Biodiversity and ecosystem services in agricultural landscapes - are we asking the right questions?

The assumed relationship between biodiversity or local richness and the persistence of 'ecosystem services' (that can sustain productivity on-site as well as off-site, e.g. through regulation of water flow and storage) in agricultural landscapes has generated considerable interest and a range of experimental approaches. The abstraction level aimed for, however, may be too high to yield meaningful results. Many of the experiments on which evidence in favour or otherwise are based are artificial and do not support the bold generalisations to other spatial and temporal scales that are often made. Future investigations should utilise co-evolved communities, be structured to investigate the distinct roles of clearly defined functional groups, separate the effects of between- and within-group diversity and be conducted over a range of stress and disturbance situations. An integral part of agricultural intensification at the plot level is the deliberate reduction of diversity. This does not necessarily result in impairment of ecosystem services of direct relevance to the land user unless the hypothesised diversity-function threshold is breached by elimination of a key functional group or species. Key functions may also be substituted with petro-chemical energy in order to achieve perceived efficiencies in the production of specific goods. This can result in the maintenance of ecosystem services of importance to agricultural production at levels of biodiversity below the assumed 'functional threshold'. However, it can also result in impairment of other services and under some conditions the de-linking of the diversity-function relationship. Avoidance of these effects or attempts to restore non-essential ecosystem services are only likely to be made by land users at the plot scale if direct economic benefit can be thereby achieved. At the plot and farm scales biodiversity is unlikely to be maintained for purposes other than those of direct use or 'utilitarian' benefits and often at levels lower than those necessary for maintenance of many ecosystem services. The exceptions may be traditional systems where intrinsic values (social customs) continue to provide reasons for diversity maintenance. High levels of biodiversity in managed landscapes are more Likely to be maintained for reasons of intrinsic, serependic ('option' or 'bequest) values or utilitarian ('direct use') than for functional or ecosystem service values. The major opportunity for both maintaining ecosystem services and biodiversity outside conservation areas lies in promoting diversity of land-use at the landscape and farm rather than field scale. This requires, however, an economic and policy climate that favours diversification in land uses and diversity among land users. (C) 2004 Published by Elsevier B.V.