Climate Change and Ice Cap History on Kilimanjaro

Glaciers around the world have retreated over the 20th century, which is also true of tropical glaciers. Although the characteristics of tropical glaciers (a high sensitivity to moisture-related climate variables) also apply to the glaciers on Kilimanjaro (Equatorial East Africa), studying their behavior requires a special view. This is because different glacier systems exist on Kilimanjaro: tabular-shaped ice bodies on the summit plateau, and slope glaciers below the summit plateau on the mountain`s steep flanks. The plateau glaciers are margined by vertical ice cliffs that - once established - lead to an irreversible areal recession of the plateau glaciers, regardless of the mass balance on the plateau glacier`s horizontal surfaces. A preceding project demonstrated that the main climatic cause of the current glacier retreat on Kilimanjaro (which commenced around 1880) is a regionally drier climate since the late 19th century, and that the glaciers show a much higher sensitivity to precipitation fluctuations than to air temperature changes. It has also become clear that the current climate is pushing the glaciers close to disappearance, which raises the question about what climate conditions were responsible for the formation and maintenance of the glaciers. The present project therefore aims at reconstructing at least 500 years of glacier history on Kilimanjaro and climate change in the tropics, to identify potential phases of an ice-free summit of Africa`s highest mountain. Since other precipitation-sensitive proxies (particularly lake levels) indicate greater climate fluctuations before 1880 than afterwards, it is likely that glacier existence on Kilimanjaro`s summit follows a relatively short-term cycle. On-site meteorological measurements obtained from automatic weather stations will continue in the proposed project, in order to run, calibrate and validate a physically-based glacier mass balance model. This model is able to quantify mass exchanges between glacier and atmosphere. Multicentury-forcing data for the mass balance model will be constructed from output of paleoclimate simulations with global climate models (coupled atmosphere-ocean general circulation models, AOGCMs). To transfer the AOGCM output to local Kilimanjaro conditions, a regionalization technique (statistical downscaling) will be employed. Finally, a multicentury history of mass turnover on the Kilimanjaro glaciers will be simulated with the mass balance model - which will indicate potential cyclicity of the Kilimanjaro glaciers that can be compared to (a) climate reconstructions from other environmental systems (lakes, ice cores) and (b) climate dynamics in the AOGCMs. The experimental part of the project concerns both the treatment of vertical ice walls in the model, as well as the ability of the regionalization techniques to produce data with high temporal resolution. If the proposed modeling chain is successful, it will provide a valuable tool for predicting the future evolution of glaciers on Kilimanjaro, since AOGCM output is also available for future climate scenarios.

The goal of this project was to establish the multicentury history of the relation between climate and glaciers on Kilimanjaro (East Africa), and to reveal the role of ice cliffs in this history, which margin the present ice fields on the summit plateau. Our research was guided by (a) consideration of links in the climate system over different space-time scales, and (b) physical principles (rather than statistical relationships). Major results are: (1) Glaciers on Kilimanjaro expanded and shrunk in concert with wet and dry periods of the regional climate over the last millennium. (2) Based on the largest slope glacier we quantified the difference in local climate for the last maximum extent (late 19th century) compared to present, which was characterized by 200 40 mm more precipitation per year on the summit of Kilimanjaro. (3) Glacier changes on Kilimanjaro are therefore a bad indicator of global warming, but a very good indicator of dynamical linkages in the climate system, since the local and regional climate mentioned above are governed by sea surface temperature distributions in, and atmospheric flow over, the equatorial Indian Ocean. (4) The ice cliffs on the summit are aligned east-west. At our sample site (oriented to the south) there is strong recession during the sunlit period ( 13 cm/month), and little retreat during the period when the sun is north of Kilimanjaro ( 1 cm/month). The particular radiation geometry is therefore the main driver of these cliffs. (5) Ice cliff formation is unavoidable in dry climate (as at present), and once they exist the life time of the plateau glaciers is roughly 165 10 years. The present plateau (not slope!) glaciers are hence expected to disappear by mid 21st century if the dry climate persists. Social implications of our research are: (1) The recession of Kilimanjaro`s glaciers is another indicator for the presence of drought in East Africa over the last 120 years, and policy measures must therefore target to secure regional water resources. (2) We also find that man-made vegetation change on the slope of the mountain tends to decrease precipitation in the forest belt (the major local water reservoir), so a strict control of the forest belt against illegal logging is of utmost importance. (3) In the light of global climate change, we have identified a key in the climate system to understand shifts in tropical precipitation zones in a warming world. This will help to assess projections of regional precipitation in the tropics, where societies are most sensitive to the availability of water.