No description available.

This research evaluated the application of artificial neural networks and hydrodynamic models for river flood modelling and prediction. The study has been completed in three main parts: First part focused on the application of artificial neural networks for flood prediction in ungauged catchments. Catchment descriptors were used as input data and the index flood was the output of the model. Different types and numbers of catchment descriptors (17 descriptors and more than 1000 catchments) were used to choose those that gave the best relationship with the hydrological behaviour and flood magnitude. ANN models with different architectures were developed and applied to training and validation sets of data to find the best type of ANN for this application. Selection of pooling groups of catchments either randomly or according to geographical proximity did not produce desirable results. Therefore hydrologically similar catchments were clustered using the FEH-Software before entering descriptors into the ANN model. This improved the accuracy of predicted floods. The second part of the research aimed to model river flow in a multi-gauging station catchment and provide real-time prediction of peak flow downstream. Three types of ANN (Multi-Layer Perception (MLP), Recurrent and Time Lag Recurrent) were adapted to evaluate the applicability of this technique. The study area covers the Upper Derwent River, a tributary of the River Trent in the UK. River flow was predicted at the subject site with lead times of 3, 6, 9 and 12 hours. Tests were completed using different lengths of input data to evaluate the effect of input data size in model outputs. The number of gauging sites to be used as data sources in the model as also evaluated. In the final part, the application of artificial neural networks (ANN) to optimise the results obtained from a hydrodynamic model of river flow as evaluated. The study area is Reynolds Creek Experimental Watershed in southwest Idaho, USA.

Potassium is one of the three most important elements in fertiliser, but in comparison to the mobile anions nitrate and phosphate, there has been little study of its transport within drainage basins. Potassium is a monovalent cation that sorbs strongly to soil constituents; leaching losses are expected to be minimal. However, data from a variety of catchments indicate that concentrations of dissolved potassium increase up to ten-fold during storm events, suggesting that leaching does occur. Possible sources of potassium are vegetation, the soil, mobilised sediment on the hillslope and sediment within the river channel. Previous work has only considered vegetation as a source. Field data at three scales: the hillslope, first-order subcatchment and fourth-order catchment, are used to investigate the sources of potassium and the mechanisms involved during transport from the hillslope to the drainage basin outlet. The drainage basin studied is that of the River Gara, near Slapton in south west Devon, UK. A temporally and spatially intensive sampling programme of mobile and immobile soil waters at the hillslope scale was conducted during storm events in the winter of 1995-6. At this scale, old water that has been resident in the soil for some time is of higher potassium concentration than new water (precipitation inputs for that storm). Discharge of such water to the surface as return flow is particularly significant in potassium transport. The subcatchment response during storm events is consistent with these observations: maximum potassium concentrations occur at the same time as old water contributes to the storm hydrograph. Vegetation and mobilised sediment on the hillslope are demonstrated to increase the potassium concentration of overland flow, but the relative significance of these two mechanisms could not be assessed. Fertiliser applications of potassium are efficiently retained within the soil; no changes in the subcatchment response are apparent after a fertiliser application. At the catchment scale, the maximum potassium concentrations during storm events occurred at the same time as new water contributed to the storm hydrograph. This response is not consistent with the hillslope and subcatchment response. It is interpreted to indicate that the link between potassium transport from the hillslope to the basin outlet is complex, with storages and sinks of potassium reducing the significance of hillslope-scale processes in the catchment response. Suspended sediment within the river channel is demonstrated to contribute to the increase in potassium concentrations during storm events, although they do not account for all of the increase. An alternative explanation is that the land use of the subcatchment differs from that of the whole catchment. The percentage of land put to arable crops in the subcatchment was higher than in the whole catchment; this is likely to reduce the potassium concentration of overland flow (composed of new water), since organic material is more evenly distributed throughout the soil profile in this case. An autumn increase in potassium concentrations has previously been observed; this has been attributed to autumn leaf fall. Experimental data indicate that potassium lost from fallen leaves is efficiently retained within the soil. Weekly data from two gauging stations of Slapton Ley Field Centre (1987-1996) display a highly variable potassium response, and demonstrate no seasonal trend.

Master of Arts / Department of Geography / Kendra K. McLauchlan / The importance of local parent material has been recognized as a fundamental control on the geochemistry of lake sediments, but there have been relatively few broad-scale surveys of catchment sources of terrigenous lake sediments. In this paper, I present a geochemical study of catchment parent materials and lake sediments from four lakes in Northern Minnesota. Similar climate and vegetation conditions are present at all four lakes, which vary mainly in catchment parent material and lake morphometry. Geochemical data including major, trace and rare earth elements (REEs) from catchment parent material samples was compared with lake sediment geochemical data using PCA, linear regression, geological indices and elemental ratios. In homogenous till-dominated catchments, patterns of elemental variation in the catchment till could be extended to predict elemental concentrations in the lake sediments. Simple ratios, which are commonly used to analyze lake sediment geochemical data, were not good predictors of lake sediment composition, however. Catchments with mixed bedrock and till were compositionally heterogeneous, and comparison with lake sediments was difficult. Lack of grain size control and biogenic silica measurements further confounded analysis. However, ΣREE/Y ratio was found to be diagnostic of the catchment parent materials and present within the lake sediments. This study makes a contribution to an improved understanding of lacustrine sedimentary archives by analyzing the spatial linkages among catchment, water and sedimentary geochemistry.

Thesis (M.Sc.)--University of the Witwatersrand, Faculty of Science, 2011 / A multifactor ecological integrity index, focusing on freshwater ecosystems on a quaternary catchment scale, can be of great benefit to conservation planning. No ecological integrity index has previously been developed for South African quaternary catchments. In this study an index was developed based on three environmental surrogates: land cover, river integrity and fish species conservation status, with the intention of identifying quaternary catchments of highest conservation concern. By developing such an index, the aim was to provide a general indication of the degree to which catchments have been transformed from a natural environment to a human altered environment, thereby identifying catchments most in need of conservation. For the three available datasets, indices were developed using a five category point-scoring system. A score of one indicates a completely degraded environment and a score of five indicates a pristine environment. The original land cover data consisted of 49 different land cover types which were reduced to five land cover transformation scores. Available river integrity data already existed in five categories and a numerical score of one to five was applied to each category. Fish species conservation status was scored according to the IUCN red data list classifications on a similar basis. Subsequently, a weighted mean score expressed as a percentage was calculated for the three indices for each quaternary catchment. These indices indicate the degree of change/transformation from a natural system (100%) to a largely degraded system (20%). Ultimately, an ecological integrity index was calculated as a mean value of the three related but independent indices. However, the results of the developed ecological integrity index were not representative of real world conditions. This is largely attributed to the lack of complete data found in two out of the three datasets used in the study. Some of the main limitations encountered were the lack of river segment definitions within each catchment and the incomplete and un-systematic collected fish species data records. The land cover data, on the contrary, was of high definition and high standard. It is recommended that in the interim, the developed land transformation index, based on a detailed analysis of land cover, be used as an indicator index of ecological integrity of catchments

MSc., Faculty of Science, University of Witwatersrand, 2011 / A multifactor ecological integrity index, focusing on freshwater ecosystems on a quaternary catchment scale, can be of great benefit to conservation planning. No ecological integrity index has previously been developed for South African quaternary catchments. In this study an index was developed based on three environmental surrogates: land cover, river integrity and fish species conservation status, with the intention of identifying quaternary catchments of highest conservation concern. By developing such an index, the aim was to provide a general indication of the degree to which catchments have been transformed from a natural environment to a human altered environment, thereby identifying catchments most in need of conservation. For the three available datasets, indices were developed using a five category point-scoring system. A score of one indicates a completely degraded environment and a score of five indicates a pristine environment. The original land cover data consisted of 49 different land cover types which were reduced to five land cover transformation scores. Available river integrity data already existed in five categories and a numerical score of one to five was applied to each category. Fish species conservation status was scored according to the IUCN red data list classifications on a similar basis. Subsequently, a weighted mean score expressed as a percentage was calculated for the three indices for each quaternary catchment. These indices indicate the degree of change/transformation from a natural system (100%) to a largely degraded system (20%). Ultimately, an ecological integrity index was calculated as a mean value of the three related but independent indices. However, the results of the developed ecological integrity index were not representative of real world conditions. This is largely attributed to the lack of complete data found in two out of the three datasets used in the study. Some of the main limitations encountered were the lack of river segment definitions within each catchment and the incomplete and un-systematic collected fish species data records. The land cover data, on the contrary, was of high definition and high standard. It is recommended that in the interim, the developed land transformation index, based on a detailed analysis of land cover, be used as an indicator index of ecological integrity of catchments

<p class=MsoNormal><span style="mso-spacerun: yes">?????????????????? </span>Stable isotopic analysis of nitrate (¹⁵N/¹⁴N and ¹⁸O/¹⁶O) was used to trace nitrate sources and cycling under undisturbed conditions and following harvest at the Turkey Lakes Watershed (TLW), located near Sault Ste. Marie, Ontario, Canada. <span style="mso-spacerun: yes">?? </span> <p class=MsoNormal><span style="mso-spacerun: yes">?????? </span><span style="mso-spacerun: yes">????????????</span>Bulk precipitation collected biweekly at the TLW from 1995 to 2000 had nitrate isotope values that ranged from +42. 4 to +80. 4&permil; for <span style='font-family:Symbol'>d</span>¹⁸O and -6. 3 to +2. 8&permil; for <span style='font-family:Symbol'>d</span>¹⁵N. <span style="mso-spacerun: yes">?? </span>An incubation experiment indicated that the isotopic composition of atmospheric nitrate was not compromised by collection methods whereby unfiltered bulk precipitation samples remain in the collector for up to two weeks. <span style="mso-spacerun: yes">?? </span> <p class=MsoNormal><span style="mso-spacerun: yes">?????????????????? </span>The first direct measurement of the isotopic composition of microbial nitrate produced <i>in situ</i> was obtained by eliminating precipitation inputs to three forest floor lysimeters and subsequently watering the area with a nitrate-free solution. <span style="mso-spacerun: yes">?? </span>Microbial nitrate had <span style='font-family:Symbol'>d</span>¹⁸O values that ranged from +3. 1 to +10. 1&permil; with a mean value of +5. 2&permil;, only slightly higher than values predicted based on the <span style='font-family:Symbol'>d</span>¹⁸O-H₂O of the watering solution used. <span style="mso-spacerun: yes">?? </span><span style='font-family:Symbol'>d</span>¹⁸O values of soil O₂ (+23. 2 to +24. 1&permil;) down to a depth of 55cm were not significantly different from atmospheric O₂ (+23. 5&permil;) and therefore respiratory enrichment of soil O₂ did not affect the <span style='font-family:Symbol'>d</span>¹⁸O values of microbial nitrate produced at the TLW. <span style="mso-spacerun: yes">?? </span> <p class=MsoNormal><span style="mso-spacerun: yes">?????????????????? </span>Nitrate export from two undisturbed first-order stream basins was dominated by microbial nitrate, with the contribution of atmospheric nitrate peaking at about 30% during snowmelt. <span style="mso-spacerun: yes">?? </span>Clear-cutting of catchment 31 in 1997 resulted in elevated nitrate concentrations, reaching levels that exceeded the drinking water limit of 10 mg N/L. <span style="mso-spacerun: yes">?? </span>Isotopic analysis indicated that the source of this nitrate was predominantly chemolithoautotrophic nitrification. <span style="mso-spacerun: yes">?? </span>The <span style='font-family:Symbol'>d</span>¹⁸O values of microbial nitrate in stream 31 progressively increased during the post-harvest period due to an increase in the proportion of nitrification that occurred in the summer months. <span style="mso-spacerun: yes">?? </span>Despite drastic alteration of nitrogen cycling in the catchment by the harvest, <span style='font-family:Symbol'>d</span>¹⁵N-nitrate values in shallow groundwater did not change from the pre-harvest. <span style="mso-spacerun: yes">???? </span>Denitrification and plant uptake of nitrate in a small forested swamp in catchment 31 attenuated 65 to 100% of surface water nitrate inputs following harvest, reducing catchment-scale nitrate export by 35 to 80%.

>Magister Scientiae - MSc / Improved knowledge is required on the quantity and source of water resources, particularly evident during periods of drought currently being faced in South Africa. There is inadequate knowledge with regards to the flood attenuating properties of wetlands, particularly evident in the ungauged catchments of Southern Africa. This study aims to improve the knowledge on the contribution of flow from tributaries with headwaters in mountainous regions to low lying areas and the effects of wetlands on river flow patterns. Several river flow monitoring sites were established along the major upper tributaries of the Nuwejaars River at which daily water levels were recorded and bi-weekly discharge measurements were conducted. Weather data was collected using four automatic weather stations and three automatic rain gauges’ setup throughout the catchment. Rainfall data coupled with rating curves and daily discharges were used to assess the flow responses of these tributaries to rainfall events. Additionally, stable isotope analysis and basic water quality analysis was used to determine the major sources of flow within the major tributaries. The rainfall and river flow data collected, coupled with the characterization of the wetland was used to determine the flood attenuation capabilities of the wetland. Lastly, a conceptual model based on a basic water balance was developed to further explain the role of the wetland and its effects on river flows. The results showed a 27-hour lag time in peak flows from the upper tributaries at the inflows of the wetland to the outflow. Two of the upper tributaries had flow throughout the year and were fed by springs in the upper mountainous regions of the catchment and all tributaries were largely reliant on rainfall for peak flows. The temporary storage of flows within the wetland occurred as a result of the Nuwejaars River bursting its banks, filling of pools, or ponds and the Voëlvlei Lake. It was concluded that the wetland increased the travel time and decreased the magnitude of flows of the Nuwejaars River. However, due to the fact that wetlands are interlinked on a catchment scale and have a collective effect on flood attenuation this study may be improved by looking at the wetlands within the catchment holistically.

<p class=MsoNormal><span style="mso-spacerun: yes">          </span>Stable isotopic analysis of nitrate (¹⁵N/¹⁴N and ¹⁸O/¹⁶O) was used to trace nitrate sources and cycling under undisturbed conditions and following harvest at the Turkey Lakes Watershed (TLW), located near Sault Ste. Marie, Ontario, Canada. <span style="mso-spacerun: yes">  </span> <p class=MsoNormal><span style="mso-spacerun: yes">    </span><span style="mso-spacerun: yes">      </span>Bulk precipitation collected biweekly at the TLW from 1995 to 2000 had nitrate isotope values that ranged from +42. 4 to +80. 4&permil; for <span style='font-family:Symbol'>d</span>¹⁸O and -6. 3 to +2. 8&permil; for <span style='font-family:Symbol'>d</span>¹⁵N. <span style="mso-spacerun: yes">  </span>An incubation experiment indicated that the isotopic composition of atmospheric nitrate was not compromised by collection methods whereby unfiltered bulk precipitation samples remain in the collector for up to two weeks. <span style="mso-spacerun: yes">  </span> <p class=MsoNormal><span style="mso-spacerun: yes">          </span>The first direct measurement of the isotopic composition of microbial nitrate produced <i>in situ</i> was obtained by eliminating precipitation inputs to three forest floor lysimeters and subsequently watering the area with a nitrate-free solution. <span style="mso-spacerun: yes">  </span>Microbial nitrate had <span style='font-family:Symbol'>d</span>¹⁸O values that ranged from +3. 1 to +10. 1&permil; with a mean value of +5. 2&permil;, only slightly higher than values predicted based on the <span style='font-family:Symbol'>d</span>¹⁸O-H₂O of the watering solution used. <span style="mso-spacerun: yes">  </span><span style='font-family:Symbol'>d</span>¹⁸O values of soil O₂ (+23. 2 to +24. 1&permil;) down to a depth of 55cm were not significantly different from atmospheric O₂ (+23. 5&permil;) and therefore respiratory enrichment of soil O₂ did not affect the <span style='font-family:Symbol'>d</span>¹⁸O values of microbial nitrate produced at the TLW. <span style="mso-spacerun: yes">  </span> <p class=MsoNormal><span style="mso-spacerun: yes">          </span>Nitrate export from two undisturbed first-order stream basins was dominated by microbial nitrate, with the contribution of atmospheric nitrate peaking at about 30% during snowmelt. <span style="mso-spacerun: yes">  </span>Clear-cutting of catchment 31 in 1997 resulted in elevated nitrate concentrations, reaching levels that exceeded the drinking water limit of 10 mg N/L. <span style="mso-spacerun: yes">  </span>Isotopic analysis indicated that the source of this nitrate was predominantly chemolithoautotrophic nitrification. <span style="mso-spacerun: yes">  </span>The <span style='font-family:Symbol'>d</span>¹⁸O values of microbial nitrate in stream 31 progressively increased during the post-harvest period due to an increase in the proportion of nitrification that occurred in the summer months. <span style="mso-spacerun: yes">  </span>Despite drastic alteration of nitrogen cycling in the catchment by the harvest, <span style='font-family:Symbol'>d</span>¹⁵N-nitrate values in shallow groundwater did not change from the pre-harvest. <span style="mso-spacerun: yes">   </span>Denitrification and plant uptake of nitrate in a small forested swamp in catchment 31 attenuated 65 to 100% of surface water nitrate inputs following harvest, reducing catchment-scale nitrate export by 35 to 80%.

The aquifer vulnerability of the Molototsi (B81G) and Middle Letaba (B82D) quaternary catchments was assessed to determine the influence of the vadose zone on the groundwater regime. Anecdotal evidence indicated that the aquifers may be vulnerable to pollution. The aquifer vulnerability was assessed by developing a new method RDSS. The RDSS method was developed by combining relevant vulnerability parameters of DRASTIC, GOD, EPIK, SEEPAGE, COP and SINTACS. RDSS evaluates the vadose zone as a pathway for pollutants by using the following four parameters namely: Recharge, Depth to water table, Soil type and Slope. Recharge was estimated using the Chloride-mass balance method. Depth to water table was measured in the field using a dipmeter. For inaccessible boreholes, data was requested from Groundwater Project Consulting Company. The seepage behaviour (soil type) was determined using parameters such as hydraulic conductivity, infiltration and percolation. Percolation and hydraulic conductivity was determined by undertaking percolation tests in accordance with SABS 0252-2:1993. Infiltration was determined using the double ring infiltrometer. Slopes were determined from the digital elevation method using ArcGIS software. High recharge was revealed in the lower parts of both B81G and B82D. Shallow depth to water level was revealed on the upper part of B82D and extended towards the lower part of B81G. Soil type relates to saturated vertical hydraulic conductivity, which was rated to be high in the northeast of B81G. Gentle (high influence due to preferential infiltration to runoff) slopes extend from the south towards the northern parts of both B81G and B82D. The four parameters (recharge, depth to water table, soil type, and slope) were overlaid using Weighted Sum, Weighted Overlay and Raster Calculator to produce the final vulnerability map. When using Weighted Overlay and Weighted Sum, rasters were given different percentages of influence in different scenarios. The Weighted Overlay tool inputs multiple rasters and sets all weights equal to 100%. The Weighted Sum tool inputs multiple rasters and sets all weight equal to 1.0. When using the Raster Calculator, rasters were evaluated by being added together without multiplying by the percentage of influence. The results obtained are discussed in detail with reference to the degree of vulnerability of these two densely populated rural areas. / Dissertation (MSc)--University of Pretoria, 2013. / Geology / unrestricted