In this study, geophysical measurements (Electrical Resistivity Tomography-ERT) and remote sensing
techniques were applied in the Thukela river basin at various scales to complement the classical
hydrometeorological networks. Detailed process hydrological studies were carried out at the Potshini
catchment in the Thukela river basin to provide an in-depth understanding of the influence of different
land use management practices, notably the impact of conservation tiJlage practices, on runoff
generation and soil moisture retention characteristics at field scale. The general trend that was
observed in the field studies is that conservation tillage systems influenced the partitioning of rainfall,
by significantly reducing surface runoff over agricultural lands under conservation tillage practices,
with a reduction ranging from 46 to 67%. The field soil-water balance studies also indicated that more
soil moisture was retained in plots under conservation tillage practices compared to plots under
conventional tillage and hence the wider adoption of such a practice could influence the partitioning of
rainfall across scales.
The field based study was integrated into catchment process studies where a classical hydrometrical
network was complemented with geophysical measurements (ERT) along catchment transects to
determine the interaction of the surface and sub-surface water and the relative contribution of the
subsurface water to catchment response. The study revealed that the shallow ground water contributes
significantly, close to 75%, of the stream flows in the Potshini catchment, especially during the dry
seasons, with the response of the shallow ground water being a function of both the rainfall intensity
and daily total amount.
The potential of integrating the catchment process studies with the larger river basin scale was
explored through the evaporative term of the water balance by applying the Surface Energy Balance
Algorithm for Land (SEBAL), a remote sensing methodology, to estimate total evaporation (ET) from
the Moderate Imaging Spectroradiometer (MODIS) satellite images. This was validated with ground
measurements from a Large Aperture Scintilometer (LAS) installed in the Potshini catchment. Good
comparison was established between the remotely sensed estimates and LAS measurements with a
deviation range of between -14 to 26% on discrete days, where the deviation was defined as the
departure of the remotely sensed estimates of ET from the respective LAS measurements. The results
from this study compare well with results from similar studies in other countries with different
climatic conditions. Subsequently, the evaporative water use of various land uses in the upper
Thukela river basin was assessed using MODIS images. Commercial forestry was identified to be the land use with a consistent and relatively high evaporative water use In the study area. High
evaporation rates over water bodies were observed during the wet summer season when both the
natural and man made water bodies were at full capacity. Nevertheless, it is recognized that the
inherent low resolution ofthe MODIS images could have impacted on the SEBAL results.
Finally, a conceptual framework, drawing the strengths of classical hydrometeorological networks,
geophysical measurements, isotope tracers and remote sensing is suggested with the potential of
enhancing our understanding and conceptualization of hydrological determinants across scales. The
relevance of the framework to water resources management is highlighted through its application to
the Potshini catchment and the Thukela river basin using results and findings from this study. / Thesis (Ph.D.)-University of KwaZulu-Natal, Durban, 2008.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:ukzn/oai:http://researchspace.ukzn.ac.za:10413/578 |
| Date | January 2008 |
| Creators | Kongo, Victor M. |
| Contributors | Jewitt, Graham Paul Wyndham., Lorentz, Simon A. |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0025 seconds