Thesis (MScAgric (Soil Science))--University of Stellenbosch, 2011. / Includes bibliography. / ENGLISH ABSTRACT: The topic of landscape heterogeneity has captured the imagination of ecologists and agriculturists alike and has been extensively investigated in this dissertation. Heuweltjies are landscape features putatively created by the termite Microhodotermes viator through their burrowing and nest-building activities. They have been closely examined in the natural veld of the Western Cape in the recent past and are the focus of many ecological studies. However, the effect of heuweltjies in cultivated landscapes (e.g. vineyards, orchards and wheat lands) remains unexplored. This study contributes significantly to our understanding of soil modifications associated with heuweltjies, as well as the physiology of vines growing on and off heuweltjies and wine made from these vines. It was hypothesized that heuweltjies occurring in cultivated areas can significantly affect crop yield and quality, thereby establishing itself as a potentially important role player in the agricultural economy of the Western Cape. This study was conducted in two climatic regions of the Western Cape, Stellenbosch (Mediterranean climate, Cabernet Sauvignon) and Robertson (semi-arid climate, Shiraz) to better understand how differences in heuweltjie characteristics correspond to differences in rainfall and temperature. In both study areas, heuweltjie soils were compared to non-heuweltjie soils with respect to physical and chemical. Grapevines associated with these heuweltjies were also compared to those growing on the adjacent, non-heuweltjie soils to determine any variation in vine vigour, physiology, phenology, berry characteristics and wine quality. Through the use of ANOVA's and Fisher's LSD posthoc tests to indicate statistical significance in soil and grapevine characteristics, it was apparent that heuweltjies induce substantial changes in soil and vine properties. Significant differences in the water content exist between the soils of the heuweltjies and non-heuweltjie areas. Heuweltjie soils exhibited higher values in comparison to the non-heuweltjie soils in the Stellenbosch study area, with opposite results in Robertson. Heuweltjie soils also displayed higher exchangeable calcium and magnesium and higher total carbon and total nitrogen values than non-heuweltjie soils in both study areas. Differences in physiology were more subtle, but vine vigour was severely altered on the heuweltjie-associated vines, exhibiting excessive vegetative growth in Stellenbosch, leading to variations in berry characteristics on and off the heuweltjies. Again, the opposite was observed for Robertson. Lower sugar and alcohol percentages and higher titratable as well as malic acid concentrations were observed in the wines emanating from the heuweltjies in Stellenbosch. Sensory analyses proved significant, as lower astringency and alcohol burn were detected in the Cabernet Sauvignon heuweltjie wines than the non-heuweltjie wines in the Stellenbosch study area. Chemical differences in the wines from the Robertson study are were insignificant. However, a significantly lower fruitiness was observed in the Shiraz heuweltjie-wines when compared to the non-heuweltjie wines. Differences in soil water content between heuweltjies and its adjacent soils was the most influential factor in this study, and affected all of the soil-grapevine interactions to a large extent. Further research need to be conducted to better understand and clarify the reasons behind these variations, as well as possible effects of global warming on heuweltjie functioning in different climatic regions of the Western Cape. / AFRIKAANSE OPSOMMING: Die onderwerp van landskap heterogeniteit het die verbeelding van beide ekoloë en landboukundiges aangegryp en is op `n omvangryke wyse ondersoek in hierdie tesis. Heuweltjies is landskap eienskappe geskep deur die uitgrawe en nesbou aktiwiteite van die termiet Microhodotermes viator. Heuweltjies in die natuurlike veld is intensief bestudeer in die onlangse verlede en is die fokus van `n verskeidenheid ekologiese studies. In bewerkte landskappe (bv. wingerde, vrugteboorde en koringlande) is daar egter `n groot tekort aan navorsing oor die invloed van heuweltjies. Hierdie studie dra grotendeels by tot ons begrip van heuweltjie-geassosieërde grondverandering, asook die fisiologie van wingerd wat groei op en af van heuweltjies, sowel as die wyn afkomstig van hierdie wingerdstokke. Hipoteties sal heuweltjies in bewerkte areas die gewasopbrengs asook –kwaliteit betekenisvol beïnvloed en word so dus gevestig as `n potensieël belangrike rolspeler in die landbou-ekonomie van die Wes-Kaap. Die studie was onderneem in twee klimaatstreke van die Wes-Kaap, Stellenbosch (Mediterreënse klimaat, Cabernet Sauvignon) en Robertson (semi-ariede klimaat, Shiraz) om vas te stel hoe verskille in heuweltjie eienskappe ooreenstem met verksille in reënval en temperatuur. In beide studie areas is heuweltjie gronde met nie-heuweltjie gronde vergelyk met betrekking tot fisiese en chemiese eienskappe. Wingerdstokke geassosieër met heuweltjies is ook vergelyk met dié wat heuweltjies omring om enige variasie in groeikrag, fisiologie, fenologie, druifkorrel eienskappe en wynkwaliteit te bepaal. Deur gebruik te maak van ANOVA's en Fisher se LSD posthoc toetse om statistiese betekenisvolheid in grond- en wingerdeienskappe aan te dui, was dit duidelik dat heuweltjies wesenlike veranderinge in dié eienskappe teweegbring. Betekenisvolle verskille heers in die waterinhoud tussen die heuweltjie en nie-heuweltjie gronde. Heuweltjie gronde het hoër waardes getoon in vergelyking met die nie-heuweltjie gronde in die Stellenbosch studie area, met teenoorgestelde bevindings in Robertson. Heuweltjie gronde het ook `n hoër konsentrasie uitruilbare kalsium en magenesium, en totale koolstof en stikstof in vergelyking met nie-heuweltjie gronde. Verskille in fisiologie was baie meer subtiel, maar wingerd groeikrag was beduidend verskillend op die heuweltjie-geasossieërde wingerdstokke, met oormatige vegetatiewe groei in Stellenbosch wat lei tot `n verskil in druifkorrel eienskappe op en af van die heuweltjie. Weereens is die teenoorgestelde waargeneem in Robertson. Laer suiker en alkohol persentasies, asook hoër titreerbare - en appelsuur konsentrasies was te bespeur in die wyne afkomstig van die heuweltjies in Stellenbosch. Sensoriese analise het betekenisvolle verskille bewys, met `n laer vrankheid en alkohol-brand sensasie waargeneem vir die Cabernet Sauvignon heuweltjie wyne as die nie-heuweltjie wyne in die Stellenbosch studie area. Chemiese verskille in die wyne vanaf Robertson was onbeduidend. Sensoriese analise het egter anders bewys, met `n betekenisvolle laer vrugtigheid te bespeur in die nie-heuweltjie wyne as die heuweltjie wyne in die Robertson studie area. Verskille in die grondwaterinhoud tussen heuweltjies en omringende gronde was die mees invloedryke faktor in hierdie studie en beïnvloed tot `n groot mate al die grond-wingerdstok interaksies. Verdere navorsing is nodig om helderheid te verskaf agter die redes vir hierdie variasies, sowel as moontlike gevolge van aardverwarming op funksionering van heuweltjies in verskillende klimaatstreke van die Wes-Kaap.

The effects of crop load and berry exposure on the composition of Marlborough Sauvignon blanc grapes and wine from the Brancott vineyard, Blenheim, were explored. Commercially grown, 2-cane and 4-cane Sauvignon blanc vines were used with a row orientation of north-south. Two exposure treatments were imposed in the following manner: complete leaf removal was undertaken in the fruit zone and 50% shade cloth was erected to give a uniform shading treatment to half the trial vines. Weekly thirty-berry and whole bunch samples were taken from each of the 32 plots with the exception of the veraison period when two samples per week were taken. Vine vigour was assessed using pruning and leaf area per vine data. Harvest occurred on different dates for 2-cane and 4-cane pruned vines so that fruit attained from both treatments had similar °Brix. Fruit was processed at the Lincoln University winery. Must analysis and wine analysis were undertaken. As expected, 4-cane vines had almost double the yield of 2-cane vines. Higher crop load significantly reduced leaf area per shoot and shoot thickness. Lower leaf area to fruit ratio for 4-cane berries resulted in delayed onset of veraison and slowed the rate of sugar accumulation. Crop load, which limited leaf area to fruit ratio, appeared to be the dominant factor in determining timing of grape physiological ripeness as expressed by °Brix over other factors such as fruit exposure. Malic acid, tartaric acid, IPMP (iso-propylmethoxypyrazine) and IBMP (iso-butyl-methoxypyrazine) were lower at equivalent °Brix in 4-cane compared with 2-cane berries. Significantly higher concentrations of quercetin were found in exposed compared to shaded berries. Must analysis showed a significant influence of crop load on berry titratable acidity and pH, reflecting berry ripening results. Exposure significantly increased the concentrations of nitrogenous compounds in 4-cane must yet showed no influence on 2-cane must. After wine processing lower malic acid concentrations in wines made from 100% exposed fruit became evident in lower wine titratable acidity but showed no influence on wine pH. Bentonite addition to wines had a small but statistically significant influence on wine by reducing pH, titratable acidity and alcohol. Bound sulphur concentrations were significantly higher in 4-cane versus 2-cane wines. At harvest, methoxypyrazine levels in grapes and wines were very low; IBMP concentrations where significantly lower than those normally found in Sauvignon blanc wines from Marlborough. This was attributed to the absence of basal leaves from the shoots of ripening berries. The results suggest that leaf area to fruit ratio is a powerful determinant of grape and wine quality.

The influence of soil texture on canopy growth, vine yield, and fruit composition of Sauvignon blanc were assessed on a mature vineyard from the Rapaura area in Marlborough. The subject vines were mature Sauvignon blanc in a commercial vineyard trained to four cane VSP and planted with a north south orientation. Row and vine spacings were typical of the area at 3m x 1.8m. Four areas of different soil texture were identified using trunk circumference measurements, visual assessment of the surface soil and aerial photographs to identify changes in vine growth. Soil pits were excavated at a later date to determine root numbers and record the soil texture in the different areas. Two crop treatments were imposed on half of the plots at approximately 50% veraison: unthinned crop and 50% crop. A shading treatment was also imposed at 50% veraison where three sets of tagged shoots had bunches that were exposed to sunlight, bunches that had some natural shading from leaves or bunches that had paper bags fastened over each to provide a completely shaded environment. Vine vegetative vigour for each plot was assessed during the growing season using the Point Quadrat method and at pruning using bud counts and pruning weights. Thirty berry samples were collected from the 32 plots and analysed weekly for soluble solids and berry weight. Harvest of bunches from the tagged shoots occurred on two different dates with the first harvest picked when fruit from one soil and crop treatment had reached 21 °Brix. The second picking took place just before the commercial harvest, which coincided with last soil and crop treatment reaching 21 °Brix. Bunches from each tagged shoot were weighed, frozen and later analysed for soluble solids, pH, organic acids, and methoxypyrazines. Vines on very stony textured soils had small trunk circumferences with a high root density and yield to pruning weight ratio compared to those growing on deep silt soils. Vines on deep silt soils had more vigorous canopies with large shoots and a higher leaf layer number. The target soluble solids was reached first by the vines on the very stony plots despite the higher crop load with soil texture appearing to be a dominant factor by influencing the time of flowering. Cluster shading decreased soluble solids, consistent with other studies, whilst crop thinning resulted in an earlier harvest. This was contrary to popular opinion that crop thinning at veraison would have no impact on sugar accumulation. pH and organic acids were unaffected by shading or bunch thinning. Vines growing on deep silt soils had a significantly higher level of total acidity and malic acid than those on the stony soils. At harvest, methoxypyrazine levels were very low compared to previously reported figures for Marlborough, which may have been a result of sample preparation. IBMP was significantly higher on deep silt soils, however, with no impact from the shading or crop treatment. The findings are consistent with the hypothesis that IBMP synthesis may be increased by the number of leaves surrounding bunches (Roujou de Boubee, 2003). The results suggest that soil texture plays a dominant role in influencing Sauvignon blanc flavour and aroma and due consideration should be given to vineyard layout and site selection prior to new plantings taking place.

Vineyards in the Mid-Atlantic often have large, vigorous vines that can be costly to manage and produce inadequate fruit for wine production. Dense canopies increase the incidence of fungal disease, require greater allocation of resources to manage, and inhibit fruit development. The primary objective of these studies was to determine effective vine-size modification treatments that would optimize fruit quality, while reducing labor and chemical control. Research factors included root manipulation, under-trellis ground cover, and rootstock. Treatment levels were root bag (RBG) or no root manipulation (NRM); under-trellis cover crop (CC) or herbicide strip (HERB); and one of three rootstocks: 101-14, Riparia Gloire, or 420-A. Effects of these treatments were measured in two experiments: Experiment I compared combinations of all three treatments, while Experiment II explored the individual effects of root restriction using root bags of varying volumes. Root restriction consistently demonstrated the ability to reduce vegetative growth and vine water status. In the first experiment fruit-zone photosynthetic photon flux density (PPFD) was increased by 234% in RBG vines. Timed canopy management tasks indicated that RBG canopies required about half the labor time of NRM canopies. Anthocyanin concentration and total phenolic content were increased by 20% and 19% respectively in RBG fruit. CC increased fruit-zone PPFD by 62%, and increased soluble solids and color compounds. The 420-A rootstock reduced potassium uptake, resulting in lower must potassium concentration. Results demonstrated that these treatments significantly reduce vegetative growth in a humid climate, decrease management labor, and produce higher quality fruit. / Master of Science / Vineyards in the Mid-Atlantic often have large, vigorous vines that can be costly to manage and produce inadequate fruit for wine production. Dense canopies increase the incidence of fungal disease, require greater allocation of resources to manage, and inhibit fruit development. The primary objective of these studies was to determine effective vine-size modification treatments that would optimize fruit quality, while reducing labor and chemical control. Research factors included root manipulation, under-trellis ground cover, and rootstock. Treatment levels were root bag (RBG) or no root manipulation (NRM); under-trellis cover crop (CC) or herbicide strip (HERB); and one of three rootstocks: 101-14, Riparia Gloire, or 420-A. Effects of these treatments were measured in two experiments: Experiment I compared combinations of all three treatments, while Experiment II explored the individual effects of root restriction using root bags of varying volumes. Root restriction consistently demonstrated the ability to reduce vegetative growth and vine water status. Sunlight exposure to grape berries was significantly increased in RBG vines. Timed canopy management tasks indicated that RBG canopies required about half the labor time of NRM canopies. Sugar and color concentration were both increased in RBG fruit. CC also increased sunlight exposure, as well as sugar and color concentration. The 420-A rootstock reduced potassium uptake, resulting in lower must potassium concentration. Results demonstrated that these treatments significantly reduce vegetative growth in a humid climate, decrease management labor, and produce higher quality fruit.

[ES] La industria vitivinícola contemporánea enfrenta varios desafíos clave: optimizar la producción, minimizar la huella ambiental, reducir costos y mejorar la calidad del vino. La integración de la tecnología y herramientas de teledetección ha demostrado ser una solución eficiente para estos problemas. La agricultura de precisión, que incluye el manejo del riego, evaluación y control del follaje, y la identificación de los mejores emplazamientos para cada variedad de uva, es especialmente beneficiosa para esta industria. Un aspecto crucial en la producción de vino es el seguimiento de la madurez de las bayas de uva, mediante la determinación del contenido de azúcar y acidez, lo cual permite establecer la fecha óptima de vendimia. Tras la producción del vino, se evalúan su color y contenido de polifenoles, indicadores de calidad. Las imágenes satelitales son fundamentales en la teledetección para la agricultura de precisión, permitiendo monitorear la salud de los viñedos y controlar las condiciones meteorológicas. Este estudio utilizó imágenes satelitales para relacionar parámetros de calidad del vino (pre y poscosecha) con índices y bandas espectrales derivadas de las imágenes Sentinel-2 obtenidas antes de la cosecha. La investigación abarcó las temporadas de vendimia de 2017 y 2018, estudiando 135.8 hectáreas de viñedos de uvas Cabernet Sauvignon en la comuna de Marchigüe, Valle de Colchagua, Chile Central. El objetivo principal fue desarrollar modelos utilizando índices y variables espectrales de imágenes Sentinel-2 para aproximar parámetros de calidad de uva y vino, como °Brix y pH, así como índice de color, antocianinas totales y fenoles totales del vino. Se espera que estos modelos, aplicados en toda la superficie de trabajo, permitan una gestión diferenciada del cultivo, mejorando la calidad del vino al predecir indicadores pre y poscosecha que influyen en las características sensoriales y la diferenciación de marcas. En las variables precosecha, las bandas de onda corta y el índice de vegetación son más frecuentes, mientras que en las variables poscosecha destacan las bandas de borde rojo, infrarrojo cercano y el índice de humedad. Estos resultados sugieren la necesidad de realizar más estudios o explorar otras especies. Las imágenes satelitales se valoran como herramientas valiosas para mejorar el proceso artesanal y obtener vinos de mejor calidad, destacando la metodología de trabajo adaptable a diversas condiciones ambientales. Uno de los hallazgos más interesantes es la determinación del periodo fisiológico óptimo para el análisis predictivo, específicamente la quincena previa a la vendimia, permitiendo una planificación precisa y oportuna del proceso de cosecha y proporcionando datos esenciales para los enólogos. En cuanto a los coeficientes de determinación obtenidos para °Brix, índice de color y antocianinas totales en la primera temporada, los resultados son altamente positivos, con valores entre 65% y 77%, lo que sugiere una predicción fiable y contribuye a la definición de la cosecha selectiva. Otras variables muestran relaciones menos concluyentes, con coeficientes de determinación entre 42% y 57% en las campañas 2017 y 2018, indicando la necesidad de mejorar el estudio mediante un mayor número de muestras o refinando el proceso de muestreo. Los resultados positivos y la disponibilidad de imágenes satelitales ofrecen la oportunidad de identificar áreas para estudios detallados y cosechas diferenciadas. La selección precisa de uvas con características similares es crucial para mejorar la calidad y productividad del vino. Este estudio resalta la viabilidad de la teledetección y la vendimia selectiva como herramientas clave para gestionar la calidad y estilo del vino, proporcionando nuevas perspectivas en la viticultura. / [CA] La indústria vitivinícola contemporània enfronta diversos reptes clau: optimitzar la producció, minimitzar la petjada ambiental, reduir costos i millorar la qualitat del vi. La integració de la tecnologia i les eines de teledetecció ha demostrat ser una solució eficient per a aquests problemes. L'agricultura de precisió, que inclou el maneig del reg, avaluació i control del fullatge, i la identificació dels millors emplaçaments per a cada varietat de raïm, és especialment beneficiosa per a aquesta indústria. Un aspecte crucial en la producció de vi és el seguiment de la maduresa de les baies de raïm, mitjançant la determinació del contingut de sucre i acidesa, cosa que permet establir la data òptima de verema. Després de la producció del vi, se n'avaluen el color i el contingut de polifenols, indicadors de qualitat. Les imatges satel·litàries són fonamentals en la teledetecció per a l'agricultura de precisió, permetent monitoritzar la salut de les vinyes i controlar les condicions meteorològiques. Aquest estudi va utilitzar imatges satelitals per relacionar paràmetres de qualitat del vi (pre i postcollita) amb índexs i bandes espectrals derivades de les imatges Sentinel-2 obtingudes abans de la collita. La investigació va abastar les temporades de verema de 2017 i 2018, estudiant 135.8 hectàrees de vinyes de raïm Cabernet Sauvignon a la comuna de Marchigüe, Vall de Colchagua, Xile Central. L'objectiu principal va ser desenvolupar models utilitzant índexs i variables espectrals d'imatges Sentinel-2 per aproximar paràmetres de qualitat de raïm i vi, com °Brix i pH, així com índex de color, antocianines totals i fenols totals del vi. S'espera que aquests models, aplicats a tota la superfície de treball, permetin una gestió diferenciada del cultiu, millorant la qualitat del vi en predir indicadors pre i postcollita que influeixen en les característiques sensorials i la diferenciació de marques. A les variables precollita, les bandes d'ona curta i l'índex de vegetació són més freqüents, mentre que a les variables postcollita destaquen les bandes de vora vermella, infraroig proper i l'índex d'humitat. Aquests resultats suggereixen la necessitat de fer més estudis o explorar altres espècies. Les imatges satel·litàries es valoren com a eines valuoses per millorar el procés artesanal i obtenir vins de millor qualitat, destacant la metodologia de treball adaptable a diverses condicions ambientals. Una de les troballes més interessants és la determinació del període fisiològic òptim per a l'anàlisi predictiva, específicament la quinzena prèvia a la verema, permetent una planificació precisa i oportuna del procés de collita i proporcionant dades essencials per als enòlegs. Pel que fa als coeficients de determinació obtinguts per a °Brix, índex de color i antocianines totals a la primera temporada, els resultats són altament positius, amb valors entre 65% i 77%, cosa que suggereix una predicció fiable i contribueix a la definició de la collita selectiva. Altres variables mostren relacions menys concloents, amb coeficients de determinació entre 42% i 57% a les campanyes 2017 i 2018, indicant la necessitat de millorar l'estudi mitjançant un nombre més gran de mostres o refinant el procés de mostreig. Els resultats positius i la disponibilitat d'imatges satel·litals ofereixen l'oportunitat d'identificar àrees per a estudis detallats i collites diferenciades. La selecció precisa de raïms amb característiques similars és crucial per millorar la qualitat i la productivitat del vi. Aquest estudi ressalta la viabilitat de la teledetecció i la verema selectiva com a eines clau per gestionar la qualitat i estil del vi, proporcionant noves perspectives a la viticultura. / [EN] The contemporary wine industry faces several key challenges: optimizing production, minimizing the environmental footprint, reducing costs and improving wine quality. The integration of remote sensing technology and tools has proven to be an efficient solution to these problems. Precision agriculture, which includes irrigation management, canopy assessment and control, and the identification of the best sites for each grape variety, is especially beneficial for this industry. A crucial aspect of wine production is the monitoring of grape berry maturity by determining the sugar and acidity content, which allows the optimal harvest date to be established. After the wine is produced, its color and polyphenol content, quality indicators, are evaluated. Satellite imagery is fundamental in remote sensing for precision agriculture, allowing the health of vineyards to be monitored and weather conditions to be controlled. This study used satellite imagery to relate wine quality parameters (pre- and post-harvest) to indices and spectral bands derived from Sentinel-2 images obtained prior to harvest. The research covered the 2017 and 2018 harvest seasons, studying 135.8 hectares of Cabernet Sauvignon grape vineyards in the commune of Marchigüe, Colchagua Valley, Central Chile. The main objective was to develop models using indices and spectral variables from Sentinel-2 images to approximate grape and wine quality parameters, such as °Brix and pH, as well as color index, total anthocyanins and total phenols of the wine. It is expected that these models, applied across the entire work area, will allow for differentiated crop management, improving wine quality by predicting pre- and postharvest indicators that influence sensory characteristics and brand differentiation. In the pre-harvest variables, shortwave bands and vegetation index are more frequent, while in the post-harvest variables, red-edge bands, near infrared and moisture index stand out. These results suggest the need to conduct further studies or explore other species. Satellite images are valued as valuable tools to improve the artisanal process and obtain better quality wines, highlighting the work methodology adaptable to diverse environmental conditions. One of the most interesting findings is the determination of the optimal physiological period for predictive analysis, specifically the fortnight prior to harvest, allowing precise and timely planning of the harvest process and providing essential data for winemakers. Regarding the coefficients of determination obtained for °Brix, color index and total anthocyanins in the first season, the results are highly positive, with values between 65% and 77%, suggesting a reliable prediction and contributing to the definition of selective harvesting. Other variables show less conclusive relationships, with coefficients of determination between 42% and 57% in the 2017 and 2018 seasons, indicating the need to improve the study by increasing the number of samples or refining the sampling process. The positive results and the availability of satellite images offer the opportunity to identify areas for detailed studies and differentiated vintages. Accurate selection of grapes with similar characteristics is crucial to improve wine quality and productivity. This study highlights the feasibility of remote sensing and selective harvesting as key tools for managing wine quality and style, providing new perspectives in viticulture. / Fredes Castro, SN. (2024). Modelación de parámetros de calidad de uva vinífera con índices satelitales provenientes de imágenes de resolución espacial media para el manejo diferenciado de viñedos en Chile [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/211473