Hydrology

Permanent URI for this collection

https://hdl.handle.net/10530/367

Browse

Browsing Hydrology by Author "Kelbe, B."

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Characterization of the hydrostratigraphic units of the Sodwana area using the electrical resistivity method
    (University of Zululand, 2015) Nweze, Augustine; Simonis, J.J.; Kelbe, B.
    This study was aimed at defining the geological facies of the Sodwana area. It formed part of a bigger project designed to investigate the linkages between the hydrological and ecological drivers of the Mgobolezeni Catchment, Sodwana. Monitoring boreholes were drilled by the Department of Water Affairs (DWA) at several locations in the study area for the purpose of monitoring groundwater changes in the Catchment. There was a need to interpolate the geology and groundwater conditions between the monitoring boreholes. This research provided the geological conditions between the monitoring boreholes using the electrical resistivity geophysical method. The Geotron (G41) resistivity meter was used to conduct the survey. Vertical electrical sounding (VES) were conducted initially in close proximity to DWS monitoring boreholes for calibration purposes. The VES data were plotted and interpreted using IPI2win computer software. Hydrostratigraphic models were developed for all VES locations in close proximity to monitoring boreholes. These models were used for interpolation of the geology between the boreholes. A total of 20 VES were conducting in the study area. 16 VES results showed 5-layer models while 4 VES results showed 4-layer models. The top layer showed very high resistivity values (> 2000Ωm) and was interpreted as dry sand (Kwambonambi Formation). The second layer showed a resistivity range of 200Ωm – 2000Ωm. In some locations, the second layer showed a resistivity range of 120Ωm - 200Ωm and was interpreted as the shallow aquifer in the Kwambonambi Formation. The third layer showed a resistivity range of 50Ωm - 120Ωm and was interpreted as the low transmissivity Kosi Bay/Isipingo Formations. The fourth layer had a resistivity range of 10Ωm - 50Ωm and was interpreted as the deeper aquifer unit of the Uloa Formation. The basement Cretaceous rock formed the fifth layer that showed a clear increase and higher resistivity values than that of the overlying fourth layer. Transmissivity values observed from the pumping tests conducted on the calibration boreholes were combined with VES results to define the transmissivity of the inferred deeper aquifer in between boreholes. The thicknesses of the geoelectric layers obtained from VES results showed a good relationship with corresponding geologic units in borehole logs. The varying apparent resistivity values and lithological information enabled the demarcation of the geoelectric units into hydrostratigraphic units. VES results also showed that the surface of the Cretaceous unit is deeper in the eastern side of the study area than in the western side, showing that the Cretaceous unit dips towards the east. The modelled transmissivity also showed a good statistical relationship (R2 = 0.99) v with the observed transmissivity. Therefore the VES method used for this study successfully achieved the aim of this study.
  • Thumbnail Image
    Item
    A geohydrological assessment of the behaviour and response of the Zululand coastal plain to both environmental influences and human activity
    (1991) Rawlins, Brian Kenneth; Kelbe, B.
    The Zululand coastal plain on the eastern seaboard of South Africa contains large quantities of readily available fresh water. Lake St Lucia, located within the plain, is a wetland habitat internationally recognised for- its environmental importance. Yet over the past few decades extensive forest developments have taken place within the lake catchment. The considerably higher transpiration rate of the fast growing trees planted there over those of grassland environments they replaced is well established. The effects on the hydrological balance of the lake and its environs of the greater water loss resulting from this land use modification has however been the subject of much speculation. This study analyses the hydrological regime of the eastern shores plain over the 17 years from 1973 to 1990. During this period both wet and dry climatic conditions were experienced, and in 1981 a plantation covering 1100 ha was established. In order to clarify the magnitude and variability of hydrological parameters under natural and under altered conditions, comparisons were made between wet and dry periods, and between grasslands and forested areas. The extent to which plantations have modified the hydrological balance could thus be determined and placed in regional perspective. The study began with reviews of the physiography of the eastern shores catchment area and of the existing data base (chapter 2). Following a substantial upgrading of the hydrometric network (chapter 3), specific data collection took place. The components of the hydrological cycle were then assessed in order to identify differences in hydrological response both to climatic variability and to land use. The results of these assessments are presented as individual sections on meteorology (chapter 4), surface hydrology (chapter 5), and geohydrology (chapter 6). It is the conclusion of the study that in this sub-tropical environment, with shallow depths to groundwater, commercial forests are responsible for an additional consumptive use of water of between 150 and 175 mm/year. In the regional perspective of the water balance of Lake St Lucia, the total forested area of 25 000 ha will thus be seen to lower average inflow to the lake by between 10% and 12%. During extreme dry periods this figure increases to approximately 30%. Although they are greatest during dry climatic cycles, the effects of such a reduction in water yield are at all times significant. In the first place, the water balance of the lake is seriously affected. In the second, with.a reduction in its fresh water input, the salinity balance of the lake cannot but be influenced. In the third, changes in this salinity balance will have a direct impact on the flora and fauna of the lake and its environs. It is to be hoped that following the proposed future removal of the plantations from the eastern shores, the situation will improve. However, the remaining 20 000 ha under commercial forestry on the western shores will continue to have a significant effect upon the hydrology of Lake St Lucia.
  • Thumbnail Image
    Item
    The Impact of Flooding Characteristics on Cotton Cultivation in Lower Kano Plain in Nyando District, Western Kenya
    (2010) Ocholla, Peter Omondi; Kelbe, B.; Rawlins, B.
    Flooding continues to be a common environmental hazard in both developed and developing countries. Kenya has not been spared by the destruction that is usually associated with floods. Crops, settlement and infrastructure are usually impaired wherever flooding occur. The severity of damage as a result of floods has been documented to have had a relationship with the flood magnitude, flood frequency and occupation of the flood prone floodplains of large rivers. In the Lower Kano Plains of Western Kenya, damage to crops by floods is exacerbated by occupation of the lower reaches of the Nyando River. This study sets out to assess the impact of flooding characteristics of the Nyando River on cotton cultivation in the Lower Kano Plains. In particular, the study examined the characteristics of the Nyando River Basin with the aim of describing how the river morphometry could have influenced flooding in the Lower Kano Plains. Also investigated, is the change in the flood magnitude and frequency with time and space, and finally, what anecdotal data (perception of cotton farmers) are available to support the assessment of flooding on cotton cultivation. The study deployed both quantitative and qualitative research methods in examining the variability of rainfall and flow and the consequent impacts of flooding on cotton cultivation. Households living downstream in Lower Kano Plains were the target social unit of analysis. Multi-stage sampling technique was used to sample the respondents whom were interviewed through the use of a self administered questionnaire schedule. Descriptive and inferential statistical techniques were used in data analysis. Relevant probability models were used to analyze the flood magnitude and flood frequency. Generally, the findings related to the research question have shown that flooding in the Lower Kano Plains has inhibited cotton cultivation and lower crop acreage. Furthermore, output has significantly declined for the period when spate has either denied farmers the ability for early planting or destroyed cotton already in the fields. The inter-seasonal and intra-seasonal variability of rainfall and flow show their first peaks are dominant in April and May, while October and November present the second cycle. There is however, a shift in the cycles towards August and September, making the annual flow cycle highly variable in terms of decisions for crop growing. These peaks interfere with the cotton growing calendar, and results in delays in planting or destruction of cotton already planted. The results from the spectral analysis revealed a strong annual and biannual cycle of both rainfall and flow, and an oscillating 4 months cycle that exhibited climate instability. A strong seasonal signal was evidenced between 6 and 12 month period that correspond with the flooding peaks. Similarly, wavelet results demonstrated a strong 12 month spectrum of both rainfall and floods with a large frequency of oscillation in the later period of the 31-year time series. The high power band of 2 to 5 years for both the raw and filtered rainfall and flow time series revealed a Quasi Biennial Oscillation (QBO) and a shift in the rainfall and flow cycle. The findings of the correlation also revealed that (r2 = 0.278) rainfall in catchment explains 27.8% of high flow. The rest (72.2%) is attributed to other factors such as anthropogenic or hydro geologic characteristics of the study area. The study area was revealed to be prone to between 3 and 7 years flood return frequency with an average magnitude of 400 m3/sec. Further results showed that out of the 31 years of continuous time series flow the Nyando River recorded 18 years of bankful flow (200 to 387.6 m3/sec). The high frequency of bankful flow illustrates that Nyando River has limited channel capacity and is therefore vulnerable to flooding downstream. The 1 year return frequency characterizes the Lower Kano Plains to crop damage by annual spates, and thus, demands a shift in the cropping pattern and or change of crop variety to those ones that withstand poor drainage. Other factors that have exacerbated the decline in cotton production include poor price of lint, competition from synthetic fibres and rising cost of cotton production. Because of poor remuneration from cotton production, farmers are shifting to the growing of other crops such as rice and sugarcane that are less affected by flooding conditions. The two latter crops are said to be highly viable, cost effective and reliable. The problem of cotton cultivation in the study area is therefore due to variability of hydrologic conditions and economic factors. There is a need for a sound flood mitigation policy as well as the adoption of appropriate agronomic practices that would enhance cotton cultivation and improve output in the flood prone areas for it to be profitable in the present economic climate.
  • Thumbnail Image
    Item
    An information base for a decision support system for management of the Mhlathuze River
    (2000) Snyman, Nina-Marie; Kelbe, B.
    Effective integrated catchment management is dependant on suitable information describing the physical, social and economic conditions in the catchment. It also depends on knowledge of the processes controlling the environmental systems in the catchment. This thesis describes the development of an information base (IB) for a computer-based decision support system (DSS) to support the effective management of the Mhlathuze River system. The IB has been developed to contain most of the available data, which are analysed and processed by models, for utilization by catchment managers. The DSS and accompanying IB was developed during the formulation and implementation of the new Water Act (1998) in South Africa. It attempts to support the implementation of this Water Act, which stipulates the integrated management of water resources on a catchment basis. The database was designed to contain all hydrologically relevant data and information on the Mhlathuze River catchment. Extensive data collection has identified information on rainfall, evaporation, flow measurements, the soil type map of the area, 1:50 000 topographical maps and 1:500 000 maps, 1991 census data, lithology and groundwater, national, regional and local boundaries, pollution and water quality monitoring points, water consumption and details of water users, etc. The database is still growing with the identification and collection of more data and the creation of additional information about the Mhlathuze River catchment from various models. A structure was developed in the IB to give access to the megabytes of information on the database in a structured manner. The operating environment of Arcview 3 (running on Windows 95) was used to develop a user - database interaction system. The Windows concept of interactive icons was used to customise the user interface by incorporating buttons and tools to the IB. Scripts, written in Avenue, were attached to these buttons and tools, to add to the functionality of the IB. A Digital Elevation Model (DEM) was developed from 100-metre elevation contours which originated from 1:500 000 maps. These were supplemented by digitized contours from the 1:50000 maps. The DEM has a horizontal resolution of 125 metre by 125 metre per cell, and covers the whole extent of the Mhlathuze River catchment, stretching 120 km in the east-west direction, and 60 km in the north-south direction. A land use model for the catchment was developed from satellite imagery (7 bands from the Landsat TM satellite). Two different techniques, involving supervised and unsupervised classification methods were applied to identify the land cover classes. The supervised classification method used the maximum likelihood technique, while the unsupervised classification method applied a cluster analysis technique of classification. For the hydrological run-off modelling of the Mhlathuze River system, the HYdrological Modelling System (HYMAS), utilising the Variable Time Interval (VTT) model, was chosen to simulate short duration hydrological events in a distributed manner. The model was used in an Instream Flow Requirements (IFR) study to identify the ecological reserve of the river. Information derived from the DEM and land use model were used during the hydrological simulations. Output from the simulations was compared to the few observed flow measurements which are available for the catchment.