A biome-scale assessment of the impact of invasive alien plants on ecosystem services in South Africa.

This paper reports an assessment of the current and potential impacts of invasive alien plants on selected ecosystem services in South Africa. We used data on the current and potential future distribution of 56 invasive alien plant species to estimate their impact on four services (surface water runoff, groundwater recharge, livestock production and biodiversity) in five terrestrial biomes. The estimated reductions in surface water runoff as a result of current invasions were >3000 million m(3) (about 7% of the national total), most of which is from the fynbos (shrubland) and grassland biomes; the potential reductions would be more than eight times greater if invasive alien plants were to occupy the full extent of their potential range. Impacts on groundwater recharge would be less severe, potentially amounting to approximately 1.5% of the estimated maximum reductions in surface water runoff. Reductions in grazing capacity as a result of current levels of invasion amounted to just over 1% of the potential number of livestock that could be supported. However, future impacts could increase to 71%. A 'biodiversity intactness index' (the remaining proportion of pre-modern populations) ranged from 89% to 71% for the five biomes. With the exception of the fynbos biome, current invasions have almost no impact on biodiversity intactness. Under future levels of invasion, however, these intactness values decrease to around 30% for the savanna, fynbos and grassland biomes, but to even lower values (13% and 4%) for the two karoo biomes. Thus, while the current impacts of invasive alien plants are relatively low (with the exception of those on surface water runoff), the future impacts could be very high. While the errors in these estimates are likely to be substantial, the predicted impacts are sufficiently large to suggest that there is serious cause for concern.

Measuring conditions and trends in ecosystem services at multiple scales: the Southern African Millennium Ecosystem Assessment (SAfMA) experience.

The Southern African Millennium Ecosystem Assessment (SAfMA) evaluated the relationships between ecosystem services and human well-being at multiple scales, ranging from local through to sub-continental. Trends in ecosystem services (fresh water, food, fuel-wood, cultural and biodiversity) over the period 1990-2000 were mixed across scales. Freshwater resources appear strained across the continent with large numbers of people not securing adequate supplies, especially of good quality water. This translates to high infant mortality patterns across the region. In some areas, the use of water resources for irrigated agriculture and urban-industrial expansion is taking place at considerable cost to the quality and quantity of freshwater available to ecosystems and for domestic use. Staple cereal production across the region has increased but was outstripped by population growth while protein malnutrition is on the rise. The much-anticipated wood-fuel crisis on the subcontinent has not materialized but some areas are experiencing shortages while numerous others remain vulnerable. Cultural benefits of biodiversity are considerable, though hard to quantify or track over time. Biodiversity resources remain at reasonable levels, but are declining faster than reflected in species extinction rates and appear highly sensitive to land-use decisions. The SAfMA sub-global assessment provided an opportunity to experiment with innovative ways to assess ecosystem services including the use of supply-demand surfaces, service sources and sink areas, priority areas for service provision, service 'hotspots' and trade-off assessments.

Complementarity as a biodiversity indicator strategy.

Richness, rarity, endemism and complementarity of indicator taxon species are often used to select conservation areas, which are then assumed to represent most regional biodiversity. Assessments of the degree to which these indicator conservation areas coincide across different taxa have been conducted on a variety of vertebrate, invertebrate and plant groups at a national scale in Britain, Canada, USA and South Africa and at a regional scale in Cameroon, Uganda and the USA. A low degree of spatial overlap among and within these selected indicator conservation areas has been demonstrated. These results tend to suggest that indicator conservation areas display little congruence across different taxa. However, some of these studies demonstrate that many conservation areas for indicator taxa capture a high proportion of non-target species. Thus it appears that indicator conservation areas might sample overall biodiversity efficiently. These indicator conservation areas may, however, exclude species essential for effective conservation, e.g. rare, endemic or endangered species. The present study investigated the value of indicator taxa as biodiversity surrogates using spatial congruence and representativeness of different indicator priority conservation areas. The conservation status of species excluded by the indicator approaches is also assessed. Indicator priority conservation areas demonstrate high land area requirements in order to fully represent non-target species. These results suggest that efficient priority area selection techniques must reach a compromise between maximizing non-target species gains and minimizing land-use requirements. Reserve selection procedures using indicator-based complementarity appear to be approaches which best satisfy this trade-off.