At the moment, about a fifth of the Kruger National Park's 19,500km2 has been consumed by fire. Most fires have been planned patch burns ignited by section rangers in an effort to meet their burning targets for this fire season, although for ecological reasons park management does not distinguish between the different sources of fire in the park. Fires have also burnt across the eastern border from Mozambique.
Over the years Kruger's fire policy has evolved as scientists have come to better understand the role of fire in the ecosystem. Along the way, the scientists and park managers have found that what determines how much of the park burns each year is not what action management takes in either starting or fighting fires, but rather how much grass is standing in the bush.
This is in turn determined by how much rain has fallen in previous years. This is part of the reason why the park does not distinguish between fires started by careless tourists, illegal immigrants, lightning or section rangers, although these details are still recorded for operational reasons. The fires started by the section rangers are intended to break Kruger up into a patch-mosaic of burnt and unburnt areas. This acts a means of preventing runaway veld fires that are started by other means.
Planned fires are usually set when there is still some green grass in the veld, which makes for a cooler fire. Cooler fires do less damage to young trees, making them less likely to turn into stunted, multi-stem trees. However, sometimes the park burns with the intention of dealing with bush encroachment.
When this is the case, the fire regime is altered to include hot fires (usually started at this time of year) and later cool fires that kill the encroaching trees and give the grass in the area a chance to flourish. To determine where planned burning should take place, the section rangers and other staff regularly visit more than 520 vegetation monitoring sites throughout the park, where they measure how much grass is standing in the veld.
The measurements taken on these plots are then fed into a model, which determines how much land should be burnt in each section. This analysis is carried out by Navashni Govender, programme manager for fire ecology, who then informs each section ranger how much of their area should be burnt that year.
It is then up to the section rangers to decide when, where and under what weather conditions to start their fires to break up the fuel load in the veld. The section rangers have recently been assisted in their burn planning by the presence of new automatic weather stations at Shingwedzi and Satara.
These weather stations take readings of the weather and translate them into a fire danger index, which is sent out twice a day to the rangers. High temperatures, low humidity and high berg winds are likely to create runaway fires, so planned burning is avoided at these times. Some burning is also carried out by scientific services, who are in charge of one of the world's longest running experiments on the effects of fire on savanna ecosystems.
The burns conducted in Kruger's four strings of Experimental Burn Plots are guided by a historic fire regime which includes annual burning in winter, no burning and burning every two and three years in winter, autumn, spring and early and late summer.
The areas of the park that have been burnt can be determined from satellite images. Twice daily, the park downloads images from the MODIS satellite. These images are analysed to reveal burnt patches, which are backed up by ranger reports from the ground. The digital maps are kept in the fire ecology section of scientific services.