Abstract: | Periodic burning of the vast Imperata grassland areas of Indonesia is thought to have a number of negative consequences, both on and off site. We assess on-site biophysical and economic consequences by adapting a bioeconomic model to conditions prevalent in the Indonesian uplands. Burning is shown to exacerbate soil degradation both directly through loss of soil nutrients and indirectly through erosion. Despite this, it is demonstrated that burning is the most profitable method of Imperata control in an upland ‘shifting cultivation’ system Changing factor prices may alter this A 25% reduction in herbicide prices would make herbicide use more attractive than burning. If off-site costs are considered, Imperata control with herbicide may be preferable to burning. Upland cropping within an Imperata fallow system is marginally profitable under prevailing economic and biophysical conditions Abandonment of these upland areas by smallholders, or adoption of more productive farming systems, seems inevitable. |