Thesis (MSc)--Stellenbosch University, 2005. / ENGLISH ABSTRACT: The agricultural sector in South Africa relies heavily on the use of pesticides to protect crops against pest organisms. Pesticides can affect non-target organisms such as the meso- and macrofauna in the soil detrimentally. Since these organisms play an important role in the processes of mineralization and decomposition in the soil and contribute to soil fertility, it is important that they are protected. A large amount of published literature exists on the biological importance of soil meso- and macrofauna and the effects that various agricultural practices have on them. The main aim of this study was to investigate the influence of agricultural practices on the abundance and diversity of meso- and macrofauna in different vineyard soils. A comparative study was conducted of an organically managed, conventionally managed and an uncultivated control soil. A secondary aim was to determine the effect of these agricultural management practices on the biological activity of these animals. Soil samples were taken, from which mesofauna (Collembola and Acari) were extracted with a modified Tullgren extractor, identified and counted. Earthworms were extracted from the soil using hand sorting methods. Soil parameters such as pH, water holding capacity, organic matter content, soil texture and soil respiration were determined. Bait lamina and litter-bags were also used to help determine the biological activity within the soil. The mesofauna diversity was quantified using the Shannon Weiner diversity index, as well as a diversity index described by Cancela da Fonseca and Sarkar (1996). Differences in abundance of both the meso-and macrofauna were statistically measured using ANOVA's. Biological activity results were also interpreted using ANOV A's. Results indicate that the abundance of the meso fauna was the highest at the organically treated vineyard soil and lowest in the conventionally managed soil where pesticide application took place. The earthworms also showed the same trend as the mesofauna, but were much more influenced by seasonal changes. Biological activity, according to the bait lamina and the litter-bag results, was higher in both the conventionally and organically managed soils than in the control, but no statistical significant differences were found between the two experimental soils. The soil respiration (C02-flux), also indicating biological activity, was highest in the organically treated soil and lowest in the conventionally treated soil. The different sampling techniques used gave variable results and although the organically managed soil proved to have higher abundances of both meso- and macrofauna, the biological activity did not show the same trends. In conclusion the data did not give enough evidence as to whether organic management practices were more beneficial than conventional management practices for the maintenance of soil biodiversity. / AFRIKAANSE OPSOMMING: Die Suid Afrikaanse Landbousektor steun hewig op die gebruik van verskillende chemiese pestisiede om oeste teen pes organismes te beskerm. Pestisiede kon ook verskeie ander nie-teikenorganismes soos die meso- en makrofauna in die grond negatief affekteer. Hierdie organismes behoort beskerm te word omdat hulle 'n belangrike rol speel in grondprosesse soos mineralisering, en die afbreek van organiese materiaal. Hierdie organismes dra ook by tot die vrugbaarheid van die grond. Daar is heelwat gepubliseerde literatuur beskikbaar wat verband hou met die biologiese belangrikheid van grond meso- en makrofauna en die effekte wat verskeie landbou behandelings op hulle het. Die primêre doel van hierdie studie was om vas te stel watter invloed konvensionele landboupraktyke op die hoeveelheid en diversiteit van meso- en makrofauna in verskillende wingerdgronde het. 'n Vergelykende studie is gedoen om wingerdgronde wat konvensioneel en organies behandel is sowel as 'n onbehandelde kontrolegrond met natuurlike plantegroei met mekaar te vergelyk. 'n Sekondêre doel van hierdie studie was ook om die effek van die verskillende boerderymetodes op die biologiese akitiwiteit in die grond te ondersoek. Grondmonsters is geneem, waaruit die meso fauna (Collembola en Acari) deur middel van 'n aangepaste Tullgren ekstraktor ge-ekstraheer, geïdentifiseer en getel. Die erdwurms is deur middel van handsorteringsmetodes versamel. Die volgende grond parameters is gemeet: pH, waterhouvermoë, organiese materiaal inhoud, grondtekstuur en grondrespirasie. "Bait lamina" en "litter bags" is ook gebruik om biologiese aktiwiteit in die grond te bepaal. Die diversiteit van mesofauna is bepaal met die Shannon Weiner diversiteitsindeks, as ook 'n diversiteitsindeks wat deur Cancela da Fonseca en Sarkar (1996) ontwikkel is. Die resultate van beide die meso- en makrofauna hoeveelhede in die verskillende wingerdgronde is met mekaar vergelyk deur van ANOV A's gebruik te maak. Die resultate van die biologiese aktiwiteit is ook deur middel van ANOVO's statisties met mekaar vergelyk. Die resultate het aangetoon dat die hoeveelheid mesofauna die hoogste in die organies behandelde grond en die laagste in konvensionele grond was. Die erdwurms het dieselfde patroon as die mesofauna getoon, maar is baie meer deur seisoenale faktore geaffekteer, bv. reënval. Volgens die resultate van die "bait lamina" en die "litter bags" was die biologiese aktiwiteit in die grond hoër in beide die eksperimentele grond as in die kontrolegrond. Die grondrespirasie (C02-puIs) was hoër in die kontrolegrond as in die ander eksperimentele gronde. Daar was groot variasie tussen die resultate wat met die verskillende tegnieke verkry is en alhoewel die organiese perseel hoër hoeveelhede van beide meso- en makrofauna gehad het, het die biologiese aktiwiteit nie dieselfde tendens gewys nie. Vanuit die data wat verkry is kon daar dus nie met sekerheid afgelei word dat organiese boerderymetodes beter vir die biodiversiteit van gronde,soos hier gemeet, is as konvensionele boerderymetodes nie.

Thesis (MScAgric (Soil Science))--University of Stellenbosch, 2007. / This study forms part of an ARC Nietvoorbij research project on the long-term effect of different groundcover management practices in vineyards. This study concentrated on the root distribution of the vines after different groundcover practices were used for one decade. The vineyard on the research farm of the Agricultural Research Council at Robertson, South Africa, was divided into 56 equally distributed trail blocks, 14 treatments with four replicas each. For the study Chardonnay cultivar on 99 Richter rootstock was used. This study concentrated on only eight treatments. The eight different cover crop practices consisted of the mechanical treatment, chemical treatment, straw mulch treatment, permanent cover crop treatment and four annual cover crop treatments. The annual cover crop treatments consisted of triticale cover crop that was sprayed with an herbicide before bloom, a triticale cover crop that was left to die naturally, a grazing vetch cover crop that was sprayed with an herbicide before bloom and a grazing vetch cover crop that was left to die naturally. The vineyard was established in November 1992 while the different treatments started in April 1993. Root studies were conducted during the 2002/2003 season. In this study the auger method was used to determine root density (mm/cm³). The root density was correlated with data from the penetrometer studies, yield and shoot mass collected during the same season.

Thesis (PhD(Agric) (Agronomy))--University of Stellenbosch, 2007 / A selection of species suitable for cover crop management in the different wine grape regions is required to enable more producers to apply this environment friendly practice in a sustainable manner as part of an integrated production strategy. The correct management practice(s) to be applied to these species over both the short and long term in a cooler and warmer wine grape region needed clarification. The effect of seeding date on the dry matter production (DMP) and weed control efficacy of seven grasses and sixteen legumes, as well as varieties of three of these species, was determined during 1991 and 1992. The decomposition rate of the mulches was determined. In the cooler climate of Stellenbosch (33o55'S, 18o52'E), the Medicago species, subterranean clovers, pink Seradella and three Vicia species did not compete effectively with the winter weeds if the weekly precipitation from mid-March to mid-May (autumn) exceeded 18 mm. The two oat species, as well as rye and triticale produced more than five t/ha of dry matter if the precipitation exceeded 18 mm per week. The DMP of the above-mentioned species indicated that these species could be considered for cover crop management in Lutzville (31o35'S, 18o52'E), if full surface irrigation of 18 mm per week could be applied for 10 weeks directly after sowing, followed by 18 mm fortnightly. Seeding date had a significant effect on DMP in both regions. A highly significant correlation (r = 0.85, p ≤ 0.0001) existed between the decomposition rate of the mulches and the initial amount of dry matter present on the soil surface.

Largement utilisés par l’agriculture intensive, les pesticides sont aujourd’hui présents dans les différents compartiments environnementaux et ce à un niveau mondial. Les recherches de ces travaux de thèse se sont plus particulièrement portées sur deux grandes thématiques, l’une liée à l’écodynamique des pesticides en Garonne estuarienne par la caractérisation des sources, la présence et le devenir des molécules, et l’autre sur leur transfert dans les compartiments eau, air et sols d’un château viticole. Des méthodes d’analyse simples et rapides à mettre en oeuvre ont été optimisées, tout en étant sensibles afin d’être compatibles avec l’étude des milieux faiblement contaminés.Les pesticides les plus fortement quantifiés dans les stations d’épuration (STEP) bordelaises sont le diuron, l’imidaclopride et le fipronil (≈ 50-100 ng.L-1). N’étant pas éliminés par les différents traitements, ils sont présents dans les rejets (apports de l’ordre du g.j-1) qui peuvent alors être une source de contamination pour le milieu. Si les pesticides majoritairement présents dans la Garonne sont le métolachlore et ses métabolites (≈ 10-800 ng.L-1), qui présentent des pics printaniers en lien avec l’application de la molécule, les pesticides caractéristiques de STEP sont également quantifiés. Les concentrations sont plus faibles (0,5-20 ng.L-1) et moins variables mais ne peuvent cependant pas être négligées, certaines molécules étant fortement toxiques même à faible concentration, comme le fipronil (PNEC = 0,77 ng.L-1).Les sols du château viticoles sont caractérisés par des pics printaniers (50-1000 ng.g-1) en pesticides en lien avec les applications, comme le boscalid ou l’azoxystrobine. Des molécules non utilisées depuis plusieurs années, comme le diuron, ont également été mis en évidence à plus faibles niveaux de concentrations (ng.g-1) et moins variables ce qui témoigne de la persistance de certaines molécules. Le compartiment atmosphérique est également caractérisé par des pics saisonniers en lien avec l’application et la volatilité des molécules, comme le folpel ou le chlorpyrifos-methyl. Des pesticides non utilisés par le château mais potentiellement appliqués par des exploitations voisines comme le métolachlore ont également été détectés, ce qui démontre bien du transport des molécules. Des pesticides interdits (diuron), non utilisés (métolachlore) et utilisés (boscalid) par le château ont été mis en évidence dans la nappe phréatique. / Intensive use of pesticides led to a widespread contamination of water, air and soils compartments. The present study focused on two main subjects, one based the ecodynamic of pesticides in Garonne River by characterization of sources, presence and outcome of these molecules and one other based on their transfer to air, water and soil compartments of a vineyard. Easy and fast methods of extractions were optimized although they were sensitive in order to be applied for mildly contaminated samples.Most quantified pesticides in Bordeaux wastewater treatment plants (WWTP) were diuron, imidaclopride and fipronil (≈ 50-100 ng.L-1). As treatments were not efficient to remove pesticides, effluents are a significant source for the Garonne River (daily discharge around 5 g). Even if major pesticides in Garonne River were metolachlor and its metabolites (≈ 10-800 ng.L-1) which showed spring peaks, typical WWTP pesticides were also quantified. Concentrations were lower (0,5 - 20 ng.L-1) and variable but they cannot be neglected as they can be highly toxic. Fipronil was for instance detected at concentrations sometimes exceeding its PNEC (0.77 ng.L-1).Vineyard soils were characterized by spring peaks of pesticides (50-1000 ng.g-1) that can be related to applications, like boscalid or azoxystrobine. Pesticides forbidden for a long time have also been detected, like diuron but in lower concentrations levels (ng.g-1) and less variable, that highlighted persistence of such molecule. Atmospheric compartment was also characterized by spring peaks than can be related to applications and molecule volatility, with the presence of folpet and chlorpyrifos-methyl. Pesticides non used by the vineyard were also detected such as metolachlor, which highleted the transport of such molecules. Forbidden (diuron), non-used (metolachlor) and currently used (boscalid) pesticides were detected in the aquifer.

Vineyards in New Zealand suffer bird damage caused by several avian species, including blackbirds and silvereyes. The introduced European Blackbird takes whole grapes which reduces yield. The self-introduced Australasian Silvereye pecks on grapes, leaving them on the vine to be further attacked by fungi and bacteria, and the subsequent off-odours can cause grapes to be refused by the winery or to suffer a price-reduction. Bird control methods remain primitive and largely ineffective during the long ripening period of wine grapes. An ecologically sound method to manage and reduce bird pressure requires deeper understanding of why some birds eat grapes, especially since grapes are not particularly nutritious. This work investigated the extent to which blackbirds and silvereyes are attracted by various compounds in ripening grapes. Since in natural grapes these compounds develop and change simultaneously, I developed an artificial grape in which a single parameter could be investigated. Artificial grapes (and sometimes nectar) were presented on a bird feeder table and the responses of birds to hexose sugars, the aromas 2-3-isobutylmethoxypyrazine and geraniol, tartaric and malic acids, grape tannins, and purple and green colour were recorded on timelapse video and analysed.

<p>In this thesis we explore and extend the theory of persistent homology, which captures topological features of a function by pairing its critical values. The result is represented by a collection of points in the extended plane called persistence diagram.</p><p>We start with the question of ridding the function of topological noise as suggested by its persistence diagram. We give an algorithm for hierarchically finding such epsilon-simplifications on 2-manifolds as well as answer the question of when it is impossible to simplify a function in higher dimensions.</p><p>We continue by examining time-varying functions. The original algorithm computes the persistence pairing from an ordering of the simplices in a triangulation and takes worst-case time cubic in the number of simplices. We describe how to maintain the pairing in linear time per transposition of consecutive simplices. A side effect of the update algorithm is an elementary proof of the stability of persistence diagrams. We introduce a parametrized family of persistence diagrams called persistence vineyards and illustrate the concept with a vineyard describing a folding of a small peptide. We also base a simple algorithm to compute the rank invariant of a collection of functions on the update procedure.</p><p>Guided by the desire to reconstruct stratified spaces from noisy samples, we use the vineyard of the distance function restricted to a 1-parameter family of neighborhoods of a point to assess the local homology of a sampled stratified space at that point. We prove the correctness of this assessment under the assumption of a sufficiently dense sample. We also give an algorithm that constructs the vineyard and makes the local assessment in time at most cubic in the size of the Delaunay triangulation of the point sample.</p><p>Finally, to refine the measurement of local homology the thesis extends the notion of persistent homology to sequences of kernels, images, and cokernels of maps induced by inclusions in a filtration of pairs of spaces. Specifically, we note that persistence in this context is well defined, we prove that the persistence diagrams are stable, and we explain how to compute them. Additionally, we use image persistence to cope with functions on noisy domains.</p> / Dissertation

No abstract available. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2004.

Thesis (MScAgric)--University of Stellenbosch, 2006. / Bibliography. Also available via the Internet.

Thesis (MScAgric)--University of Stellenbosch, 2002. / ENGLISH ABSTRACT: In an attempt to reduce the pH of juice and wine, different fertiliser applications and canopy management practices were evaluated in a field trial. Fertiliser treatments consisted of no, CaS04, Ca(OH)2, and MgS04 fertilisation. Canopy management was as follows: suckering (leaving only two shoots per bearer), tipping, vertical shoot positioning and removal of lateral shoots and yellow leaves in the bunch zone (Canopy 1); suckering (leaving three shoots per bearer), vertical shoot positioning as well as topping (Canopy 2); vertical shoot positioning and topping (Canopy 3). The field trial was conducted in the Paardeberg region on the farms Meerlus and Kersfontein. The vineyard at Meerlus was Cabernet franc/R99 with a high canopy density and a good root distribution, established on a sandy loam soil of granite origin, with a low subsoil pH and a high K content. The vineyard at Kersfontein was Cabemet Sauvignon/101-14 Mgt with a lower canopy density and a less extensive root distribution, also established on a sandy loam soil of granite origin, but with a low top- and subsoil pH and a higher K content. Fertilisation had no significant influence on the K content of juice and wine. Fertilisation with Ca and Mg reduced the pH of juice significantly in the case of Meerlus. In contrast, Mg fertilisation increased the pH of juice significantly at Kersfontein. Lastly, fertilisation had no significant effect on the pH of the wine. The K content of the juice at Meerlus was significantly reduced by Canopy 3 in comparison with Canopy 1 and 2. However, in contrast with Canopy 1 and 3, Canopy 2 significantly increased the pH of juice at Meerlus. The K content of the juice at Kersfontein was significantly reduced by Canopy 1, compared to Canopy 2 and 3, with no significant effect on the pH of the juice. Canopy management had no significant effect on wine pH. It appears to be possible to reduce the pH of juice in the case of Cabernet franc/99R, situated on granite soils, by means of Ca and Mg fertilisation. Because Ca(OH)2 and MgS04 fertilisation increased the maturity of Cabernet Sauvignon/101-14 Mgt grapes, it was impossible to evaluate the effect of fertilisation on the pH of juice at Kersfontein. The general phenomenon that a high canopy density results in a high pH of juice and wine, was not observed in this field trial. The possible reasons for this were the low canopy density of the control plots, as well as the difference in maturity between canopy management treatments. / AFRIKAANSE OPSOMMING: In 'n poging om die pH van sap en wyn te verlaag, is die effek van verskillende bemesting- en lowerbestuursfaktore in 'n veldproef geêvalueer. Bemesting het bestaan uit geen, CaS04, Ca(OH)2 en MgS04 toediennings. Lowerbestuur was: suier tot twee lote per draer, tip, vertikale lootposisionering, verwydering van sylote en geel blare in trossone (Lower 1); suier tot drie lote per draer, top en vertikale lootposisionering (Lower 2); top en vertikale lootposisionering met geen suier nie (Lower 3). Die veldproef is op twee plase nl. Meerlus en Kersfontein, in die Paardeberg omgewing uitgevoer. Die wingerd by Meerlus was Cabernet franc/R99 met 'n hoë lowerdigtheid en goeie wortelverspreiding, wat op 'n sandleemgrond van graniet oorsprong met In lae ondergrond-pH en hoë K-inhoud gevestig is. Die wingerd by Kersfontein het bestaan uit Cabernet Sauvignon/101-14 Mgt met 'n laer lowerdigtheid en swakker wortelverspreiding, wat op In sandleemgrond van graniet oorsprong met In deurgaans lae grond-pH en baie hoë K-inhoud gevestig is. Bemesting het geen betekenisvolle invloed op die K-inhoud van sap en wyn gehad nie. By Meerlus het Ca- en Mg-bemesting egter die pH van sap betekenisvol verlaag. In teenstelling hiermee het Mg-bemesting die sap-ph by Kersfontein betekenisvol verhoog. Bemesting het verder geen betekenisvolle invloed op die pH van wyn gehad nie. Lower 3 het die K-inhoud van sap by Meerlus betekenisvol verlaag in vergelyking met Lower 1 en Lower 2. By Kersfontein was die K-inhoud van sap by Lower 1 betekenisvollaer as by Lower 2 en Lower 3. Teenoor Lower 1 en Lower 3 het Lower 2 'n betekenisvol hoër sap-pH by Meerlus tot gevolg gehad. Lowerbehandelings het egter geen betekenisvolle invloed op die pH van wyn gehad nie. Die moontlikheid bestaan dus om die sap-pH van Cabernet franc/R99 op granietgrond betekenisvol m.b.v. Ca- en Mg-bemesting te verlaag. Aangesien Ca(OH)2- en MgS04- bemesting die rypheidsgraad van Cabernet Sauvignon/101-14 Mgt se druiwe betekenisvol verhoog het, was dit onmoontlik om die effek van bemesting op sap-pH by Kersfontein te evalueer. Die algemene verskynsel dat 'n hoë lowerdigtheid tot hoë pH's in sap en wyn lei, is nie in die proef ondervind nie. Die lae lowerdigtheid van die kontrole persele en die verskil in rypheidsgraad tussen lowerbehandelings kan moontlik as rede hiervoor aangevoer word.

À l’échelle d’un terroir viticole, le climat présente des variations significatives et joue un rôle important sur les caractéristiques des vins produits. L’adaptation de la filière viticole au changement climatique en cours nécessite la connaissance de l’évolution du climat à l’échelle locale. Cette étude vise à intégrer cette échelle dans les projections climatiques en se basant sur l’utilisation combinée de modèles dynamiques et géostatistiques. Dans un premier temps, l’utilisation d’un modèle climatique régional à haute résolution (1 km) dans les vignobles de Marlborough (Nouvelle-Zélande) a permis de cartographier les températures d’une région viticole. Les limites et les incertitudes de l’utilisation de ce type de modèle, notamment pour la représentation des variations thermiques les plus locales, ont également été étudiées. Par l’utilisation des données issues d’un réseau dense de capteurs de température, une seconde étape a consisté au développement d’un modèle statistique non linéaire permettant une cartographie fine des températures sur les appellations Saint-Émilion, Pomerol et leurs satellites. Enfin une méthode d’intégration de l’échelle locale dans les projections de changement climatique est proposée, associant modèles dynamiques et modèles géostatistiques. Cette thèse a mis en évidence que l’utilisation simultanée de différentes méthodes de modélisation des températures peut représenter une piste intéressante pour pallier aux manques qu’elles peuvent représenter individuellement et limiter ainsi l’incertitude. / At vineyard scale, climate variability can be significant in magnitude and play a key role in vine and wine characteristics. Adaptation of viticulture to climate change requires knowledge about future fine-scale climate evolution. This study aims to integrate local scale in future climate projections, coupling dynamic and statistical modelling. A first step consisted in producing temperature maps at 1 km resolution using WRF in a vineyard area (Marlborough, New-Zealand) and evaluating model uncertainties. It revealed that dynamical models do not represent well local climate variations. Using a high density temperature data logger network, the second part is dedicated to developing a non-linear statistical model to map temperature at very fine scale in famous sub-appellations of the Bordeaux vineyard area (Saint-Émilion). Following, a method, coupling dynamical and statistical modelling, is proposed to integrate local scale in climate change projections. This thesis highlights that using simultaneously statistical and dynamical models can be an approach to reduce model uncertainties.