Albany thicket is prevalent in the Eastern Cape Province of South Africa. Its spread has diminished through overgrazing and heavy browsing by animals, land clearance and urban expansion. The result is highly degraded land characterized by invasion of alien species. There is a wealth of documented evidence on the high carbon sequestration ability of thicket biome, but not much has been done to assess its effect on carbon dioxide emissions from the soil. Given that the concentration of atmospheric greenhouse gases has been constantly rising since the industrial era, it is imperative to assess the influence of thicket biome as a source or sink of these gases. There is evidence of shifts in the climate in southern Africa as reflected by changes in rainfall patterns, increased temperatures, recurrent droughts and fires. As such, the historical rainfall variability in an Albany thicket region and its interaction with the temporal land use / cover changes was studied. This served to give some background information about the study area for more detailed study on C and carbon dioxide effluxes in thicket vegetation under different levels of degradation. This study thus aimed to determine the influence of thicket vegetation at various levels of degradation on soil carbon and carbon dioxide fluxes. The impact of plant photosynthetic pathway on soil C residence time and gas effluxes were analysed to elucidate on the land-use and cover patterns occurring in the area. All this was done to shed some light on the role of soil and thicket vegetation on carbon dioxide emissions and C storage in the spectrum of a shifting climate. The main area of research was Amakhala reserve in an Albany thicket in Eastern Cape Province; and it concentrated on three land cover types namely intact thicket, degraded thicket and grassland. The objectives mentioned above were achieved by assessing historical rainfall variability from 1970 to 2010 through trend and time series analysis at nine rainfall stations located at Amakhala reserve, Grahamstown, Bathurst, Port Alfred, Uitenhage and Port Elizabeth. The land use changes that have occurred in the Albany thicket region covering Amakhala reserve, Grahamstown, Bathurst and Port Alfred were also assessed for 1989, 1999 and 2009 through satellite image analysis with Idrisi Andes GIS software; then their interaction with rainfall variability were determined. To elucidate on the vegetation species composition and land use / cover changes that have occurred in the study area, plant biomass as well carbon (C) and nitrogen (N) isotope measurements were done. Plant biomass was assessed for the dominant species through use of pre-existing allometric equations that required data on plant basal diameter, canopy area, stem numbers and height. The plant carbon was then estimated through use of a conversion factor of 0.48 on above-ground biomass, while soil organic C was determined through the modified Walkely - Black method. Carbon and N isotope ratios were determined from the foliar material of three replicate samples of dominant plant species then analyzed through mass spectrometry. Soil carbon dioxide effluxes were then monitored in each of the intact thicket (IT), degraded thicket (DT) and grassland (G) over a 10 month period; by measuring the net carbon dioxide exchange rate (NCER) through the dynamic chamber method. An automated carbon dioxide exchange analyzer, coupled to a soil temperature probe and photosynthetic active radiation (PAR) sensor was used; with NCER measurements taken every 20-30 days. Soil temperature, moisture, penetration resistance and PAR readings were taken during each assay and later used to interpret the NCER. Results showed that long term variability in annual rainfall had a declining trend at Grahamstown (r = -0.59), Uitenhage and Bathurst stations (r = -0.32 at both stations), but was not significant at Amakhala, Port Alfred and Port Elizabeth stations. Most reductions in rainfall occurred in the 1980s and 1990s with the autumn, winter and summer rainfalls, the daily rainfall index and the daily rainfall subclasses of 10 mm and above showing a similar trend. The land use change detection gave a significant increase in proportion of degraded and transformed (moderately degraded) land between 1989 and 2009 with most of the increases occurring from 1989 to 1999, while farmland area decreased by 1.8 percent over the years. Thus the Albany region had over 30 percent of its land occupied by transformed vegetation, with heavy browsing and uncontrolled grazing being attributed to the destruction of pristine vegetation. Land-use change to game ranching and goat pastoralism was attributed to the reduction in farmland. Rainfall variability – land use change linkages were most significant in 1999 that recorded the least rainfall and had the lowest mean, maximum and sum of the NDVI. Grahamstown had the most significant rainfall-NDVI trends as it had the lowest NDVIs in 1999 when rainfall was lowest, the highest NDVI in 1989 when rainfall was highest and moderate NDVIs in 2009 when rainfall was moderate. Vegetation at the IT was characterized by a dense thicket with diverse growth forms of canopy trees, woody shrubs, succulent shrubs and ephemerals which mostly had the C3 type of pathway. This was in contrast with the IT soil isotopy that showed more positive C isotope ratios, indicating a switch between C3 and CAM photosynthesis in original vegetation. Most of the canopy trees had disappeared in the DT to be replaced by herbs, shrubs and grasses. As such, there was a huge difference in isotope ratios between DT plants and soils with the plants having mostly C3 metabolism while the soil showed a predominance of CAM plants in previous vegetation, indicating significant changes in land cover. The G site mostly comprised the grasses Themeda triandra and Panicum maximum and a few herbs. It maintained a dominance of C4 metabolism in both plants and soils showing very little change in species composition over the years. Because of the higher species diversity at IT, its soil organic C was quite high reaching levels of 3.4 percent (i.e. 3.4 t C / ha) in the top 10 cm then decreasing with depth (p < 0.001); but was moderate at DT (1.1-1.3 percent) and very low at G ( 0.5 percent C) (p < 0.001). In the same manner above-ground biomass was highest at IT i.e. 330 000 kg/ha; but was only 22 000 kg/ha in DT and as low as 6 700 kg/ha in G vegetation. High biomass at IT was mostly attributed to the succulent shrub Portulacaria afra and the canopy trees Euclea undulate, Rhus longispina and Schotia afra. This above-ground biomass translated to biomass C amounts of 158 000 kg/ha at IT, 10 600 kg/ha at DT and 3 200 kg/ha at G. Thus the IT had the highest while G the least and DT moderate plant and soil C sequestration ability. In all, the conversion of IT to DT led to a net loss of 147 000 Kg of biomass C / ha and 12 000kg less organic C / ha of land. Soil carbon dioxide effluxes were however variable between seasons as they were affected by differences in soil properties and seasonal weather patterns. High soil moisture levels (up to 16 percent gravimetric moisture) resulted in reduced soil penetration resistance (1 to 4 Kg/cm2) which raised effluxes at G and DT sites (up to 1.2 μmols m-2 sec-1) in winter, while low moisture (2 percent) resulted in hard dry soil (14 Kgm-2 penetration resistance) with suppressed CO2 effluxes in spring (0.2 μmols m-2 sec-1) especially in DT and G soils. Rising temperature generally caused accelerated gas emissions but only when moisture was not limiting (as was the case in IT). Thus the high summer temperatures (up to 40oC) gave lower effluxes especially in DT and G (< 1 μm-2sec-1) due to limited moisture supply (< 10 percent); while the Autumn period that had very high temperature (up to 48 oC) and good moisture (up to 16 percent) saw accelerated soil CO2 emissions (averaging 2 μmols m-2 sec-1) from all cover types. The high biomass and litter fall at IT served as ready substrate for soil respiration as long as moisture was not limiting and temperatures were favourable, while reduced cover at DT resulted in poor moisture conservation and creation of hard dry soils in spring and summer with reduced respiration. It was concluded that the DT had high CO2 effluxes in winter and reduced emissions in summer; while the opposite was true for the IT. All the cover types had minimal CO2 effluxes in spring and accelerated emissions in autumn. The grassland on the other hand was a fairly moderate source or sink of CO2 in most seasons compared with the other two covers. It was observed that an environment of good moisture and low-moderate temperatures (such as that in the winter) minimises effluxes while maintaining good plant productivity. It was concluded that thicket vegetation is a good sink of carbon that should be preserved in its natural condition to optimize its carbon sequestration potential. All three land covers served as sources or sinks of CO2 depending on soil and seasonal conditions. Thus high moisture and low penetration resistance generally increased effluxes of thicket ecosystems. The effect of increasing temperature on effluxes was only significant when moisture was not limiting. Conditions of good moisture and low-moderate temperatures gave reasonable amounts of effluxes while maintaining good plant productivity. Though the dry soil conditions significantly reduced effluxes in all land covers; they were not desirable since they decreased plant productivity and ultimately its C sequestration potential. Moreover, prolonged dry conditions only serve to exacerbate recovery of thicket plants as they increase mortality of canopy species in degraded and transformed areas in comparison with intact thicket.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:nmmu/vital:10665 |
| Date | January 2013 |
| Creators | Zengeni, Rebecca |
| Publisher | Nelson Mandela Metropolitan University, Faculty of Science |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis, Doctoral, PhD |
| Format | xiii, 130 leaves, pdf |
| Rights | Nelson Mandela Metropolitan University |
Page generated in 0.0033 seconds