Peasant agroecosystems are seen as a continuum of integrated farming units and natural ecosystems where plant gathering and crop production are actively practiced Many of these traditional agroecosystems still found throughout developing countries constitute major in situ repositories of both crop and wild plant gemzplasm. These plant resources are directly dependent upon management by human groups; thus, they have evolved in part under the influence of farming practices shaped by particular cultures. Because genetic conservation programs are more effective when preserving the ecosystems in which the resources occur, maintenance of traditional farming systems and adjacent natural ecosystems is proposed as a sensible strategy for in situ preservation of crop and wild plant genetic resources. It is here argued that preservation efforts should be linked to rural development projects that take into account the eth‐nobotanical knowledge of rural people and that emphasize both food self‐sufficiency as well as local resource conservation. Preservation of these traditional agroecosystems cannot be achieved when isolated from maintenance of the culture of the local people. Therefore, projects should also emphasize maintenance of cultural diversity.

No abstract

No abstract

No abstract

Fires burn vast areas of the monsoonal savannas of northern Australia each year. This paper describes the contemporary fire regimes of two ecologically similar, relatively large national parks (Litchfield—1464 km2; Nitmiluk—2924 km2) in the Top End of the Northern Territory, over 8 and 9 years, respectively. Fire histories for both parks were derived from interpretation of LANDSAT TM imagery, supplemented with NOAA-AVHRR for cloudy periods at the end of the 7-month dry season (c. April–Oct). Data concerning seasonality, extent and frequency of burning were analysed with respect to digital coverages for the park as a whole, landscape units, vegetation types, infrastructure and tenure boundaries. Ground-truth data established that interpreted accuracy overall, for 2 assessment years, ranged between 82 and 91% for both parks. Over 50% of Litchfield and 40% of Nitmiluk was burnt on average over this period, with Litchfield being burnt substantially in the earlier, cooler, and moister, dry season, and Nitmiluk mostly in the parched late dry season, after August. On both parks the current frequency of burning in at least low open woodland / heath habitats is ecologically unsustainable. Both parks are prone to extensive fire incursions. The data support earlier regional assessments that the average fire return interval is around 2 years in at least some areas of northern Australia. Nevertheless, comparison of contemporary fire regimes operating in three major regional national parks shows distinct differences, particularly with respect to the extent and seasonality (hence intensity) of burning in relation to different landscape components. Management implications are considered in discussion.