Using Palaeolimnology to Inform Freshwater Restoration in Madagascar

Madagascar is amongst the hottest of the world's biodiversity hotspots, with extremely high levels of species richness and endemicity(1). Malagasy people are among the world's poorest, being heavily dependent on natural resources.
Consequently, Madagascar has placed the conservation of natural capital at the heart of its National Development Plan and recently has tripled its Protected Area network(2). However, conservation attention has been heavily focused on forests; of the 46 Protected Areas established before 2011, 45 protect forest ecosystems, yet compared to forests, Malagasy wetlands show similar rates of endemism and threats to species with higher rates of destruction(3). Additionally many Malagasy livelihoods are wetland dependent: half of the population rely on unimproved water from rivers and lakes, and wetlands supply most of the country's rice, which is the staple foodstuff.
Recent analysis shows degradation is almost ubiquitous in wetlands on Madagascar's central plateau, and has multiple causes. Extensive deforestation and burning over the last 1000 years has resulted in extremely high rates of soil erosion and siltation. A large proportion of the marshes surrounding waterbodies have been converted to rice agriculture, and a survey of 37 wetlands revealed that 82% of marsh habitats had been converted to agriculture(3). There has also been extensive introduction of invasive alien fish species since the 1950s and uncontrolled pesticide application to agricultural
wetlands since the 1990s.
Rather than a simple focus on the conservation of the remaining high quality wetlands, the pervasive nature of extreme degradation means that active restoration is needed to bring back biodiversity and ecosystem services. There are two prerequisites for restoration; firstly, there needs to be a better understanding of pre-disturbance conditions, which can serve as targets for restoration measures, and secondly, finding sustainable community-led solutions can be challenging in Madagascar, and therefore focusing restoration efforts on removing the most important pressures is vital. However, the
scarcity of undisturbed sites means that it is difficult to use contemporary ecology to set targets for restoration of degraded sites. Furthermore, disentangling the impacts of the individual stressors using spatial comparisons is impossible because they are so ubiquitous and spatially correlated.
The causes of degradation in Madagascar have though operated at different temporal scales, and therefore a long-term perspective of the extent and type of impacts on Malagasy wetlands can be used to identify the most pertinent causes of degradation, and establish restoration targets. Although there is a paucity of long-term monitoring data, it is possible to
extend the timescale of freshwater datasets by complementing contemporary data with that derived from lake sediment cores (palaeolimnology). This approach has been used in Europe to contextualise a wide range of environmental issues and to define reference conditions(4) but hasn't been adopted widely elsewhere. Undisturbed cores taken from the accumulation zone of a waterbody are accurately dated and sliced to provide different time-depth intervals. These are examined by a range of techniques to provide direct evidence of community composition (by reconstructing species distributions and abundance), indirect indications of physio-chemical conditions (by complementing ecological knowledge or using transfer functions), and trophic relationships (e.g. using stable isotope analysis).
This research will use a palaeolimnological approach to establish reference conditions for wetlands in central Madagascar and to set targets for restoration at Lake Sofia, where reintroduction of a critically endangered bird is underway. 1Myers et al 2000 Nature, 403, 853-858
2UN 2007. Madagascar Action Plan
3Bamford et al in Press PLoS ONE
4Bennion et al 2011 J Palaeol. 45 415-431