holocene Sahara is supported by geological data. Although sparsely distributed pollen records reveal that a large part of the Sahara in the mid-Holocene was covered with steppe or xerophytic shrubs (10), some reconstructions suggest that the western part of the Sahara was more humid than its eastern part (11).


Figure 1 Fig. 1. Simulated difference between mid-Holocene and present-day vegetation cover, revealing changes mainly in the Western part of the Sahara, the focus of earlier hypotheses of abrupt ecosystem change in the Sahara. Gray shading indicates changes from mid-Holocene vegetated state to present-day desert (4); the green shaded area has more than 30% reduction in vegetation cover during the Holocene (3). A strong positive biogeophysical feedback between vegetation cover and precipitation in the Western Sahara found in (3, 4) may result in an abrupt shift from wet to dry regime. The blue shaded area indicates the area of abrupt vegetation changes attributed to a nonlinear response of vegetation to precipitation from simulation by Liu et al. (14). The stars indicate the position of Lake Yoa (19.03°N, 20.31°E) and Ocean Drilling Program site 658C (20.45°N, 18.35°W). The record from Lake Yoa does not seem to show evidence of abrupt ecosystem transitions (2), whereas the marine sediment core reveals an abrupt change in dust transport from the continent (13). Yellow circles indicate sites with mid-Holocene steppes or xerophytic shrubs from the BIOME-6000 database (10). [View Larger Version of this Image (70K GIF file)]
 

The first computer simulation of transient Holocene dynamics of the Sahara was undertaken by using the CLIMBER-2 model (12) which does not resolve any difference between the western and the eastern part of the Sahara. In this simulation, the transition from a "green" to a barren Sahara took place within several hundred years about 5000 years before the present (yr B.P.) at a much faster pace than the driving force of changes in summer insolation. The geological records of dust transport into the North Atlantic (13) exhibit an even faster transition from weak to strong dust flux around 5500 yr B.P. Certainly, the dust records should be interpreted with care because the abrupt change in the dust depositions may reflect not only a decrease in vegetation cover but also an increase in the source area of the dust caused by lake desiccations.

Later model results about abruptness of North African vegetation changes in mid-Holocene give a more mixed picture. In the transient simulation with the ECBILT-CLIO-VECODE model (3), high climate variability masked out the abrupt climate changes in the Western Sahara. However, increased climate instability about 6000 yr B.P. was interpreted as a system shift from moist to dry regime. Liu et al. (14) used another fast climate-vegetation model, FOAM-LPJ, and found the abrupt changes not in the western but in the eastern part of the Sahara (Fig. 1), although the later results do not seem to be supported by the Lake Yoa record (2). Abrupt vegetation changes in the Eastern Sahara simulated by FOAM-LPJ (14) are associated not with the strong biogeophysical feedback but with a nonlinear response of vegetation cover to precipitation. In contrast, model studies (3, 4) that simulate mid-Holocene expansion of vegetation cover into the Western Sahara (Fig. 1) attribute a substantial part of this expansion to the strong coupling between land and atmosphere. Detailed understanding of model differences regarding the strength of biogeophysical feedback requires further investigation.

In conclusion, the jury is still out on the invalidation of the earlier assertion of abrupt ecosystem changes in the Sahara. To prove or falsify the hypothesis on abrupt changes in the Western Sahara, a convincing summary of terrestrial paleoclimatic records from this region has to be presented. Until then, the conclusion of Kröpelin et al. (2) on weak biogeophysical feedback is not justified in general.


References and Notes

Received for publication 17 July 2008. Accepted for publication 30 October 2008.

To Advertise     Find Products

ADVERTISEMENT

Featured Jobs

Science. ISSN 0036-8075 (print), 1095-9203 (online)