De facto on-farm conservation by farmers without any external incentives is the most important way of conserving PGRFA at present. For instance, the Arguarana Jivaro community in the Peruvian Amazon grows 61 distinct cultivars of cassava, while some small communities in the Andes grow 178 locally named potato varieties (Brush, 1991; Bolster, 1985). From the national perspective, the area cultivated with the traditional varieties can be minimized while maintaining diversity. The high opportunity costs for land with farmers' varieties calls for action. While such a level of agrobiodiversity currently in situ may represent an optimum level of in situ conservation, the large group of marginalized farmers is utilizing far too much land with farmers' varieties than is needed to maintain a national optimal level, given the few alternatives open to them.
Although the high opportunity costs are reason enough to reduce the area under traditional varieties as much as possible, the high fixed costs of in situ conservation programs and projects make these questionable. Hence, a system of in situ conservation has to be found which is flexible enough to react when needed and is less expensive.
A preliminary approach at estimating the costs of in situ conservation is based on the introduced "controlled in situ conservation" system.6 Considering these details and taking into account that all agricultural crops consist of approximately 3 million distinct varieties (FAO, 1998), an MSS of roughly 15 million hectares of arable land must be utilized for a safe, but minimized in situ conservation; 1 % of the existing 1.4 billion hectares arable
5 As average figures they do not show the differences between countries or regions, so for instance, in sub-Saharan Africa, modern wheat varieties are planted only on 52% of the wheat area and in rice it may be even less.
6 See the explanations of the MSS on page 23 and Virchow (1999) for detailed information.
land (Engelman and LeRoy, 1995) is necessary to conserve the estimated 3 million distinct varieties.7
As long as the economic and technological development has not yet transformed all marginalized agricultural areas into high potential agricultural areas, much more than 1% of the arable land is still utilized with traditional varieties and landraces. In India, more than 60% of arable land is still cultivated with traditional or farmers' varieties (CMIE, 1988). Hence, in most cases no intervention is required; no financial costs will be incurred. In situ conservation, however, is supposed to be sustainable even in the face of changes of comparative advantages of traditional varieties to modern varieties. Consequently, a rough estimate has to guide the decision whether in situ conservation will be justifiable in the light of scarce financial resources.
In the first place, financing the controlled in situ conservation may lay in the hand of each country, which has the sovereignty over its genetic resources, as was stressed by CBD (UNEP, 1994). There may, however, be countries, which need more than the calculated 1% of their arable land for the in situ conservation of all distinct varieties because of their richness in agrobiodiversity. On the other hand, there may be countries, which need less than the average 1%. Because of the interdependency in PGRFA already utilized and needed in the future, the countries having to spend less than average may compensate countries which need more than 1% of their arable land.
India and Germany are examples of agrobiodiversity-rich and -poor countries; it can be stated that while in India land under agricultural use is still expanding to meet the increasing food needs, in Germany 12% of arable land is left as fallow due to the EC's set-aside programs (GCR, 1995). Consequently, the social costs on a national level for controlled in situ conservation, in terms of opportunity costs for foregone benefit, will be much higher in India than in Germany. From a simplistic point of view, the resources would seem allocated best if Germany could take over India's controlled in situ conservation. The different ecological conditions, however, restrict a transfer of in situ conservation activities from India to Germany. Consequently, Germany and other agrobiodiversity-poor countries could compensate India for the conservation, which is in a global interest similar to the joint implementation programs, known from the UN Framework Convention on Biological Diversity.
7 For the sake of comparison, India has set aside 4% of the country's surface for almost 500 wildlife protected areas (ICR, 1995).
Was this article helpful?