Water and available nitrogen as co-determinants of a mesic savanna in Kruger National Park, South Africa

• Main Author: Keretetse, Moagiemang Thomas
• Other Authors: February, Edmund C
• Format: Dissertation
• Language: English
• Published: University of Cape Town 2014
• Subjects: Botany
• Online Access: http://hdl.handle.net/11427/9276

Includes bibliographical references (leaves 66-75). === Savanna is a summer rainfall, tropical and subtropical vegetation type with a continuous grass cover and discontinuous cover of trees. Rainfall plays an important role in savanna as it triggers many ecosystem processes such as nitrogen mineralization. This rainfall is however extremely variable with cycles of wet and dry years. The impacts of rainfall variability on N mineralization rates and tree-grass dynamics are not well known. This study seeks to provide a quantitative understanding on the influence of rainfall amount on N mineralization rates, and how this interaction (of water and available N) may affect structural diversity in savannas. Rainfall manipulation treatments were set up to simulate different rainfall amounts. These treatments included a wet treatment which received normal rainfall regime plus irrigation( 150 % of rain). The control treatment had no rainfall manipulation and received the normal rainfall regime of the year (l00 % rain). The dry treatment received half of the normal rainfall (50 %). Furthermore, the treatments had different plant cover combinations of tree-grass, grass only and tree only. Volumetric soil water content and nitrogen mineralization rates were determined in all treatments from August to June for two growing seasons. N mineralization rates were determined by incubating soil cores in situ for +1- 28 days. Grass biomass, tree saplings height and basal area were measured to determine the treatment effects on plant growth. Leaf area and nutrient contents of leaves were quantified to establish physiological response of saplings to different rainfall regimes was also determined. N mineralization rate was highest in wet treatments (0.55 - 0.29 flg NI g soil! day) and lowest in dry treatments ( 0.33 - 0.11 flg NI g soil! day). Grass biomass increased in wet treatments (1901.3 - 2079.2 Kglha) and decreased in dry treatments (722.3 - 880.6 Kg/ha), while the saplings' growth decreased in wet treatments (190 cm) and increased in dry treatments (265 cm). However, saplings experienced greatest growth when grass was removed, than when grass growth was suppressed by dry conditions. These results show that increase in water and available N in wet treatments led to high grass productivity and intense competition on tree sapling which showed little growth. In dry treatments, grasses were limited by water and low N availability as shown by the reduction in grass productivity. Reduced grass competition in dry treatments, and lack of competition in grass removal treatments opened an opportunity for saplings to grow faster. This shows that grass competition plays an important role as it can limit saplings from growing beyond sizes where they are vulnerable to fire and browsing. This study shows that it is not only the impact of rainfall that is driving tree-grass coexistence and structural diversity in savannas, but the combined effects of rainfall and available