Tropical peatlands are one of the largest near-surface reserves of terrestrial organic carbon, and hence their stability has important implications for climate change. In their natural state, lowland tropical peatlands support a luxuriant growth of peat swamp forest overlying peat deposits up to 20 metres thick. Persistent environmental change-in particular, drainage and forest clearing-threatens their stability, and makes them susceptible to fire. This was demonstrated by the occurrence of widespread fires throughout the forested peatlands of Indonesia during the 1997 El Niño event. Here, using satellite images of a 2.5 million hectare study area in Central Kalimantan, Borneo, from before and after the 1997 fires, we calculate that 32% (0.79 Mha) of the area had burned, of which peatland accounted for 91.5% (0.73 Mha). Using ground measurements of the burn depth of peat, we estimate that 0.19-0.23 gigatonnes (Gt) of carbon were released to the atmosphere through peat combustion, with a further 0.05 Gt released from burning of the overlying vegetation. Extrapolating these estimates to Indonesia as a whole, we estimate that between 0.81 and 2.57 Gt of carbon were released to the atmosphere in 1997 as a result of burning peat and vegetation in Indonesia. This is equivalent to 13-40% of the mean annual global carbon emissions from fossil fuels, and contributed greatly to the largest annual increase in atmospheric CO(2) concentration detected since records began in 1957 (ref. 1).

In 1997-98, fires associated with an exceptional drought caused by the El Niño/Southern Oscillation (ENSO) devastated large areas of tropical rain forests worldwide. Evidence suggests that in tropical rainforest environments selective logging may lead to an increased susceptibility of forests to fire. We investigated whether this was true in the Indonesian fires, the largest fire disaster ever observed. We performed a multiscale analysis using coarse- and high-resolution optical and radar satellite imagery assisted by ground and aerial surveys to assess the extent of the fire-damaged area and the effect on vegetation in East Kalimantan on the island of Borneo. A total of 5.2 +/- 0.3 million hectares including 2.6 million hectares of forest was burned with varying degrees of damage. Forest fires primarily affected recently logged forests; primary forests or those logged long ago were less affected. These results support the hypothesis of positive feedback between logging and fire occurrence. The fires severely damaged the remaining forests and significantly increased the risk of recurrent fire disasters by leaving huge amounts of dead flammable wood.

Proteus mirabilis colonies exhibit striking geometric regularity. Basic microbiological methods and imaging techniques were used to measure periodic macroscopic events in swarm colony morphogenesis. We distinguished three initial phases (lag phase, first swarming phase, and first consolidation phase) followed by repeating cycles of subsequent swarming plus consolidation phases. Each Proteus swarm colony terrace corresponds to one swarming-plus-consolidation cycle. The duration of the lag phase was dependent upon inoculation density in a way that indicated the operation of both cooperative and inhibitory multicellular effects. On our standard medium, the second and subsequent swarm phases displayed structure in the form of internal waves visible with reflected and dark-field illumination. These internal waves resulted from organization of the migrating bacteria into successively thicker cohorts of swarmer cells. Bacterial growth and motility were independently modified by altering the composition of the growth medium. By varying the glucose concentration in the substrate, it was possible to alter biomass production without greatly affecting the kinetics of colony surface area expansion. By varying the agar concentration in the substrate, initial bacterial biomass production was unaffected but colony expansion dynamics were significantly altered. Higher agar concentrations led to slower, shorter swarm phases and longer consolidation phases. Thus, colony growth was restricted by higher agar concentrations but the overall timing of the swarming-plus-consolidation cycles remained constant. None of a variety of factors which had significant effects on colony expansion altered terracing frequencies at 32؇C, but the length of the swarming-plus-consolidation cycle was affected by temperature and medium enrichment. Some clinical isolates displayed significant differences in terracing frequencies at 32؇C. Our results defined a number of readily quantifiable parameters in swarm colony development. The data showed no connection between nutrient (glucose) depletion and the onset of different phases in swarm colony morphogenesis. Several observations point to the operation of density-dependent thresholds in controlling the transitions between distinct phases.Proteus mirabilis colonies have fascinated microbiologists for over a century (13). On typical laboratory media, mature P. mirabilis swarm colonies display striking patterns characterized by circular symmetry and regularly spaced concentric terraces or zones (Fig. 1) (6). These terraces develop as a result of periodic events during colony growth, most notably the cyclic repetition of alternating phases: swarming (active migration) and consolidation (growth without movement of the colony perimeter) (2, 3). Colony expansion is a dynamic process involving movement over the solid substrate by multicellular rafts of specially differentiated swarmer cells (7,18,30,32). The swarmer cells are elongated and hyperflagellated but have the same DNA/length ratio as the shorter oligofla...