more likely to adopt improved seeds and high yield varieties, and apply more fertilizer and manure. Many farmers could benefit from these technologies, no major infrastructural development is needed, and the benefits are more equitable than large-scale irrigation projects. One possible drawback of these approaches is that they often have a high labor demand and that may deter adoption particularly where HIV/ AIDS rates are high.

 In SSA, unlike most other regions, water resources typically are not over-exploited (a key exception being South Africa). Most countries have enough water to meet their near-future needs—though these resources are often as yet untapped. Yet, though there is considerable scope for increased exploitation, most countries in SSA are not currently making the necessary investments to exploit the water resources (Molden and de Fraiture, 2004). Therefore an immediate challenge for many countries in SSA is to exploit the existing water resources more fully. Water scarcity is likely to become a much larger issue in the future, and is already causing localized conflicts in some countries (for example, the Ewaso Ng’Iro North Basin in Kenya) (Weismann, 2000) and so mechanisms are required to ensure that water exploitation is technically and economically efficient and that equitable access to water resources is taken into account.

Irrigation. In the past, there was a considerable focus of AKST on the use of large-scale irrigation for agricultural systems. Although such irrigation systems can have a positive impact on poverty reduction, they have at the same time often proven incompatible with environmental concerns where water off-take for agriculture has a negative impact on water-related ecosystems and ecosystem services. Moreover, research from Asia suggests that research into rainfed areas offers greater productivity increases and greater reductions in poverty than similar investments in irrigated agriculture (Fan et al., 2000a,b; Bindraban and Rabbinge, 2003).

Therefore, the potential for irrigation needs to be considered in the context of alternative water management strategies, external costs imposed by an irrigation scheme and distributional considerations. Investment in irrigation requires coordination among a number of farmers and significant upfront funds. NEPAD proposes that countries set up public-private partnerships for managing basic irrigation infrastructure, and encourage the private sector to invest in irrigated agriculture in parallel. These investments are only likely to occur however if the legal framework is sufficiently transparent and credible for the private sector to be willing to make long-term investments.

Water resources in SSA have typically been managed within administrative boundaries. A more logical approach is for water resources to be managed within the boundaries of a river basin (UNEP, 1999). Such an approach requires institution building and sharing of information. Further, organizational structures most likely will need to be adapted to reflect realities such as the increasingly artificial divide between rainfed and irrigated agriculture (Molden and de Fraiture, 2004). The development of water-harvesting techniques and small-scale irrigation are likely to be hindered by the current sectoral distinction between rain fed and irrigated

agriculture, reinforced by the current professional divide between, for example, agronomists who work on rain-fed agriculture and irrigation engineers (Molden and de Fraiture, 2004), and institutional divide—these two areas typically fall under different government ministries. Either new explicit institutional linkages are required, or the merging of responsibilities within one particular ministry. In parallel, those involved with separate research into rainfed or irrigated agriculture can be provided with opportunities to work more closely both with villagers and each other.

5.5.3 Incentives and motivation for change
Farmers and researchers rarely consider fully the costs of environmental degradation. Farmers themselves may not be sufficiently aware of the costs on their own farms, or the damage that they are causing occurs on land other than their own and they do not bear the costs. In Cameroon many farmers do not regard soil fertility as a problem (despite a general consensus that in west Africa soil degradation is the biggest problem for the sustainability of agriculture), in part because there are still opportunities for more extensive slash and burn agriculture (Sanchez, 2000). Similarly, researchers developing new approaches to crop intensification or pest management, for example, may not take into account environmental costs, as these may be cumulative over time, external to the individual farmer, or resources may be priced at below their “social cost” (subsidized water and electricity).         Ultimately, farmers are more likely to undertake longterm investments in improving the resource base on their farms if they face the true cost of any environmentally destructive practice (polluter pays principle), if they produce cash crops and have good access to markets for outputs and inputs, access to credit, and access to extension services (Reardon et al., 1995). Machakos, Kenya is a much cited example of an area where land degradation has been reversed and agricultural production increased despite increases in population. Factors that contributed to this success include good transport infrastructure to markets, secure land tenure and above average rural education and health (Toure and Noor, 2001).

Unless the full costs of environmental degradation and resource exploitation to farmers themselves (on-farm degradation), to the community (degradation of common pool resources such as forests), or to other sectors (pollution of down-stream water supplies) are quantified (both for current practices and proposed new practices) it will be difficult to persuade policy makers or farmers to adopt technologies and approaches that reduce the degradation.

The enabling and institutional environment is particularly important with respect to increased water exploitation. For farmers to choose to adopt efficient water techniques, not only must they be affordable for farmers, but appropriate institutions and incentives need to be in place, and farmer motivations and the links between water use and soil fertility better understood.

In the long run, realigning farmers’ incentives over their water use is essential for improving water efficiency and water equity. This entails appropriate mechanisms for allocating water—whether pricing, allocation of property rights, regulation, social pressure, or negotiation. The appropriate