http://ag.arizona.edu/OALS/oals/npc/newcrops.html - Overview
Roughly one-third of the world's land mass is chaparral, scrub, or dry desert where conventional crops cannot be produced without irrigation. However, the world's burgeoning demands for food and fiber have placed increasing pressure on these marginal regions where attempts to cultivate crops are being initiated now more than ever before. Generally, the consequences have ranged from disappointing to disastrous.
When prolonged irrigation and other intensive agricultural practices become established in arid and semi-arid lands, soil salinity buildup and severe erosion inevitably follow. These problems are now major factors contributing to a worldwide epidemic of desertification and cropland loss. Approximately one-third of the world's 160 million hectares of irrigated lands are already affected by salinity problems, and cost-intensive methods are required to prevent more widespread damage. A further problem with irrigation is that it often relies on non-renewable underground water reserves. As the aquifer is depleted, the pumping costs and energy requirements increase - thus compounding the already high price of irrigation. Farm profits are routinely swallowed by these costs in the southwestern U.S., where thousands of acres of abandoned croplands stand as evidence of a single fact: conventional agriculture in arid lands is ultimately a self-limiting endeavor.
The most practical solution for the long term is likely to involve a change in current agricultural strategies, ultimately leading to the cultivation of crops that are better adapted to the desert. Diversified cropping schemes requiring little or no irrigation would bring with them corresponding reductions in the problems of erosion and soil salinization; such crops also would have greatly reduced growing costs. And a crop uniquely adapted to an arid habitat could be grown on land where agriculture is not impractical or impossible, thus conferring an additional cost advantage by avoiding competition with more traditional crops.
Conventional wisdom would imply that farming in the desert is a pipe dream In fact, many areas where rainfall cannot support conventional crops are still surprisingly productive of biomass. Uncultivated desert and semidesert regions generate about one-fourth as much of the world's primary productivity as do all the cultivated lands. Clearly, the potential exists for beneficial use of these low-rainfall areas.
The key to obtaining resources from marginal lands without causing environmental damage lies in understanding the unique adaptations, productivity, and limitations of plant life in the desert. These ecosystems are characterized by low standing biomass, although many desert-adapted species exhibit a very high photosynthetic conversion efficiency during favorable parts of the year and are able to make use of short and erratic periods of rainfall.
Traditionally, plants that have economic or ethnobotanical importance in the desert, other than food crops, are most likely to be species that produce compounds rich in carbon and hydrogen, such as rubber, paraffins, resins and lower-molecular-weight specialty chemicals. Although they have a slower overall growth rate than conventional crops, these plants show increased water economy by producing a high energy content per unit of dry weight biomass. When water is limited, any plant - regardless of its ability to tolerate heat and drought - will produce a relatively low amount of biomass. Thus, the most logical approach to bioresource utilization in the desert is to breed and cultivate plants that produce oils, resins, and other plant substances that have a high value per unit of volume.
For more information -
http://ag.arizona.edu/OALS/oals/npc/npc.html
Forum 
