In the United States, hops (Humulus lupulus L.) are grown mainly in the Pacific Northwest (PNW). For this reason, most cultural information is based on the growing conditions of the PNW. Growing conditions in Virginia differ drastically and present unique disease and production challenges. Three studies were conducted with the intent of increasing hop cultivation knowledge for Virginia growers. For the first study, 13 cultivars of hops grown at the Virginia Tech hop yard were compared for growth, yield, and quality. Mean cone fresh weight per plant ranged from 12.00 g for Mt. Hood to 1002.87 g for Crystal in 2016 and from 97.98 g for Mt. Hood to 900.33 g for Cascade in 2017. In 2016, only Alpharoma, Cascade, Mt. Rainier, and Southern Cross had alpha acid levels, an indicator of cone quality, within the accepted range. In 2017, Alpharoma, Centennial, Mt. Rainier, and Nugget had alpha acid levels within the expected range. Three cultivars (Cascade, Crystal, and Ultra) were above the expected alpha acid range, which indicates more bittering potential for beer brewers. In a nutrient deficiency study, hop plants were grown in hydroponic solutions, and deficiencies were induced for nitrogen (N), phosphorus (P), and potassium (K). After visual deficiency symptoms had been induced, leaf tissue samples were taken and analyzed for nutrient content. Images were taken at each deficiency stage. For N, incipient deficiency symptoms were observed at a mean of 3.18% dry weight in leaf tissue samples. Visual symptoms included a chlorotic appearance, undersized leaves, and red petioles. Incipient symptoms for P were observed at a mean of 0.307% dry weight in leaf tissue samples. Necrotic spots, leaf cupping, and undersized leaves were apparent with this deficiency. Incipient symptoms for K were observed at a mean of 1.21% dry weight in leaf tissue samples. Symptoms included rounded leaf tips, blue veins, and marginal scorch. In the third study, a whole leaf powdery mildew (Podosphaera macularis) assay was developed and tested using five hop cultivars, Alpharoma, Cascade, Comet, Sorachi Ace, and Tahoma. Leaves were inoculated with powdery mildew (PM) using a settling tower. This method was used to rapidly assess the resistance of cultivars. Leaves were successfully inoculated and PM colonies were allowed to grow for two weeks. Images of the PM colony development on inoculated leaves were compared using ImageJ to determine percentage of coverage. Tahoma was the only cultivar found to produce a significantly different mean percent coverage (19.5%) compared with the resistant cultivars Cascade and Comet (<1%). / Master of Science in Life Sciences / Hops (Humulus lupulus) were first grown in the United States in Massachusetts in the early 1600s. Production of this crop eventually spread throughout the Northeastern US. By the mid-1800s, commercial production spread to southern states such as Virginia. Infestation of pests, e.g., hop aphid and diseases such as downy mildew in eastern states, and prohibition on the production of alcohol caused a significant shift in hop production, which favored the Pacific Northwest (PNW). As a result, in Virginia specific knowledge of hop cultivation is now lacking. Three studies were conducted to increase our knowledge of hop cultivation for the region. For the first study, 13 cultivars grown at the Virginia Tech hop yard were compared for growth, yield, and quality. The cultivars Cascade and Alpharoma had alpha acid levels equal to or higher than expected for the 2016 and 2017 seasons. The alpha acid levels are an indicator of the hop quality from the Cascade and Alpharoma cultivars. Cascade was also a top producer of hop cones with a mean fresh weight of 989.67 g and 900.33 g in 2016 and 2017, respectively. In a nutrient deficiency study, plants were grown in Hoagland hydroponic solutions, which contain all essential nutrients needed for plant growth. This was done to provide photographic documentation to assist local growers with deficiency identification. Deficiencies were induced for nitrogen (N), phosphorus (P), and potassium (K) by removing each of the specific nutrient of interest from each treatment solution. After deficiency symptoms were induced, leaf tissue samples were taken and analyzed for nutrient content. Nutrient deficiencies were documented using photographs. For N, visual symptoms included a chlorotic appearance, undersized leaves, and red petioles. Incipient symptoms for P included necrotic spots, leaf cupping, and undersized leaves. For K, leaf cupping, blue green veins, marginal scorch, and rounded leaf tips. Disease resistance of cultivars is important information for growers and can heavily influence hop yard planning. As such, in another study, an assay for powdery mildew (Podosphaera macularis) was developed that allows for rapid low cost testing of hop cultivars. The assay was tested on the following cultivars: Alpharoma, Cascade, Comet, Sorachi Ace, and Tahoma. Leaves were inoculated with powdery mildew (PM) using a settling tower. Powdery mildew colonies were allowed to grow for two weeks and were then analyzed using ImageJ software to determine percent coverage. At the end of the experiment, Tahoma had significantly greater PM coverage compared to the other cultivars, indicating that Tahoma is less resistant to the specific PM strain. / Hops (Humulus lupulus) were first grown in the United States in Massachusetts in the early 1600s. Production of this crop eventually spread throughout the Northeastern US. By the mid-1800s, commercial production spread to southern states such as Virginia. Infestation of pests, e.g., hop aphid and diseases such as downy mildew in eastern states, and prohibition on the production of alcohol caused a significant shift in hop production, which favored the Pacific Northwest (PNW). As a result, in Virginia specific knowledge of hop cultivation is now lacking. Three studies were conducted to increase our knowledge of hop cultivation for the region. For the first study, 13 cultivars grown at the Virginia Tech hop yard were compared for growth, yield, and quality. The cultivars Cascade and Alpharoma had alpha acid levels equal to or higher than expected for the 2016 and 2017 seasons. The alpha acid levels are an indicator of the hop quality from the Cascade and Alpharoma cultivars. Cascade was also a top producer of hop cones with a mean fresh weight of 989.67 g and 900.33 g in 2016 and 2017, respectively. In a nutrient deficiency study, plants were grown in Hoagland hydroponic solutions, which contain all essential nutrients needed for plant growth. This was done to provide photographic documentation to assist local growers with deficiency identification. Deficiencies were induced for nitrogen (N), phosphorus (P), and potassium (K) by removing each of the specific nutrient of interest from each treatment solution. After deficiency symptoms were induced, leaf tissue samples were taken and analyzed for nutrient content. Nutrient deficiencies were documented using photographs. For N, visual symptoms included a chlorotic appearance, undersized leaves, and red petioles. Incipient symptoms for P included necrotic spots, leaf cupping, and undersized leaves. For K, leaf cupping, blue green veins, marginal scorch, and rounded leaf tips. Disease resistance of cultivars is important information for growers and can heavily influence hop yard planning. As such, in another study, an assay for powdery mildew (Podosphaera macularis) was developed that allows for rapid low cost testing of hop cultivars. The assay was tested on the following cultivars: Alpharoma, Cascade, Comet, Sorachi Ace, and Tahoma. Leaves were inoculated with powdery mildew (PM) using a settling tower. Powdery mildew colonies were allowed to grow for two weeks and were then analyzed using ImageJ software to determine percent coverage. At the end of the experiment, Tahoma had significantly greater PM coverage compared to the other cultivars, indicating that Tahoma is less resistant to the specific PM strain.

L'utilisation des champignons mycorhiziens à arbuscules (CMA) pourrait constituer une alternative potentielle aux traitements fongicides conventionnels pour lutter contre les maladies cryptogamiques des plantes. Notre travail a consisté à étudier l'éventuel effet protecteur de la mycorhization arbusculaire chez le blé tendre (triticum aestivum L.) contre Blumeria graminis f. sp. tritici (Bgt), un champignon biotrophe responsable de l'oïdium, une maladie affectant les parties aériennes de la plante. L'inoculation mycorhizienne du blé avec le CMA Funneliformis mosseae (FM), en conditions controlées et optimisées, nous a permis d'obtenir parallèlement à un taux de mycorhization de 38% des plants de blé, une amélioration significative de la biomasse et un taux de protection contre Bgt estimé à 78%. Ces résultats suggèrent l'induction d'une résistance systémique des réactions de défense du blé par mycorhization (Mycorrhiza-Induced Resistance, MIR). Cette protection serait liée à une accumulation de composés phénoliques et de preoxyde d'hydrogène dans les cellules épidermiques des feuilles de blé mycorhizé, au niveau du site de pénétration de Bgt. Une surexpression des gènes POX, PAL, CH11 et NPR1 codant pour des marqueurs de défense a également été mise en évidence dans les feuilles en absence d'infection par Bgt. Enfin, nos travaux ont également souligné l'intégration de divers paramètres pour optimiser l'utilisation des CMA comme agents de biocontrôle chez le blé. La meilleure protection contre l'oïdium a été obtenue aec un apport en phosphore réduit de 5 fois par rapport à celui préconisé au champ et un inoculum mycorhizien à base de Fm, que ce soit chez un cultivar modérément sensible ou un cultivar plus résistant. / The use of arbuscular mycorrhizal fungi (AMF) could be an innovative alternative to chemicals against fungal plant diseases. Our work aimed at studying the possible protective effect of arbuscular mycorrhization in the bread wheat (Triticum aestivum L.) against Blumeria graminis f. sp. tritici (Bgt), a biotrophic fungi responsible of wheat powdery mildew, a disease affecting the aerial plant organs. Wheat mycorrhizal inoculation by Funneliformis mosseae (Fm), under controlled and optimized conditions, allowed us to obtain concomitantly a micorrhizal rate of 38%, a significant increase of plant biomass and a protection level against Bgt estimated at 78%. These results suggest the induction of systemic wheat defense reactions resulting from mycorrhization (Mycorrhiza-Induced Resistance - MIR). This protection is linked to an accumulation of phenolic compounds and hydrogen peroxide at the Bgt penetration sites in epidermal leaf cells of mycorrhized wheat plants. Up-regulations of POX, PAL, NPR1 and CH11 genes encoding for defense markers were also pointed out in leaves of mycorrhizal wheat in the absence of Bgt infection. Moreover, our study highlighted the importance of taking into account various parameters to optimize the use of AMF as biocontrol agents. The highest protection against powdery mildew was obtained with a 5-fold reduced phosphorus input compared to that recommended in the field and with the mycorrhizal inoculum Fm, in both a moderately susceptible or a more resistant cultuva

O meloeiro é uma cultura de relevância econômica no Brasil, principalmente na região Nordeste, onde grande parte da produção é exportada. Entre os fatores limitantes da cultura está o oídio, que pode ser controlado através do uso de variedades resistentes. No entanto, para que isto seja possível, é necessário que se tenha conhecimento sobre qual é o agente causal em ocorrência na região de produção e quais são as fontes de inóculo desta doença. O presente trabalho teve como objetivos o levantamento das espécies e raças de oídio com ênfase na região Nordeste, a verificação da patogenicidade de isolados provenientes de outras cucurbitáceas em relação ao meloeiro, a avaliação de germoplasma de meloeiro para resistência a P. xanthii raça 2 (francesa), a proposição de um sistema de cultivares de meloeiro para caracterização de raças de P. xanthii e a otimização de um protocolo de extração de DNA de P. xanthii. No total de 65 isolados analisados, observou-se que Podosphaera xanthii foi a única espécie encontrada entre as amostras. Dentre estas, as raças 1 e 2 foram as prevalentes. Também foram observadas as raças 0, 3, 4 e 5 em menor freqüência. As culturas de abóbora, abobrinha, melancia e pepino apresentaram-se como fontes de inóculo de oídio para o meloeiro, uma vez que todos os isolados provenientes destas culturas colonizaram discos foliares de meloeiro. Em geral, os genótipos de melão cultivados comercialmente no Brasil apresentaram-se resistentes a raça 2 (francesa) de P. xanthii. A partir da observação das raças ocorrentes no país e de dados da literatura, propôs-se um sistema de cultivares de meloeiro diferenciadoras para raças de P. xanthii a ser empregado em monitoramentos do patógeno. Um protocolo para extração de DNA de P. xanthii também foi obtido com a finalidade de facilitar estudos genéticos sobre o patógeno com base em marcadores moleculares. / Melon is an economically important crop in Brazil, mainly in the north-east region, where most of the production is exported. Among the limiting factors of the crop is powdery mildew, which can be controled by using resistant varieties. However, to make it possible, it is necessary to know the causal agent in occurrence in the production region and what are the inoculum sources of this disease. The present work had as objectives the survey of the powdery mildew species and races with emphasis in the north-east region; the verification of the strains pathogenicity deriving from others cucurbits in melon; the evaluation of melon germplasm to Podosphaera xanthii race 2 (french) resistance; the proposition of a system of melon cultivars to characterize the P. xanthii races and the optimization of a protocol to P. xanthii DNA extraction. Over all of the 65 strains analysed, it was observed that P. xanthii was the unique specie encountered between the samples. Within that, the races 1 and 2 were prevalents. Also were observed the races 0, 3, 4 and 5 at low frequency. Squash, zucchini, watermelon and cucumber crops showed up like inoculum sources of powdery mildew to melon. Usually, the melon genotypes commercially cultivated in Brazil showed up resistant to P. xanthii race 2 (french). After the observation of races in occurrence and datas from literature, a system of melon cultivars was proposed to differentiate P. xanthii races to be used in surveys of the pathogen. A protocol to extraction P. xanthii DNA also was proposed with the purpose to facility genetics studies about the pathogen basis on molecular markers.

At present, the global population is increasing, while soil and fresh water resources for crop production are declining. It is important to adopt sustainable practices to optimise the use of limited natural resources without compromising the environment, and to enhance continuous production in the long term. The rapid growth of UK strawberry industry has been achieved through the precision use of varieties, nutrients and polythene tunnels. This intensive production has caused significant environmental impacts especially Greenhouse Gas (GHG) emissions from the production. Strawberry powdery mildew (Podosphaera aphanis) is a major fungal disease affecting strawberry production worldwide particularly in polythene tunnels. The disease can result in yield losses of up to 70% of the crop. A ruleQbased system was used in the field trials to predict high risk days of P. aphanis development, taking into account the optimal environmental conditions conducive to conidial germination and disease development. The results (Chapter 3) showed that the use of this prediction system achieved satisfactory control of P. aphanis in commercial strawberry production, with reduced fungicide applications compared with commercial spray programme. The results were consistent in two consecutive years and on different varieties. In addition, it was suggested that the use of the prediction system may also lead to lower GHG emissions associated with fewer fungicide applications, thereby benefit strawberry growers both environmentally and economically. Results from 2014 & 2015 silicon fertigation trials showed that the use of a silicon nutrient via the fertigation system reduced the strawberry susceptibility to P. aphanis and twoQspotted spider mites (Tetranychus urticae Koch) in two consecutive years on different varieties (Chapter 4). In both years, crops received the silicon nutrient only without fungicides had both lower rate of epidemic (r) and lower value of Area Under the Disease Progress Curve (AUDPC) (r = 0.0036, AUDPC = 475 in 2014; r = 0.001, AUDPC = 267 in 2015) compared with the untreated control (r = 0.0042, AUDPC = 662 in 2014; r = 0.0011, AUDPC = 281 in 2015). Silicon also delayed the epidemic buildQup in the silicon nutrient only treatment for approximately two weeks compared with the untreated control. Crops from the silicon nutrient plus fungicides treatment had lower susceptibility (r = 0.0012 in 2014; r = 0.0004 in 2015) than those from the fungicides only treatment (r = 0.0017 in 2014; r = 0.0005 in 2015) suggesting that the silicon nutrient may also enhance fungicides performance in reducing the epidemic buildQup when used together. Moreover, the presence of T. urticae on strawberry leaves was significantly lower (P < 0.001) in plants treated with the silicon nutrient than those without. In addition, initial results suggested that silicon may play a positive role in raising °Brix of strawberry leaf petiole, improving pollen viability, and influencing the length of flower receptacle and stamens. Maltmas Farm has a wide range of semiQnatural habitats that provide food and nesting resources for wild pollinators. Hoverflies, bumblebees and solitary bees were found to be the main wild pollinators that pollinate commercial strawberries at Maltmas Farm (Chapter 5). The number of pollinators in tunnels or open fields significantly correlated with the abundance of strawberry flowers (P < 0.05). Pollinator presence also differed between groups throughout the day and over the seasons. Hoverflies appeared early in the day and were abundant in summer months; bumblebees and solitary bees were present most of the day and throughout the season, whereas honeybees were only active in sunny days. Temperatures, relative humidity and cloud coverage also affected pollinator presence. In addition, pollinator activity was not significantly (P > 0.05) affected by the application of the silicon nutrient via the fertigation system. The integrated use of the prediction system (to reduce fungicide applications and subsequent GHG emissions), the silicon nutrient (to reduce crop susceptibility to P. aphanis and T. urticae), and sustainable farmland management (to encourage the presence of wild pollinators) could help strawberry growers to achieve a more sustainable production.

Made available in DSpace on 2017-07-20T13:41:59Z (GMT). No. of bitstreams: 1 HELENTON CARLOS DA SILVA.pdf: 14476259 bytes, checksum: 0a42181cc955a30bfda103115227d8ab (MD5) Previous issue date: 2016-03-31 / This paper aims to propose a treatment system for wastewater from the construction and demolition waste washing water (RCD). Therefore, the physico-chemical characterization of a lot of stuff, from washing effluent in different proportions of water was performed. A mass of material was washed following the criteria established in the Brazilian Norm Standard MERCOSUR 46 of 2003, for determination by physical removal of the pulverulent material (particles smaller than 75 μm), which damage the quality of the mortar. The RCD has been classified as fine aggregate with a maximum size of 4,8 mm and feature average fineness modulus of 2,77. For washing RCD according to the specified in the standard no specific amount of water for performing the process. Thus, the present study aimed to optimize this amount obtaining an optimum rate of mass RCD for a water volume of 1,67 : 1,00. The resulting effluent from the washing of ash RCD presented pH between 7,54 and 8,34, acidity from 9,09 to 15,25 mg CO2.L-1 and 10,33 to 17,33 mg of CaCO3.L-1 . The turbidity of 70,93 NTU varied more than 300 NTU. Following proposed treatment for the effluent with three treatment options: a sedimentation tank; by combined sedimentation tank in series with sand filter; processes for coagulation, flocculation and sedimentation. And after the analysis of these three systems, it was observed that all showed efficiency in some parameters analyzed, highlighting that the turbidity was removed by 99 % in the three cases. The process was less efficient coagulation, flocculation and settling by the use of the coagulant of aluminum sulfate in 100 mg.L-1 concentration. It also emphasizes that the COD was removed more than 90 % in the three proposed processes. Thus, it was concluded that the best treatment system was observed by physical and biological combined removal by settling tank in series with a sand filter. / Este trabalho tem por objetivo propor um sistema de tratamento para o efluente proveniente da água de lavagem de resíduos de construção e demolição (RCD). Para tanto, foi realizada a caracterização físico-química de um lote do material, do efluente proveniente da lavagem em diferentes proporções de água. Uma massa do material foi lavada seguindo os critérios estabelecidos na Norma Brasileira Regulamentadora Norma MERCOSUL 46 de 2003, para a determinação por remoção física do material pulverulento (partículas menores que 75 μm), que prejudicam a qualidade da argamassa. O RCD foi classificado como agregado miúdo,com dimensão máxima característica de 4,8 mm e módulo de finura médio de 2,77. Para a lavagem do RCD de acordo com o especificado na norma não há quantidade específica de água para a realização do processo. Desta forma o presente estudo visou otimizar essa quantidade obtendo-se uma taxa ótima de massa de RCD para um volume de água de 1,67 : 1,00. O efluente resultante da lavagem do RCD cinza apresentou pH entre 7,54 e 8,34, acidez de 9,09 a 15,25 mg de CO2.L-1 e 10,33 a 17,33 mg de CaCO3.L-1. A turbidez variou de 70,93 NTU a mais que 300 NTU. Na sequência foi proposto o tratamento para o efluente com três opções de tratamento: por tanque de sedimentação; por tanque de sedimentação combinado em série com filtro de areia; por processos coagulação, floculação e sedimentação. E após a análise desses três sistemas, observou-se que todos apresentaram eficiência em alguns parâmetros analisados, destacando-se que a turbidez foi removida em 99 % nos três processos. O processo menos eficiente foi à coagulação, floculação e sedimentação com a utilização do coagulante sulfato de alumínio em 100 mg.L-1 de concentração. Destaca-se ainda que a DQO foi removida mais de 90 % nos três processos propostos. Assim, concluiu-se que o melhor sistema de tratamento observado foi por remoção física e biológica por tanque de sedimentação combinada em série com filtro de areia.

A cultura do meloeiro é de grande importância econômica no Brasil, sendo a região Nordeste a principal produtora e exportadora do fruto. A baixa resistência aos principais patógenos é um dos fatores limitantes para a competitividade brasileira da cultura no mercado internacional. Uma das doenças mais importantes é o oídio, causado por Podosphaera xanthii. Geralmente, o controle de P. xanthii é obtido com o uso de fungicidas. Porém, é necessário empregar medidas mais econômicas e que não agridam o ambiente, como a utilização de cultivares resistentes. O objetivo deste trabalho foi identificar marcadores moleculares de vários tipos ligados ao gene Pm-1 de resistência à raça 1 de P. xanthii com o intuito de auxiliar programas de melhoramento genético. Para tanto foram utilizadas duas linhagens quase-isogênicas (LQI) de melão AF426pm1 (P1) e AF426Pm1 (P2), ambas pertencentes à variedade inodorus, que são contrastantes para ausência e presença, respectivamente, do gene Pm-1. Para a análise de co-segregação entre gene e marcadores candidatos, foi utilizada uma população de retrocruzamento RC1F1, fenotipada para resistência a raça 1 de oídio, obtida a partir do cruzamento entre linhagens. As técnicas de LM-PCR e AFLP resultaram em marcadores polimórficos, porém somente os encontrados com a técnica de AFLP puderam ser utilizados como marcadores no teste de co-segregação. Foram encontrados dois marcadores AFLP. No entanto, somente um, denominado HF155 mostrou-se ligado a uma distância de 4,9 cM do gene Pm-1. Devido à proximidade deste marcador ao gene este pode ser utilizado em programas de seleção assistida por marcadores, visando o melhoramento de linhagens com resistência ao P. xanthii. / Melon crop is of great economical importance in Brazil being the Northeast region the main producer and exporter of the fruit. The low levels of resistance to pathogens is one of the restricting factors for the Brazilian competitiveness worldwide. One of the most important diseases is powdery mildew caused by Podosphaera xanthii. Generally, the control of P. xanthii is achieved with the use of fungicides. However it is necessary to use less expensive control methods with a minimum environmental impact such as resistant cultivars. The objective of this work was to identify molecular markers linked to the Pm-1 gene which confers resistance to race 1 of P. xanthii with the purpose of assisting boding program. Two near isogenic lines (LQIs) of melon Agro AF426pm1 (P1) and AF426Pm1 (P2), both belonging to the inodorus variety, which are respectivally contrasting for abscence and presence of Pm-1 gene were analyzed. For the cosegregation analyses between gene and candidate markers, it was used the BC1F1 population was used screened for resistance to powdery mildew and obtained from a cross between the LQIs. The LM-PCR and AFLP techniques were efficient in the polymorphisms detection, however only those ones which were found using AFLP technique could be used as markers in the co-segregation test. It was found two markers with this technique but just the HF155 is located 4.9 cM of the Pm-1 gene. Due the proximity of the HF155 marker to the Pm-1 gene this one can be used in markerassisted selection aiming to develop melon cultivars resistant to P. xanthii.

La phénologie du débourrement est un caractère majeur d’adaptation des arbres à leurenvironnement en milieu tempéré. Notre objectif a été de caractériser les contraintes biotiques (oïdium) etabiotiques (températures hivernales et printanières / gels tardifs) s’exerçant sur le débourrement afind’expliquer les patrons de variation phénologique intra et inter populationnelle observés chez le chêne(Quercus petraea) le long d’un gradient altitudinal. Nous avons utilisé une approche combinantobservations in situ, expérimentation, et modélisation. Nous avons mis en évidence que l’évitement desgels tardifs printaniers est un caractère adaptatif majeur le long du gradient altitudinal. La tardiveté dudébourrement pourrait être due à des besoins plus importants en température de forcing. Par ailleurs, lechampignon n’est pas adapté localement à la phénologie de son hôte et les individus et les populationssont alors inégalement exposés à la maladie. En montant en altitude, les chênes sont de plus en plusexposés au champignon, mais les facteurs environnementaux sont défavorables à une plus forte infection.A basse altitude, l’oïdium et les gels tardifs favorisent des phénotypes phénologiques opposés(respectivement précoces vs. tardifs) ; la combinaison des deux contraintes pourrait donc contribuer aumaintien de la forte diversité phénologique observée. D’autre part, nous avons observé que l’infection parl’oïdium engendre une augmentation du polycyclisme chez les semis de chêne au cours de la saison decroissance, ce qui les rend moins résistants aux gels hivernaux. Nous montrons qu’il est important que lesmodèles phénologiques à visée prédictive intègrent la phase de chilling aboutissant à la levée dedormance. Le manque de chilling ne semble pas encore un facteur limitant, mais la tendance actuelle à undébourrement de plus en plus précoce sera probablement freinée voire inversée au milieu du siècle enbasse altitude, dans la marge sud de distribution de Q. petraea. / Budburst phenology is a major adaptive trait of trees to the environment in temperateclimate. Our aim was to characterize the biotic (powdery mildew) and abiotic (winter and springtemperatures / spring frost) constraints acting on budburst in view to explain the patterns of intra and interpopulations’ phenological variation observed in sessile oak (Quercus petraea) along an elevation gradient.We based our approach on in situ monitoring, experimentation and modeling. Our results highlight that theavoidance of late spring frosts is a major adaptive trait along the elevation gradient. The lateness inbudburst might be due to higher requirements in forcing temperatures. Otherwise, the fungus is not locallyadapted to its host phenology so oak individuals and populations are unequally exposed to the disease.With increasing elevation, oaks are more and more exposed to the fungus, but the environmental factorsare unfavorable to higher infection. At low elevation, powdery mildew and late spring frosts favor oppositephonological phenotypes (early-flushing vs. late-flushing trees, respectively); the combination of the twopressures may thus contribute to the maintenance of the observed high phenological diversity. We alsoshowed that powdery mildew infection induced an increased polycyclism during the growing season in oakseedlings, which made them less resistant to winter frosts. Predictive phenological models will have toinclude the chilling phase which conditions dormancy breaking. Although the lack of chilling is not yet alimiting factor, the current trend in increasingly advanced budburst will certainly be slowed or even reversedin the middle of the century at low elevation, in the southern margin of the distribution area of Q. petraea.

La cendrosa de l'ordi, causada pel fong Blumeria graminis (D.C.) E. O. Speer f. sp. hordei Em. Marchal, sin. Erysiphe graminis D.C.: Fr. f. sp. hordei Em. Marchal, anamorf: Oidium monilioides (Nees) Link, és la malaltia aèria més important en aquest conreu a Catalunya. El 1987 es van iniciar les investigacions sobre l'etiologia de les principals malalties dels cereals. Es realitzà una prospecció per tota la l’àrea cerealícola de la geografia catalana, que va permetre estimar la distribució i importància de les malalties que afecten l'ordi, el blat, el panís i l'arròs. Els resultats del període estudiat, 1987-1990, conclogueren que per a la cendrosa de l’ordi, la probabilitat d’una epidèmia severa (severitat final=60%5% en varietats susceptibles) varià entre p=0,2 i p=0,6 (entre dos i sis anys amb epidèmia severa de cada deu anys de cultiu), segons zones de conreu. Amb aquestes premisses es va anar elaborant un sistema de suport a la presa de decisions (SSPD) propi per a la cendrosa de l’ordi a l’àrea de Catalunya anomenat CENCONT, el qual prediu la malaltia tenint en compte que la cendrosa forma part d’un ‘complex de malalties’ i considerant la severitat (i l’àrea sota la corba epidèmica) com a variable independent i el possible efecte de la resta de malalties i plagues com a covariants. Per a l’aplicació pràctica de les anàlisis epidèmiques a CENCONT, es van estimar els valors dels paràmetres del model Logist com mitjanes d’un nombre de epidèmies representatives dels distints processos: de reacció varietal i de control químic. Aquests valors van resultar útils per a la predicció d’epidèmies i l’avaluació de CENCONT va demostrar la seva utilitat, tant pel que fa a la seva precisió com per a l’anàlisi econòmica fonamentada en el càlcul de la Funció de Guany. Però les assumpcions teòriques fetes en ajustar models tipus Logist, anomenats sintètics, a les dades epidèmiques de camp, han de permetre fer prediccions amb un marge d'error acceptable, en les condicions ambientals definides per cada predicció, la qual cosa s’ha de demostrar que és certa en una eina SSPD. Per aquesta raó els objectius que s’han formulat en aquest treball s’han basat resumidament en: 1) Fer una anàlisi matemàtica dels models sintètics, principalment els Logist i Richards, veient el seu comportament i estudiant les relacions y0, r i k mitjançant simulació, 2) Contrastar els resultats teòrics amb els de les epidèmies desenvolupades en condicions naturals, aplicant els models, en ajusts a epidèmies de camp de la cendrosa de l'ordi, amb la condició d'anar fixant els paràmetres o bé deixar-los lliures en el model, principalment l'asímptota k, i 3) Discutir la validesa de les assumpcions fetes en els models per ajustar les equacions a les dades experimentals, veient les interaccions entre paràmetres i les conseqüències que derivades de l'anàlisi comparativa d'epidèmies i la seva aplicació a sistemes predictius de la malaltia. Per assolir els objectius plantejats s’han utilitzat 113 processos epidèmics observats en camp en el període 1991-2002, avaluant principalment la intensitat de malaltia en el temps i basats en dos objectius experimentals: control químic i resistència varietal Paral•lelament, s’han analitzat des del punt de vista matemàtic mitjançant el programa MAPLE els models exponencial, Logist, Gompertz, Monomolecular, Von Bertalanffy-Richards i Weibull, i s’ha valorat la seva idoneïtat en la seva aplicació a malalties vegetals, especialment a la cendrosa de l’ordi. Després de l’anàlisi matemàtica, s’han aplicat els models mitjançant el programa Table Curve 2D©. En una primera fase s’han ajustat tots els models a les epidèmies de camp, excepte el model exponencial, i en una segona, de tots els models aplicats s’han seleccionat els que donaven millor rendiment estadístic pel que fa als diversos paràmetres d’avaluació. En aquest procés d’ajust, també s’han descartat les epidèmies que no aconseguien un ajust de suficient qualitat segons els criteris establerts. A més, s’han establert comparacions de les taxes relatives epidèmiques obtingudes en els ajusts mitjançant la utilització de la taxa mitjana ponderada absoluta. S’ha fet una anàlisi especial del comportament del paràmetre de forma del model Richards i s’han elaborat models de superfície de resposta per als paràmetres epidèmics y0, r i k dels models Logist i Richards. Entre les conclusions més importants s’ha trobat que el model Richards s’ajusta millor que Logist i aquest que Gompertz al conjunt d’epidèmies de cendrosa de l’ordi a Catalunya. La resta de models analitzats es van descartar per pitjor comportament. Quant als paràmetres dels models, en primer lloc no és el mateix fixar l’asímptota que no fixar-la, com a mínim des de la perspectiva estadística. En segon lloc, la taxa relativa epidèmica disminueix de valor segons si s’ajusten les dades a Richards, Logist o Gompertz respectivament, i això passa sempre respecte qualsevol variant dels models, asímptota lliure o bé asímptota fixa, i per a qualsevol objectiu epidèmic, control químic o bé resistència varietal. Respecte a la severitat inicial, no s’ha trobat una pauta de comportament definida. Sobre la predicció, es pot fer una millora d’aquesta en el sistema d’avisos CENCONT passant del model Logist al model Richards. El model Richards es pot utilitzar amb els tres paràmetres clàssics o bé amb quatre paràmetres, incorporant el paràmetre de forma però amb unes certes restriccions. Si s’agafa el model Richards triparamètric, serà necessari fixar el paràmetre de forma a un valor que raonablement pot ser inferior a 5. Aquest tipus de modelització és la recomanada per a incloure en Sistemes de suport a la presa de decisions com és el CENCONT. La millora en la predicció no implica necessàriament una millora en la presa de decisions del sistema d’avisos de la cendrosa de l’ordi a Catalunya, perquè les implicacions econòmiques no són en la major part dels casos prou importants com per provocar un canvi de decisió. La diferència màxima calculada de pèrdues associades estimades segons el model Richards, amb asímptota lliure i les associades segons el model Logist, se situa en el 2.9%, calculant-les a partir de l’ASCPM i de la severitat final. A efectes pràctics es pot considerar el mateix haver predit segons el model Richards, millor en el treball, o el Logist, el qual fou el que realment s’havia aplicat en el programa CENCONT. L’estimació Logist podria ser suficient per a la major part de les situacions (epidèmies) i que només en un petit percentatge hauria valgut la pena canviar de model. De tota manera, per al tipus de prediccions del present treball, no es veuria malament el fet d’adoptar el model Richards en un futur, amb la condició d’haver fixat el valor característic del paràmetre m per a la zona d’estudi. / El oídio de la cebada causado por el hongo Blumeria graminis (D.C.) E. O. Speer f. sp. hordei Em. Marchal, sin. Erysiphe graminis D.C.: Fr. f. sp. hordei Em. Marchal, anamorf: Oidium monilioides (Nees) Link, es la enfermedad aérea más importante en este cultivo en Cataluña. El año 1987 se iniciaron las investigaciones sobre la etiología de las principales enfermedades de los cereales. Se realizó una prospección por toda el área cerealista de la geografía catalana que permitió estimar la distribución e importancia de las enfermedades que afectan la cebada, el trigo, el maíz y el arroz. Los resultados del período estudiado, 1987-1990, concluyeron que para el oídio de la cebada, la probabilidad de una epidemia severa (severidad final=60%5% en variedades susceptibles) osciló entre p=0,2 y p=0,6 (entre dos y seis años con epidemia severa de cada diez años de cultivo), según zonas de cultivo. Con estas premisas se elaboró un sistema de soporte a la toma de decisiones (SSTD) propio para el oídio de la cebada en el área de Cataluña, llamado CENCONT, el cual predice la enfermedad teniendo en cuenta que el oídio forma parte de un ‘complejo de enfermedades’ y considerando la severidad (y el área bajo la curva epidémica) como variable independiente y el posible efecto del resto de enfermedades y plagas como covariables. Para la aplicación práctica de los análisis epidémicos en CENCONT se estimaron los valores de los parámetros del modelo Logist como medias de un número de epidemias representativas de los distintos procesos: reacción varietal y control químico. Estos valores resultaron útiles en la predicción de epidemias y la evaluación de CENCONT demostró su utilidad, tanto en lo que se refiere a su precisión como para el análisis económico fundamentado en el cálculo de la Función de Ganancias. Pero las asunciones teóricas realizadas en el ajuste de modelos tipo Logist, llamados sintéticos, a los datos epidémicos de campo, han de permitir predicciones con un margen de error aceptable, en las condiciones ambientales definidas para cada predicción, lo cual se debe demostrar como cierto en una herramienta SSPD. Por esta razón los objetivos que se han formulado en este trabajo se han basado resumidamente en: 1) Efectuar un análisis matemático de los modelos sintéticos, principalmente Logist y Richards, viendo su comportamiento y estudiando las relaciones y0, r i k mediante simulación, 2) Contrastar los resultados teóricos con los obtenidos de las epidemias desarrolladas en condiciones naturales, aplicando los modelos en ajustes a epidemias de campo del oídio de la cebada, con la condición de ir fijando los parámetros o bien dejarlos actuar libremente en el modelo, principalmente la asíntota k, y 3) Discutir la validez de las asunciones realizadas en los modelos para ajustar las ecuaciones a los datos experimentales, viendo las interacciones entre parámetros y las consecuencias derivadas del análisis comparativo de epidemias y su aplicación a sistemas predictivos de la enfermedad. Para conseguir los objetivos propuestos se han utilizado 113 procesos epidémicos observados en campo durante el período 1991-2002, evaluando principalmente la intensidad de enfermedad en el tiempo y basados en dos objetivos experimentales: control químico y resistencia varietal. Paralelamente se han analizado los modelos exponencial, Logist, Gompertz, Monomolecular, Von Bertalanffy-Richards y Weibul desde un punto de vista matemático mediante el programa MAPLE, y se ha valorado su idoneidad en la aplicación a enfermedades vegetales, especialmente al oídio de la cebada. Después del análisis matemático se han aplicado los modelos mediante el programa Table Curve 2D©. En una primera fase se han ajustado las epidemias de campo a todos los modelos excepto el modelo exponencial, y en una segunda fase de todos los modelos aplicados se han seleccionado aquellos que resultaban en un mejor rendimiento estadístico respecto a diversos parámetros evaluados. En este proceso de ajuste también se han descartado las epidemias que no conseguían un ajuste de suficiente calidad según los criterios preestablecidos. Además se han hecho comparaciones de las tasas relativas epidémicas obtenidas en los ajustes mediante la utilización de la tasa media ponderada absoluta. Se ha realizado un análisis especial del comportamiento del parámetro de forma del modelo Richards y se han elaborado modelos de superficie de respuesta para los parámetros epidémicos y0, r i k de los modelos Logist y Richards. Entre las conclusiones más importantes se destaca que el modelo Richards ajusta mejor que Logist y este que Gompertz respecto al conjunto de epidemias de oídio de la cebada en Cataluña. El resto de modelos analizados se descartaron debido a su peor comportamiento. Respecto a los parámetros de los modelos, en primer lugar no es lo mismo fijar la asíntota que no fijarla, como mínimo desde el punto de vista estadístico. En segundo lugar, la tasa relativa epidémica disminuye de valor según si se ajustan los datos a Richards, Logist o Gompertz respectivamente, sucediendo esto siempre respecto a cualquier variante de los modelos, asíntota libre o bien asíntota fija, y para cualquier objetivo epidémico, ya sea control químico o bien resistencia varietal. Respecto a la severidad inicial, no se ha encontrado una pauta de comportamiento definida. Sobre la predicción, se puede afirmar que se realizaría una mejora en el sistema de avisos CENCONT pasando del modelo Logist a utilizar el modelo Richards. El modelo Richards se puede utilizar con los tres parámetros clásicos o bien con cuatro parámetros, incorporando el parámetro de forma pero con ciertas restricciones. Si se escoge el modelo Richards triparamétrico, será necesario fijar el parámetro de forma hasta un valor que razonablemente puede ser inferior a 5. Este tipo de modelización es la recomendada para Sistemas de Soporte a la Toma de Decisiones como CENCONT. La mejora en la predicción no implica, sin embargo, una mejora en la toma de decisiones del sistema de avisos del oídio de la cebada en Catalunya, porque las implicaciones económicas no son en la mayor parte de los casos suficientemente importantes como para provocar un cambio de decisión. La diferencia máxima calculada de pérdidas asociadas, estimadas según el modelo Richards con asíntota libre y las asociadas según el modelo Logist se sitúa en el 2.9%, calculándolas a partir de l’ABCPE (Área Bajo la Curva de Progreso de la Enfermedad) y de la severidad final. A efectos prácticos se puede considerar como lo mismo el haber predicho según el modelo Richards, mejor en el trabajo, o el Logist, el cual fue el que realmente se había aplicado en el programa CENCONT. La estimación Logist podría ser suficiente para la mayor parte de las situaciones (epidemias) ya que solamente en un pequeño porcentaje habría valido la pena cambiar de modelo. De todas formas, para el tipo de predicciones del presente trabajo, no se vería como inconveniente el adoptar el modelo Richards en un futuro, con la condición de fijar un valor característico del parámetro m para la zona de estudio. / Powdery mildew caused by the fungus Blumeria graminis (D.C.) E. O. Speer f. sp. hordei Em. Marchal, sin. Erysiphe graminis D.C.: Fr. f. sp. hordei Em. Marchal, anamorf: Oidium monilioides (Nees) Link, is the most important disease on barley crop in Catalonia (Northeast of Spain). Surveys of cereal –barley, wheat, maize and rice– diseases in Catalonia were carried out on 1987 and provided the necessary data of the relative importance and crop losses of diseases on these crops. The results sowed that, in the period of 1987 to 1990, barley Powdery mildew disease had a probability of severe epidemic of p=0,2 and p=0,6 (final severity=60%5% in susceptible cultivars) depending on the sub-area. This means that frequency of critic disease is between two and six years in a ten years period of time. Those studies gave as a result the elaboration of a Decision Support System (DSS) on Powdery mildew in Catalonia called CENCONT. This DSS gives predictions of the disease as a “disease complex”, considering severity (and AUDPC) an independent variable and the rest of the diseases and pests as covariates. To build CENCONT parameters of the Logist model were estimated as means of a diversity of numerous epidemics obtained from chemical control and resistance cultivar trials. These values had showed to be useful on predicting epidemics and consequently evaluating CENCONT in the point of view of economical analysis. But theoretical assumptions of those synthetic models, as Logist model, must be sufficient to predict in an acceptable range of error in the environmental conditions of a particular prediction. This conclusion should be demonstrated. Thus, the objectives of this work had been: 1) working out of mathematical modelling using synthetic models, especially Logist and Richards, to study their behaviour and the relationships between y0, r and k by means of program simulation, 2) to check theoretical conditions applied to epidemics developed in natural conditions, adjusting models to data and testing the parameter behaviour using different starting fitting values, and 3) to validate the adjusting conditions to experimental data, knowing the interactions between parameters and their consequences in Comparative Epidemiology and Decision Support Systems. To achieve with the objectives there have been used 113 field epidemics obtained between 1991 and 2002, assessing the disease intensity vs. time and based on two experimental objectives: Chemical Control and Cultivar Resistance. Eventually, the Exponential, Logist, Gompertz, Monomolecular, Von Bertalanffy-Richards and Weibull models have been analysed mathematically using MAPLE computer program, and assessed their fitness on applied disease epidemics, particularly to powdery mildew of barley. Furthermore, the models have been applied to field epidemic data using Table Curve 2D computer program. First, all models except Exponential were tested to know their ability to fitness and later some of the models were discarded. To eliminate models statistical yield criteria on fitting were used. As well as the models some epidemics were also discarded in this process using the criteria of poor quality results of fitting. Once more, comparisons between relative epidemic rate data were done by testing the suitability of weighted mean absolute rate, and an additional analysis to study the particular behaviour of shape parameter of Richards’ model was found necessary. Finally, a 3D model of the surface response curve of epidemic parameters y0, r and k of Logist and Richards previous 2D models was developed. Main results of this work are that behaviour of Richards’ model is better than Logist model to the fitting to barley powdery mildew epidemics in Catalonia. Yields of the rest of the models had been worst and are discarded of the normal use in our conditions. With regard to parameters, it is stated that the use of fixed asymptote is statistically different in front of free asymptote. Besides, after fitting the relative epidemic rate values decrease in the order of Richards’, Logist and Gompertz models. This happens independently of considering the use of free or fixed asymptote and both chemical control or cultivar resistance objectives. Moreover, the initial severity of disease does not show any pattern of behaviour. As a consequence of the data obtained, it is obvious that should be convenient to use Richards’ model instead of Logist in the CENCONT computer program. Richards’ model could be used with his three or four parameter form, but if the second, the use of shape parameter should be controlled and restricted in a short range of values. Within the form parameter, a value les than 5 should be recommended. This modelling pattern is suitable to include in those DSS as CENCONT is. The improvement on prediction does not means that making decisions using this DSS should be better because the economic consequences are not enough frequent and relevant to decide changing the pattern. The maximum difference of crop losses calculated comparing the use of Richards’ model instead of Logist model is about 2.9%. These estimates are of final severity or AUDPC of the epidemic. In fact, the use of Richards’ model or the Logist model does not normally bear differences in the economic recommendations of CENCONT which uses Logist model. So actually the Logist model could be sufficient to the most of environmental circumstances (epidemics) and only few of them should carry a model change. Instead of this, the use of Richards’ model in next future should be considered as a clear improvement of the prediction, but fixing the shape parameter to the particular conditions of the area should be recommended as necessary in this hypothetical case.

Thesis (Ph.D.) - University of KwaZulu-Natal, Pietermaritzburg, 2008. / Full text also available online. Scroll down for electronic link.

Made available in DSpace on 2016-08-12T19:18:32Z (GMT). No. of bitstreams: 1 ElaineWLPN_TESE.pdf: 787916 bytes, checksum: 2087efc39ac35d1b19fedd7e4362e6bf (MD5) Previous issue date: 2014-04-04 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Powdery mildew, caused by Podosphaera xanthii, is the most important foliar disease of cucurbits in the Northeast Brazil. Breeding for resistance is one of the best ways to control this pathogen. The objectives aims these work were to characterize morphologically and by microsatellites markers resistant accessions of melon, to study the inheritance and to identify SNP markers associated to resistance in the accessions AM55-1A and C-AC-15. Fifty-six melon plant accessions with different geographical origins were evaluated for resistance to P. xanthii using artificial inoculation procedure under greenhouse conditions. The resistant accessions C-AC-02, C-AC-09, C-AC-15, C-AC-34 and AM55-1A were characterized with morphological traits and microsatellites markers. The accessions C-AC-02, C-AC-09, C-AC-15, C-AC-34 exhibited characteristics of momordica botanical variety while the accession AM55-1A exhibited characteristics of acidulus botanical variety. The molecular characterization confirmed he results observed in morphological characterization. The inheritance of resistance in AM55-1A and C-AC-15 in crosses with susceptible cultivar Véndrantais was studied under greenhouses conditions. The segregations ratios for resistance/susceptibility observed in the different populations (F1, F2, BC1 and BC2) indicated a monogenic and dominant inheritance in AM55-1A and two independent genes with epispastic interaction double recessive control resistance in C-AC-15. The backcross population (BC1) derived from AM55-1A x Véndrantais and F2 population from C-AC-15 x Véndrantais were utilized to identify QTLs associated to resistance to P. xanthii using bulked segregation analysis with 243 SNP molecular markers. Two QTLs (Pm-AM55-1A_1 and Pm-A55-1A_2) were identified on ligation group V from AM55-1A x Véndrantais . One QTL (Pm-C-AC-15_5) on ligation group V and other (Pm-C-AC-15_9) on ligation group IX / O oídio causado por Podosphaera xanthii, é a doença foliar mais importante das cucurbitáceas no Nordeste do Brasil. O melhoramento genético para resistência é uma das melhores maneiras de controlar esse patógeno. Os objetivos deste trabalho foram caracterizar morfologicamente e por marcadores microssatélites acessos resistentes de melão, estudar a herança e identificar marcadores SNP associados à resistência nos acessos AM55-1A e C-AC-15. Cinqüenta e seis acessos de meloeiro com diferentes origens geográficas foram avaliados para a resistência a P. xanthii usando inoculação artificial em casa de vegetação. Os acessos resistentes C-AC-02, C-AC-09, C-AC-15, C-AC-34 e AM55-1A posteriormente foram caracterizados utilizando descritores morfológicos e marcadores microssatélites. Os acessos C-AC-02 , C-AC-09, C-AC-15, C-AC-34 exibiram características da variedade botanica momordica, enquanto o acesso AM55-1A exibiu características da variedade botânica acidulus. A caracterização molecular confirmou os resultados observados na caracterização morfológica. A herança da resistência em AM55-1A e C-AC-15 em cruzamentos com a cultivar suscetível Véndrantais foi estudada em condições de casa de vegetação. As razões de segregação resistência/suscetibilidade observadas nas diferentes populações (F1, F2, RC1 e RC2) indicou herança monogênica e dominante em AM55-1A e dois genes independentes com interação epistática duplo recessiva no controle genético em C-AC-15. A população de retrocruzamento (RC1) derivado do cruzamento AM55 -1A x Véndrantais e a população F2 do cruzamento C-AC-15 x Véndrantais foram utilizados para identificar QTLs associados à resistência a P. xanthii usando análise de bulk segregante com 243 marcadores SNP. Dois QTLs (Pm-AM55-A_1 e Pm-A55-1A_2) foram identificados no grupo de ligação V no cruzamento AM55-1A x Véndrantais . Foram identificados um QTL (Pm-C-AC-15_5) no grupo de ligação V e outro (Pm-C-AC-15_9) no grupo de ligação IX