Browsing by Author "Kooverjee, Neha"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    Landscape-scale carbon stocks in the subtropical thicket of the Baviaanskloof, South Africa
    (University of the Witwatersrand, Johannesburg, 2024-08) Kooverjee, Neha; Smart, Kathleen; Fisher, Jolene; Scholes, Bob
    South Africa’s subtropical thicket biome has garnered interest for its potential for carbon sequestration through the restoration of degraded land, funded locally and internationally by the carbon market. Accurate carbon estimation of pre- and post-restoration carbon storage is key to monitoring the progress of restoration efforts and determining accurate estimates of carbon increase. Most past thicket studies have based landscape-scale carbon estimates on plot data alone, without accounting for the spatial variation in factors such as soil depth, resulting in overestimates of the average potential carbon that can be stored. The aim of this research was to estimate and create maps showing the spatial distribution of the thicket carbon stocks in the Baviaanskloof Nature Reserve, South Africa, by calculating the amount of carbon stored in the soil, woody biomass, root biomass and litter biomass across the thicket area of the reserve using plot data with remote sensing imagery. Soil organic carbon was estimated using soil profile data collected in the same primary catchment as the Baviaanskloof and extrapolated across the landscape using a Random Forest-based regression model. Woody carbon was estimated using previously recorded in situ carbon recordings. This data was then extrapolated across the landscape using raster data for vegetation height and cover, climate data, and elevation as predictor variables. Root and litter carbon were then calculated individually based on the woody carbon estimates and ratios of root and litter carbon to woody carbon from a past study. The average soil organic carbon (37.7 ± 29 t ha-1), woody carbon (18.1 ± 4 t ha-1), root carbon (3.1 ± 0.7 t ha-1), and litter carbon (2.4 ± 0.5 t ha-1) estimated in this study for intact thicket are all lower to varying degrees than those of past thicket studies that were based solely on plot data, indicating the importance of accounting for variation in the landscape and its effect on carbon stocks. The results of this study indicate that slope position, a proxy for soil depth, has a significant effect on thicket carbon stock. Woody carbon, root carbon and litter carbon estimates significantly decreased downslope, but soil organic carbon, the largest carbon pool in thicket, significantly increased downslope. In general, the results also indicate an increase in carbon after restoration, however, an illogical trend of lower potential root and litter carbon compared to current carbon in degraded thicket is present, indicating that the results of this study are not reliable for restoration planning and decision-making. This may be due to the inaccuracy of global Light Detection and Ranging (LiDAR) products, such as the Global Land Analysis and Discovery (GLAD) canopy height dataset, over non-forested areas such as thicket which are much shorter. Accurate estimates of current and potential carbon and accurate monitoring of restoration is important in tracking South Africa’s progress towards its biodiversity conservation goals under the Convention on Biological Diversity, and carbon emission reduction goals under the Paris Agreement. With access to locally captured LiDAR imagery that is suited to the short, dense structure of thicket vegetation and the topography of the landscape, and a larger database of in situ thicket carbon recordings, the methods used in this study could be employed to calculate more accurate carbon estimates and aid in directing restoration efforts to the most appropriate areas.