Limnological and ecological sensitivity of Rwenzori mountain lakes to climate warming

Research paper by Hilde Eggermont, Dirk Verschuren, Leen Audenaert, Luc Lens, James Russell, Gerrit Klaassen, Oliver Heiri

Indexed on: 25 Feb '10Published on: 25 Feb '10Published in: Hydrobiologia


An increasing number of studies forecast that anthropogenic climate change poses serious consequences for the biodiversity and ecosystem functioning of high-elevation mountain lakes, through a series of both direct and indirect effects. The impacts of future climate warming on alpine ecosystems are of particular concern, given that warming is expected to be most pronounced at high elevations around the globe. Here, we evaluate the limnological and ecological sensitivity of high-elevation lakes in the Rwenzori Mountains (Uganda-D. R. Congo) to climate change. This is done by comparing the species assemblages of larval chironomid remains deposited recently in lake sediments with those deposited at the base of short cores (dated to within or shortly after the Little Ice Age) in 16 lakes. Chironomid-based reconstructions of mean annual air temperature (MATemp) are made using a variety of inference models (with transfer functions based on weighted averaging, weighted-averaging partial least squares, and a weighted modern analogue technique), and two different calibration data sets, one covering the full regional temperature gradient and one comprising only high-elevation Rwenzori lakes and ponds. The reconstructed historical temperature change ranges between a cooling of −2.03°C and a warming of +3.22°C (with n = 16 lakes × 3 models × 2 calibration data sets). However, excluding the atypical mid-elevation lake Mahoma (2,990 m altitude), we find a three-to-one ratio of cases of inferred warming against inferred cooling, and of the 24 Δ MATemp values exceeding 0.60°C, 23 are positive and only one is negative. Chironomid-inferred temperature changes mostly fall within the error range of the regional temperature inference models. A generalized linear mixed model analysis of the combined result from all lakes (except Mahoma) nevertheless indicates significantly warmer MATemp (on average +0.38 ± 0.11°C) at present compared to between ~85 and ~645 years ago. Inferred temperature changes are independent of whether lakes are located in glaciated or non-glaciated catchments, and of the age of the core base, suggesting that at least part of the signal is due to relatively recent, anthropogenic warming. The direction of faunal change at the lakes in relation to established species–environment relationships suggests that part of the observed shifts in species composition reflect lake-specific evolution in habitat features other than temperature, such as nutrients, pH or oxygen regime, which in our present calibration data set co-vary with temperature to a greater or lesser extent. The fairly uniform and marked historical warming trend in Rwenzori lakes documented by this study highlights their ecological vulnerability and their value as early warning systems for detecting the limnological and ecological effects of global warming.