Diversity by Design

by Larry Harrington

Our common goals

There are a number of reasons for fostering diversity in agroecosystems. More diverse systems take better advantage of ecological niches. Species adapted to different stresses (e.g., waterlogging, soil acidity) can be positioned where they have a comparative advantage. Greater system diversity can also improve stability and resilience. Diverse agroecosystems offer multiple pathways for energy and nutrient cycling; consequently system productivity is not held hostage to the performance vagaries of any particular species. When properly designed, more diverse systems also can reduce problems associated with pests, diseases, and weeds and can decrease reliance on external inputs. Such diverse systems may also be associated with in-situ conservation of foodgrain land races and folk varieties.

However, agroecosystem biodiversity is not an end in itself but a means of achieving productivity, stability, resilience, improved environmental quality, and the conservation of crop genetic diversity. These in turn are part of larger societal goals -- sustainable food security, reduced poverty, and improved public health.

Societies also value natural biological diversity in the broader sense. People are concerned about the possible extinction of species because of their potential future benefits, their role in ecological balances, and simply because people place a value in their continued existence, regardless of future human benefits.

Kinds of diversity

Agroecosystem biodiversity can be understood in several different ways:

Crop genetic diversity. This embraces such factors as varietal concentration; pace of varietal change over time; genetic similarity among major cultivars; the conservation and pyramiding of favorable genes in breeders' varieties; the conservation and use of important genes present in folk varieties, land races, and wild relatives; and opportunities for expanding crop genetic diversity through wide crosses and biotechnology.

Crop species diversity over space. Spatial species diversity may be exceedingly narrow (e.g., a monocropped rice field) or exceedingly broad (e.g., a home garden featuring simultaneous cultivation of fruit trees, banana plants, coffee, spices, and several food crops). Plots with low species diversity and high species diversity often are found within the same farming system.

Crop species diversity over time. Temporal species diversity may be narrow (e.g., one maize monocrop crop per year, every year); broad within a year (e.g., an annual sequence of multiple cropping involving cereals, legumes, and horticultural crops); or broad over several years (e.g., rice-potato-wheat patterns, broken every few years by a sugarcane crop). Crop species diversity over space and over time are not necessarily related.

Agroecosystem biodiversity through crop-livestock interactions. The presence of livestock in a system tends to greatly enhance the value of non-crop components (crop residues, grazing lands, forest resources) and typically features nutrient cycling between rangeland and crop land, thus fostering improved productivity and sustainability of cropping systems and a higher potential for spatial and temporal crop species.

Natural biodiversity within agroecosystems. More diverse agroecosystems -- particularly those with greater spatial diversity, and those with trees -- may provide habitat for a wider array of wildlife.

Natural biodiversity as indirectly affected by agroecosystems. Highly productive agroecosystems can indirectly foster natural biodiversity by making it unnecessary to farm marginal or fragile areas, or to clear new forest areas for agriculture. System diversity may be broadened by increasing crop genetic diversity, expanding crop species diversity over space and time, fostering crop-livestock interactions, or improving productivity in favored agricultural areas to protect biologically diverse fragile, marginal, or forested areas from agriculture.

Sustainability in perspective

Partly because they are thought to be closely linked to issues of biological diversity, sustainability issues are commanding more attention from agricultural scientists. Many agree that the ability to quantify sustainability is fundamental to making the concept useful. Unfortunately little progress has been made in this regard.

Most commonly, indicators of sustainability are narrowly driven by definitions. This often leads to arguments that are merely circular. For example:

• If agroecosystem sustainability is defined in terms of zero external input use, then any technical change leading to reduced external input use can be said to foster sustainability.
• If agroecosystem sustainability is defined in terms of low levels of environmental pollution, then any technical change leading to less environmental pollution can be said to foster sustainability.
• If agroecosystem sustainability is defined in terms of high agroecosystem biodiversity, then any technical change leading to higher agroecosystem biodiversity can be said to foster sustainability.
• If agroecosystem sustainability is defined in terms of local self-reliance in agricultural production (i.e., avoidance of international markets), then any technical change leading to greater local self-reliance can be said to foster sustainability.

All of these definitions, and their corresponding indicators, are inadequate -- even when combined. They emphasize the plot or farm community level of analysis, ignoring higher levels. They succumb to the "fallacy of scale," in which something that appears unsustainable at one level of analysis may be a strong element in favor of sustainability at a higher level.

An example of "fallacy of scale"

Green Revolution technologies for rice and wheat in South Asia often have been criticized as unsustainable. At the agroecosystem level, these technologies may feature low species diversity, high reliance on external inputs and energy sources, and environmental pollution from pesticides and fertilizers. However, at higher levels of analysis, the diffusion of Green Revolution technologies in parts of South Asia has been associated with accelerated economic development in Bangladesh; higher incomes through employment generation in Uttar Pradesh; improvements in income distribution in Pakistan; reduced population growth in Green Revolution areas of India -- and, not least, the saving of approximately 40 million ha from the plow (or woodcutter's ax) in India alone. Without the Green Revolution technologies, India would have needed another 40 million ha of rice and wheat area to meet foodgrain demand.

The Green Revolution played a virtually unrecognized role in reducing pressure to cultivate biologically diverse fragile, marginal, or forested areas. In the absence of the Green Revolution, food prices would have been higher, employment growth (especially off-farm employment) would have been slower, poverty more widespread, and population growth more rapid -- exacerbating the threat to natural biological diversity.

So, at a higher level of analysis, resource degradation and environmental pollution in Green Revolution areas has been a cost associated with defusing longer-term threats to resource quality and natural biological diversity in biologically diverse fragile, marginal, or forested areas. Researchers and farmers must reduce this cost. Plot-level threats to sustainability in Green Revolution areas must be addressed. The challenge is to generate a "doubly green revolution" that maintains the powerful and favorable indirect consequences of highly productive agricultural technology, while improving resource quality and reducing pollution. Sustainability is not enough -- productivity must increase as well.

Diversity by design

Greater agroecosystem biodiversity -- particularly crop genetic diversity and spatial and temporal species diversity -- often can help achieve sustainable improvements in agricultural system productivity. How, then, do we foster widespread use of more biologically diverse agroecosystems? There are at least two ways:

• Diversity by design -- Researchers and farmers design more biologically diverse agroecosystems, which are widely adopted by farming communities. This process includes participatory research on indigenous technical knowledge about system diversity, with a view to extrapolating such knowledge to comparable areas.
• Demand-led diversification -- Higher incomes and reduced poverty generated by more productive agricultural practices shift the structure of food demand towards a more diverse array of products, such as fruits, vegetables, and animal products. Farmers follow market signals and diversify their farming systems.

The path of "diversity by design" is direct. It is the path taken by cropping systems and farming systems research (FSR). In Asia such research sought to diversify and intensify cropping patterns by introducing a second non-rice crop into rice-based systems (an alternative made possible by the introduction of short-duration, non-photoperiod sensitive rice varieties). In Africa, the emphasis has been less on intensification and more on reconciling food security and system sustainability requirements. Even in Africa, however, diversity has been a major theme in FSR. "Diversity by design" also has been characteristic of research on agroecology.

The lessons learned from FSR and research on agroecology have vastly improved researchers' capacity to work with farmers in understanding and improving farming systems. However, these lessons have not led to widespread adoption of more diverse, productivity-enhancing resource-conserving agricultural systems. Work is urgently needed to improve the effectiveness of research (measured by widespread adoption) aimed at designing such agricultural systems. Until then, the path of "diversity by design" is unlikely to help achieve our common goals.

Demand-led diversification

The path of "demand-led diversification" is relatively indirect but has induced widespread change in farming systems, particularly in Asia. Indonesia provides a classic example. Increased rice productivity in favored lowland areas expanded the supply of rice and reduced its price. Marginal rice areas, often on hillsides, became unprofitable and farmers in these areas ceased producing rice. However, higher incomes in rural and urban areas -- largely resulting from improved rice productivity and lower rice prices -- shifted the structure of demand for food towards fruits and vegetables. Partly as a consequence, farmers in hillside areas switched from growing rice to perennial fruit trees. The same process is evident in other areas where new technology has increased the productivity of basic grain production, for instance in the Indian Punjab.

In contrast, stagnating grain (maize) productivity in southern Africa provoked a chronic food security crisis, the expansion of maize cultivation into wildlife areas, and a relative absence of market signals that would induce farmers to diversify into cash crops. Not by coincidence, "diversification out of grain production" is not high on the southern Africa research and policy agenda.

Success in demand-led diversification is sensitive to the policy environment, requiring:

• an overall policy environment that encourages more flexible and broader cropping systems rather than commodity-support programs;
• laws and institutions that facilitate efficient marketing by establishing grades and standards for different commodities and developing and distributing farm inputs;
• public investment in physical and social infrastructure, communications, and information systems; and a rural financial system that mobilizes rural savings, makes credit available to traders, and diversifies the rural economy; and
• training and education to prepare rural people for non-agricultural jobs.

Demand-led diversification will lead to more biologically diverse agroecosystems at the aggregate (e.g. regional) level but may not ensure increased biodiversity at the plot or farm level. Moreover, plot-level trends in resource quality, external input use, and environmental pollution may increase or decrease, in accordance with practices adopted as farmers learn to manage a new set of enterprises. The path of "demand-led diversification," on its own, also may not lead to the attainment of our common goals.

Complementarity and competition

As noted earlier, the goals of sustainable food security, reduced poverty, improved public health, and conservation of natural biological diversity will be attained through widespread use of more productive, stable, resilient agroecosystems and the conservation of crop genetic diversity.

This will require an emphasis on sustainable productivity improvement in favored areas -- to reduce pressure to cultivate biologically diverse areas unsuited to agriculture, foster "demand-led diversification," and lead to the "doubly green revolution" described previously.

Agricultural research and development -- scientists, extension workers, farmers, and policy makers -- can help through: cereal varieties that are more tolerant to biotic and abiotic stresses and use nutrients more efficiently; productivity-enhancing resource-conserving crop management practices such as integrated pest management (IPM) or reduced tillage; and more effective "diversity by design" through adoption of new cropping patterns, farming systems, and land management systems (featuring staple cereals) that capitalize on the advantages of system diversity to sustainably improve productivity.

Sustainable productivity improvement in marginal, fragile areas -- acknowledging that we can reduce but not eliminate pressure to cultivate biologically diverse areas unsuited to agriculture -- would again require varieties, crop management practices, cropping patterns, farming systems, and land management systems that can offer leverage points.

As "demand-led diversification" takes hold, scientists, extension workers, farmers, and policy makers must foster sustainable management strategies. A greater temporal and spatial diversity of enterprises does not necessarily imply improved sustainability. New fruit and vegetable crops can exacerbate or ameliorate problems of erosion, soil fertility loss, water-induced land degradation, external input dependence, or environmental pollution.

In the end, it is not a case of competition, but of complementarity. It is not a case of "Green Revolution" vs. "alternative agriculture," or "diversity by design" vs. "demand-led diversification." We must use all of the tools at our disposal -- following all promising paths -- to reach our common goals.

Larry Harrington is Manager, Natural Resources Group, Centro Internacional de Mejoramiento de Maiz y Trigo (CIMMYT)

Return to CGIAR Newsletter Table of Contents