Browsing by Author "Wills, Cameron"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- ItemOpen AccessThe differences in grass species composition and the effects on fire behaviour in an African mesic savanna(2015) Wills, Cameron; Bond,William JFire is a key determinant of savanna dynamics, particularly with respect to its influence on vegetation dynamics. The grass species composition and structure of savannas have impacts on fire behaviour through their differing fuel characteristics related to each species specific phenologies and morphologies. Climate change and elevated atmospheric CO2 levels may influence the trajectory of grassy ecosystems over a large spatial and temporal scale. These dynamics operate through to competition between C4 and C3 plant, where C3 plants have a competitive advantage under conditions of elevated atmospheric CO 2 levels . This competitive advantage may be mitigated by landscape scale disturbances, including fire and small disturbances, such as game trails created by animals. Different grass clades appear to differ in productivity and moisture content with Andropogoneae (Themeda triandra and related species) being highly productive , but drying out early after exposure to drought compared to Paniceae (Panicum maximum and related species), which holds moisture for longer in the dry season. This study begins to delve in how these processes interact in these grassy mesic savanna ecosystems. The focus of this study is upon the differences in the fire ecology of grasslands dominated by Themeda triandra (Themeda) and Panicum maximum (Panicum) and the effects that game trails surrounding thicket patches have on fire continuity and spread. From a fire perspective, fuel load and moisture content are major contributors to fire spread. Grass biomass contributing to fuel was found to differ between the two respective dominant grass species. Themeda (5000 - 6500 kg/ha) had significantly higher fuel loads than Panicum (3 00 0 - 400 0 kg/ha) (p<0.001). The fuel moisture content of Panicum was significantly higher than the moisture content of Themeda, where the latter cured more rapidly than the former through the dry season (p<0.001). Therefore, rates of curing Panicum swards were influenced by season (p<0.01). It is postulated that these differences would affect the fire behaviour in an African mesic savanna. The experimental fires were affected by the timing in the dry season of the fires (p=0.03), the grass species composition (p=0.02) and the grassy fuel loads (p<0.001). The rates of spread of the fires increased with increasing fuel loads (R2 =0.71). There were higher fire intensities in grasslands dominated by Themeda (~5500 kW/m) than observed in grasslands dominated by Panicum (~2000 kW/m) (p=0.01). Themeda swards were able to burn with high fire intensities at any time in the dry season, whereas swards dominated by Panicum burnt more intensely later on in the fire season. Animals were able to impact the behaviour of fire, not only through herbivory, but through the creation of game trails around thicket patches. The presence of these trails around thicket patches were able to impede fire continuity and spread, thus "protecting" thicket patches (p<0.001). This was enhanced by the grass species composition around the thicket edges being largely dominated by Panicum; this may create an "edge effect" that could mutually reinforce the "game trail" effect in protecting thicket patches. Changing climate and elevated CO2 conditions and disturbances through fire and animals may affect the species composition, structure and functioning of mesic savanna landscapes . Managers of African savannas can manipulate fire intensity by choosing the time in the season to burn, and further by burning in years with higher or lower fuel loads. If the management objective were to maintain savannas dominated by C 4 grasses, then the management protocol would be to use of frequent, high intensity fires that may provide C 4 plants with a competitive advantage over C 3 plants and mitigate the spread of C 3 plants into the now C 4 dominated systems. This may be able to restrict the encroachment of woody species typical of the savanna thicket stable state in open savannas that has become an increasing phenomenon in these landscapes.
- ItemOpen AccessEffects of repeated fire on the savanna / forest boundary(2012) Wills, Cameron; Bond, William JSavanna and forest/thicket can exist as alternate stable states, among others, determined by fire ecology feedbacks. Bush encroachment has become an ever-increasing trend converting grassland and savanna biomes to forest/thicket. A severe firestorm occurred in the Hluhluwe-Umfolozi Game Reserve in north-eastern South Africa in September 2008. The fire penetrated closed thicket areas and opened up the landscape. The main aim of this study was to investigate whether repeated fire, following such an extreme fire event, could trigger a biome or regime shift; in this case from forest/thicket to savanna. Fire spread is determined by grassy fuel loads, primarily grass biomass in these ecosystems, and the fire weather at the time of the fire. Grass biomass was found to have significantly accumulated (p=0.0002) in the thicket areas in just three growing seasons since 2008, which allowed fires to burn the area again in 2012. In the 2012 fires, fire intensity, measured by char height on woody stems increased in relation to the increase in grass biomass up to a point after which increasing grass biomass had no effect. The 2012 fires were able to penetrate areas opened up by the 2003 firestorm despite high canopy cover created by tree resprouting. Tree mortality was cumulative with repeated burning (21% mortality post 2008 increasing to 47% mortality in 2012). This was linked to the vigour of post 2008 resprouting with much higher mortality of trees in 2012 that had few (<4) resprouting stems. Thus, this study indicates that, a fire regime of sufficient frequency and intensities could potentially cause a vegetation state shift from forest/thicket to savanna. Therefore, this may provide management options for wildlife parks and protected areas.