60 years on, the Experimental Burn Plots of the Kruger National Park continue to provide new insights for savanna ecology

By Tony Swemmer, Manager, SAEON Ndlovu Node Tweet

In 1954, conservation managers in the Kruger National Park initiated what would become one of the longest-running fire experiments in the world, amid concerns about the lack of information regarding the effects of fire on the fauna and flora of the park.

At the time, burning of natural grassland and savannas was a highly contentious issue. Many felt that fire could only have a negative effect on nature.

However, a fire suppression policy that had been in place since 1948 had led to new concerns about bush encroachment and deterioration of grass production in the complete absence of fire. Furthermore, the accumulation of dead grass during the fire suppression period was resulting in wildfires that were large and difficult to control (about a quarter of the park was burnt by unintended fires in 1953).

In response, the newly formed “Biology Section” of the Kruger National Park set up “burning trials” at four sites, in four of the major ecosystem types in the park. The design was large and ambitious - four replicated “strings”, each consisting of seven plots, were laid out at each site, with each plot an impressive seven hectare in size.

Each plot within a string was assigned a different burning treatment including a control (no burning) and various combinations of fire frequency (once a year or once every three years) and season (winter, spring or summer). These plots are clearly visible from space, along with the double roads which enclose a fire break on each side of each string.

The vegetation of the plots was surveyed at the start of the experiment, with the intention of resurveying every year. However, this proved to be too ambitious due to the number and size of the plots, and the large numbers of small trees and shrubs present.

It took three years just to do the initial survey of all plots. Management were able to successfully implement, and record, the burns planned for each plot every year, resulting in a long-term experiment that is now of immense value for understanding the role of fire on savanna ecosystems.

Researchers from around the world have recognised the value of this unique ecological research infrastructure, now known as the Experimental Burn Plots (EBPs) of the Kruger National Park, and many research projects have been conducted on the EBPs since the late 1990s. Results produced by these researchers have in turn assisted staff of the SANParks Scientific Services division to motivate for the budgets required to keep the experiment going (controlled burning of up to 144 plots a year at remote sites within a large national park is not cheap).

Impact of factors other than fire

The carefully maintained burning treatments, and associated data, of the 60-year-old EBPs are now allowing researchers to determine the roles of factors other than fire that affect the composition and functioning of savanna ecosystems. Of particular interest are two factors that have changed substantially since the initiation of the experiment: elephants and atmospheric carbon dioxide (CO₂).

In the 1950s, elephant numbers in the park were still low and elephant impacts on vegetation negligible. Atmospheric CO₂ was not much higher than pre-industrial levels. In order to test how these factors might be altering the effect of fire on vegetation, the SAEON Ndlovu Node and SANParks Scientific Services teamed up to resample the trees at all four sites over the summer of 2015-2016.

Surprising results

Analysis of these data, together with the data collected from all previous samples - in the 1950s, 1970s, 1990s and 2000s, has produced some surprising results.

The density of trees and shrubs has clearly declined at the two drier sites (Mopane and Satara) regardless of burning treatment. This reflects the increasing impacts of elephants over recent decades, and demonstrates that fire is not required in these systems to prevent bush encroachment above a certain population density of elephants.

The EBP site of intermediate rainfall, located near Skukuza camp, has experienced high variable trends in tree and shrub density, most likely reflecting complicated interactions of soil types, fire and elephants.

At the Pretoriuskop site (“Pkop” in the above graph), tree and shrub densities have increased, even on plots burnt every year. This site, which has the highest rainfall, now supports far higher densities than any of the other sites, as well as most parts of the park.

It is possible that plants at this site have been able to utilise elevated levels of atmospheric CO₂ to overcome the harmful effects of fire (the exact mechanism behind this is explained in SAEON’s “Change is in the Air” booklet). However, closer examination reveals that while increases were dramatic between the 1970s and 1990s, there has been no significant increase in tree and shrub density since then, even on plots that never burn. This suggests that any effect of CO₂ is now over, perhaps due to counteracting effects of increasing elephant impacts, or due to increasing competition between trees and shrubs.

Further research

Further research is needed to determine the exact causes of these trends, and continued implementation and sampling of the EBPs will play a critical role in helping ecologists to determine the true causes of past (and future) changes in the structure and functioning of our savannas.

For more information on the design and history of the EBPs, the following paper can be downloaded from Koedoe, the academic journal of SANParks: http://koedoe.co.za/index.php/koedoe/article/view/35