Carbon dioxide is the most important greenhouse gas and the major factor leading to the global climate change problem. In previous studies, the ocean is considered to be the major storage of anthropogenic carbon dioxide. However, evaluation of the global CO2 flux seldom includes the estuarine and coastal regions. It should be noted that the current estimate is based on a very limited data set. In particular, data from subtropical and tropical river estuaries are scarce. Many researches point out that the estuary is a CO2 source to the atmosphere, but the data is insufficient so one couldn¡¦t obtain the total CO2 flux accurately. In this study, our team sampled 25 estuaries based on field surveys covering four seasons in Taiwan, aiming to better quantify the estimation of CO2 flux in the coastal regions. The dissociation constants of carbonic acid are unavailable to calculate the fCO2 in the low salinity (S<1). Therefore, the difference (%) between measured and calculated is very large but will be reduced with increasing salinity. Furthermore, in the process of measuring total alkalinity and pH, accuracy may reduce because of humic acid and variations of ionic strength. No matter in the western or eastern estuaries, most of fCO2 values is higher than the atmosphere. And they decrease downstream with increasing salinity. The fCO2 is higher in the west than in the east, because of human activities. Neither group of estuaries shows obvious seasonal variability. The fCO2 in the estuaries has a relationship with salinity, because of mixing with sea water. Because fCO2 is controlled by biological activity, it also has a relationship with AOU, pH and nutrients (NO3- and PO43-). In the east, the fCO2 has no correlation with many parameters. It is probably that the slope is steeper and the river length is shorter in the east than in the west resulting in short resident time. So that many reactions are not complete before the water exports to the sea. The average water-to-air CO2 flux is 24.6¡Ó19.2 (mol C m-2 y-1), which is 3.5 times the value of Pearl River (6.9 mol C m-2 y-1) but similar to the world average (23.7¡Ó33.1 mol C m-2 y-1). The CO2 flux is the highest in spring (81.7¡Ó15.8 mmol C m-2 d-1) and the lowest in winter (54.1¡Ó132 mmol C m-2 d-1). Upper/mid/lower estuaries are operationally defined as those areas of estuaries with salinities below 2, between 2 and 25, and above 25, respectively. The trends of fCO2 have good relationships with AOU and PO43- in the upper estuaries. The reason is probably caused by human activities and biological respiration. The phenomenon is more complex in the mid than in the upper estuaries. Consequently, the fCO2 has a good correlation with pH and DIC in the mid estuaries as a result of organic matter decomposition. However, in the lower estuaries, the variation of fCO2 is subjected to biological respiration and mixing with sea water. The fCO2 is the highest in the upper estuaries (2228¡Ó92.0 uatm)¡Athe average water-air CO2 flux is 42.3¡Ó1.54 (mol C m-2 y-1). Measured fCO2 in the mid estuaries is 1302¡Ó353 (uatm) and the average CO2 flux is 25.8¡Ó1.26 (mol C m-2 y-1). The lowest fCO2 (559¡Ó14.9 uatm) is found in the lower estuaries and the CO2 flux is 7.38¡Ó7.45 (mol C m-2 y-1). The 106 estuaries of the globe are divided into three parts by salinity. The fCO2 is 3033¡Ó1078, 2277¡Ó626 and 692¡Ó178 uatm in the upper, mid and lower estuaries, respectively. The average CO2 flux is 68.5¡Ó25.6¡B37.4¡Ó16.5 and 9.92¡Ó15.2 mol C m-2 y-1, respectively. Geographically estuaries in all three latitude bands (¡Õ23.5o, 23.5-50o and ¡Ö50o) are generally sources of CO2. Interestingly, water-to-air fluxes do not significantly, and all fall around 24 mol C m-2 y-1 although the flux is slightly lower at high latitude. The water in estuaries release CO2 in all seasons although the flux seems to be highest in autumn (73.2¡Ó 93.4 mmol C m-2 d-1) and lowest (53.4¡Ó65.1 mmol C m-2 d-1) in winter. The average CO2 flux is 23.9¡Ó33.1 mol C m-2 y-1, and the total CO2 flux is 0.26 Pg C y-1. Next, we estimate the tropical rivers¡¦ carbon fluxes using carbon parameters concerning 175 rivers globally between 30oN and 30oS. The specific DIC yield (flux/area) are 0.63, 3.33, 9.79 and 3.38 g C m-2 y-1 in tropical Africa, the Americas, Asia and Oceania, respectively. The DIC flux in Asia is the highest among the four regions, mainly because the percentage of carbonate rock is highest there and the second highest water discharge there. The PIC fluxes are 7.40¡Ñ1012 g C y-1 in Africa, 2.82¡Ñ1013 g C y-1 in the Americas, 1.53¡Ñ1013 g C y-1 in Asia and 2.49¡Ñ1011 g C y-1 in Oceania. The DOC fluxes are 2.80¡Ñ1013, 5.82¡Ñ1013, 4.50¡Ñ1013 and 4.48¡Ñ1012 g C y-1 in tropical Africa, the Americas, Asia and Oceania, respectively, for a total DOC flux of 0.136 Pg C y-1. Tropical rivers provide 0.53 Pg C y-1 of carbon to the oceans, of which 39.8¢H is DIC, 25.7¢H is DOC, 9.7¢H is PIC and 24.8¢H is POC.

We analyze output from a domain-filling forward trajectory model in order to better understand the annual cycle of water vapor entering the stratosphere. To do this, we determine the minimum water vapor saturation mixing ratio along each trajectory (the final dehydration point or FDP) and assume that the parcel carries that much water vapor into the stratosphere. In the annual average, the tropical Western Pacific, equatorial Africa and South America, and Southeast Asia are found to be the locations of the most frequent FDPs. Looking at individual seasons, we find that FDPs in the tropical western Pacific tend to occur in the summer hemisphere, with FDPs over South America and Africa occurring predominantly during the boreal winter. During boreal summer, a dehydration maximum occurs in the Asian monsoon region. In the annual average, FDP maxima occur at 99 and 84 hPa. Looking at individual seasons, we find that FDPs occur at higher altitudes (centered at 84 hPa) during boreal winter and at lower altitudes (99 hPa) during boreal summer. The annual cycle in FDP altitude combines with the annual cycle in tropical tropopause layer temperatures to generate the observed annual variations in water vapor entering the stratosphere.

Soluble organic matter represent less than 1% of total soil organic matter (SOM) - but it contributes to many terrestrial ecosystem processes, due to its high mobility and reactivity in soil. Although it has been suggested that soluble organic matter (OM) may serve as an early indicator of soil quality changes as a result of shifts in land-use and management practices, only a few studies have addressed the dynamics of soluble OM in relation to land-use and specifically soil depth. This study focuses on two aspects of soluble OM. In the first part, I hypothesized that extractable OM obtained by aqueous solutions is a continuum of substances that depending on the extraction method can be separated into two operationally different fractions. The size and properties of these fractions may consistently differ among different land uses and at different soil depths. The objective of this part of the study was then to assess dynamics (size and properties, biodegradability and seasonality) of water extractable organic matter (WEOM) and salt extractable organic matter (SEOM) in a sequence of human dominated land-uses at topsoil and subsoil. At the second part of the study, I tested the regulatory gate hypothesis –abiotic solubilization of OM- as a primary controlling factor in soluble OM production. The objective of this study was to evaluate the impact of the microbial activity on the net production of dissolved organic matter (DOM) from the native SOM in the presence of added DOM and plant residue. For the first part of the experiment, the soil samples were collected from four land-uses under bog pine (Halocarpus bidwillii) woodland, tussock grassland (Festuca novae-zelandiae and Heiracium pilosella), cropland (Medicago sativa) and plantation forest (Pinus nigra). The selected land uses were located in the Mackenzie Basin, Canterbury, New Zealand and occurring on the same soils, topography and experienced similar climates. Soil samples were obtained from topsoil (0-20 cm) and subsoil (60-80 cm) at the end of each season (November, February, May and August) during 2007-2008. The sampled soils were adjusted to the same water status prior to extraction. While WEOM was obtained during a mild extraction procedure and using 0.01M CaCl2, SEOM was extracted with 0.5M K2SO4 at high temperature (75οC for 90 min). Both extracts were filtered through a 0.45 μm filter size. In the first part of the study, I assessed the biodegradation dynamics of WEOM and SEOM (spring samples), using a double-exponential decay model. The WEOM and SEOM were inoculated and incubated at 22°C for 90d under aerobic conditions. Subsamples were removed on days 1, 3, 7, 12, 16, 30, 42, 60, 75, and 90, filtered (0.22 μm), and analyzed for organic C and N content, UV absorption, and 13C natural abundance (δ13C). The results of the biodegradation experiment indicated a similar pattern for both C and N of SEOM and WEOM as that of previously shown for soil DOM. However, C and N mineralization rate were considerably larger in the WEOM than SEOM. The parameters of the double-exponential model suggested that regardless of the land-use and soil depth, both the WEOM and SEOM can be modeled in two biological pools, with a largely similar “fast decomposable” but different “slowly decomposable” pools. However, since the extraction was not sequentially followed, a very small portion of the SEOM was comprised of the WEOM and given the greater observed biodegradability of the WEOM, the overall biodegradable portion of the SEOM would be lower than the observed. Despite a greater biodegradability of the organic N than C of both WEOM and SEOM; mainly due to a longer HL of the slowly biodegradable pool of C; the C/N ratio of the samples did not change very much during the biodegradation. This led us to conclude that the biodegradation of soluble OM may occur as a function of N availability. Parallel to C and N loss, a considerable increase in SUVA254 of SEOM, and particularly WEOM occurred during the incubation period. The greater increase in the proportion of aromatic compounds (assessed by SUVA) in the WEOM than SEOM, implied consumption of simple compounds (vs. very humified) during decomposition and further supported the observed faster biodegradation rate of the WEOM. The data indicated a relatively strong correlation (R2=0.66 and 0.74 for the WEOM and SEOM, respectively) between the amount of biodegraded C and the increase in SUVA254. This suggested that SUVA254 can be used as a simple, low-cost but reliable approach for describing the biodegradability of soluble OM, as previously suggested by others. At the end of the bioassay, the 13C natural abundance of the WEOM was significantly depleted, and showed a clear relationship with the proportion of the biodegraded C. This confirmed the previously suggested preferential biodegradation of simple organic constituents (13C enriched), resulting in the accumulation of more depleted 13C compounds (often recalcitrant compounds). Moreover, the results of the δ13C technique revealed that the relatively greater 13C enrichment of the WEOM obtained from subsoil, seems to be due to the presence of root exudates (often highly 13C enriched). In contrast, a proportionally greater 13C depletion observed in the SEOM particularly at subsoil samples, suggests that there is a close relationship between the SEOM and the typically 13C depleted humified SOM. The results of the biodegradation model (half-life of both C and N), in addition to dynamics of SUVA254 and δ13C of the WEOM and SEOM were very comparable between top and subsoil samples. This implied that the potential biodegradability of soluble OM under laboratory conditions does not necessary reflect the reported lower in situ biodegradability at soil depth, in agreement with recent evidence suggested by others. Instead, this may be largely due to the lack of optimum conditions (oxygen, nutrients, and moisture) for the decomposer community at soil depth. Although there was a tendency for a generally greater biodegradability of the samples from the soils under the crop land (both WEOC and SEOC), along with relatively greater increase in SUVA, there was not a consistent trend of the effect of land use on the biodegradation of either WEOM or SEOM. The lower C/N ratio of the soils under the crop land seemed to be related with the observed proportionally greater biodegradability of these soils. During the second part of the study, I assessed seasonal variations of the size and properties of the previously defined WEOM and SEOM, collected from top-and subsoil from the land-uses. I observed that 10-year after conversion of the degraded tussock grassland to cropland or plantation, the total C stock of topsoil (0-20 cm) when above- and below-ground plant biomass is excluded; has remained unchanged. This was attributed to the limited biomass production of the region, more likely as a result of low productivity of the soil, but also harsh climatic conditions. Not only soil depth, but land-use affected both C concentration and C/N ratio of soil organic matter (SOM), with the greatest C concentration of soils under grassland and plantation in topsoil and subsoil, respectively. Despite the WEOM, the size of SEOM was largely unaffected by land-use and soil depth; instead, the properties of SEOM was more consistent with the effect of soil depth. Given the observed large temporal and spatial variability of the WEOM, the study suggests that the SEOM more consistently reflects the influence of land use and soil depth. No consistent effect of seasonality was observed in terms of size or properties of the SOM and the WEOM and SEOM. Overall comparison of the size and properties of the WEOM and SEOM indicated that OM extraction efficiency may vary largely, depending on extraction conditions. Using more concentrated salt solutions consistently yielded greater amount of OM (N, and especially C) release from soil with properties resembling more those of total soil OM (more humified) compared to the WEOM. The SEOM was also less variable by time and space. The last part of the study was aimed to assess biotic vs. non-biotic solubilization of OM in the presence of added plant residue. Given the need to recognize the source of the solubilized OM during the experiment, I used enriched (13C) plant residue as the source of fresh OM. The above-ground part of ryegrass was added to soil either as plant residue or residue extract (extracted with CaCl2 followed by 0.45μm filtration) -termed DOM. These two forms of added OM (residue/DOM) were conceived to represent two levels of bioavailability for the decomposer community for further assessing possible biotic solubilization of OM. Two soils similar in their OM content and other properties, but different in mineralogy were selected for the experiment. Soils were incubated for 90d under sterilized vs. non-sterile conditions and leached regularly with a dilute aqueous solution (0.05M CaCl2). Plant residue was added to soil (1:100, residue: soil, w/w) prior to the start of the incubation, but DOM was frequently applied to the soils along with each leaching experiment. The greater C and N concentration in the leachates of both sterilized residue-amended and DOM-amended soils compared to that of living soils, indicated a high microbial activity, as determined by CO2 loss, in the living soils. However, the proportion of the solubilized C (determined by 13C) from sterilized soils was largely comparable to that of living soils. This supports the recently suggested “regulatory gate” hypothesis, stating that solubilisation of OM largely occurs independent of the size or community structure of microorganisms. In addition, I observed that even with the presence of adequate amount of added fresh OM (ryegrass residue), about 70% of the solubilized C consistently originated from the humified soil OM, highlighting the role of native soil OM as the source of soluble OM in soil. In addition, in the DOM-amended soils, there was strong evidence, indicating that in the sterilized soils, the added DOM was exchanged with the humified soil OM as observed by an increase in SUVA, and humification index (HI) of the leached OM. Although the results of the study did not show a considerable difference in the solubilisation rate of added OM as a function of biological activity (either in the residue- or DOM-amended soils), there was clear evidence that the presence of microbial activity has resulted in further decomposition of the solubilised OM through biological transformations. Together, the results suggested that the proposed fractionation method can be used to separate two operationally defined pools of soluble OM with consistent differences in their size (C and N), properties (δ13C, SUVA254, and C/N ratio) and biodegradability across the land-uses and soil depth. The second part of the study supported the primary role of abiotic factors on the production of soluble OM from native soil OM. Although the abiotic mechanisms involved in the solubilization remain to be addressed by future studies. Cons and pros of the methods with some suggestions for further works have been mentioned in the last chapter.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

Apple scab (causal organism: Venturia inaequalis) is an economically devastating disease of apples that is predominantly controlled with fungicides. Of the chemical classes currently available, the sterol-inhibiting (SI) and strobilurin (QoI) fungicides are the most commonly used. Recent observations indicate that V. inaequalis populations in Virginia have developed resistance to myclobutanil and other SIs. However, little is known about the frequency and distribution of SI and QoI resistance in Virginia's scab populations. The first objective of this research was to evaluate V. inaequalis populations in Virginia for SI and QoI resistance. Fungal isolates were collected from experimental orchards at the Alson H. Smith Jr., Agricultural Research and Extension Center (AHS AREC) and from commercial orchards in Virginia and Maryland. Sensitivities were determined by assessing colony growth at 19°C on potato dextrose agar (PDA) amended with 0 or 1.0 µg ml-1 of myclobutanil (SI) (N=87) or trifloxystrobin (QoI) (N=25) at 28 days. A range of fungicide sensitivity was observed for both chemical classes. The second objective of this research was to monitor the temporal dynamics of SI resistance over five sequential field seasons. To monitor shoot growth, neon rubber bands were placed over actively growing shoot tips following myclobutanil application or sample collection. Fungal isolates were collected from the same trees from 2007 through 2010 (N=176) and compared with isolates collected from wild apple seedlings (N=3). A continuum of SI resistance was observed for each year, and the V. inaequalis population exhibited a baseline shifted toward reduced sensitivity. The third objective of this research was to examine the spatial distribution of SI fungicide resistance within the tree canopy in a lower-density orchard (less than 150 trees A-1). Leaves collected from larger trees (>8m) in a lower-density orchard at the AHS AREC were analyzed for manganese deposition, pre- and post-mancozeb application. Fungal isolates (N=105) were collected from several locations within the canopy in replicated trees in the same orchard. Weather sensors also monitored the microclimates within those tree canopies. Spray deposition, microclimate and SI resistance were influenced by canopy location. The fourth objective of this research was to investigate potential SI resistance mechanisms. Previously classified isolates were screened for point mutations within the CYP51A1 gene (Appendix C), differences in polymorphic bands (alleles) (Appendix D), and differences in metabolism of myclobutanil (Appendix E). The consensus sequences for the CYP51A1 gene were identical for all isolates tested (N=9), and results from amplified fragment length polymorphism experiment (N=82) were inconclusive. There were, however, significant differences among incubation time and myclobutanil concentration in the bioassay (N=11). Our results indicate that myclobutanil is still an effective compound for control of apple scab in many areas of Virginia. / Ph. D.

京都大学 / 0048 / 新制・課程博士 / 博士(農学) / 甲第19775号 / 農博第2171号 / 新制||農||1041(附属図書館) / 学位論文||H28||N4991(農学部図書室) / 32811 / 京都大学大学院農学研究科森林科学専攻 / (主査)教授 神﨑 護, 教授 柴田 昌三, 教授 大澤 晃, 外国人教師 鄭 克聲 / 学位規則第4条第1項該当 / Doctor of Agricultural Science / Kyoto University / DGAM

Le phénomène El Niño est le mode dominant de la variabilité du climat aux échelles de temps interannuelles dans le Pacifique tropical. Il modifie considérablement le climat régional dans les pays voisins, dont le Pérou pour lequel les impacts socio-économiques peuvent être dramatiques. Comprendre et prévoir El Niño reste un enjeu prioritaire pour la communauté climatique. Des progrès significatifs dans notre compréhension du phénomène El Niño et dans notre capacité à le prédire ont été réalisés dans les années 80, en particulier grâce à la mise en place du système d'observation dans le Pacifique tropical (programme de TOGA, en particulier, ainsi que l'émergence de l'ère des satellites). À la fin du XXe siècle, alors que de nouvelles théories scientifiques ont été proposées et testées, les progrès réalisés dans le domaine de la modélisation numérique et de l'assimilation de données ont conduit à l'idée que le phénomène El Niño pourrait être prévu avec au moins deux ou trois saisons à l'avance. Or, depuis le début du 21ième siècle, les manifestations du phénomène El Niño ont réduit cette expectative: un nouveau type d'El Niño est a été découvert - identifié par des anomalies de température moins intenses et localisées dans le centre du Pacifique équatorial. Ce phénomène, connu sous le nom CP El Niño pour El Niño Pacifique Central ou El Niño Modoki a placé la communauté scientifique devant un nouveau défi. Cette thèse est une contribution à l'effort international actuel pour comprendre la dynamique de ce nouveau type d'El Niño, dans le but de proposer des mécanismes expliquant sa présence accrue au cours des dernières décennies. Plus précisément, l'objectif de cette thèse est d'étudier le rôle des ondes longues équatoriales dans le Pacifique tropical sur la dynamique océanique et la thermodynamique associées au phénomène El Niño de type Pacifique Central. Cette thèse s'intéresse tout d'abord au premier CP El Niño du 21ième siècle, le phénomène El Niño 2002/03, à partir des sorties d'un modèle de circulation océanique général. Ensuite, nous documentons les caractéristiques des ondes équatoriales de Kelvin aux fréquences Intra Saisonnières (ISKw) sur la période 1990-2011, fournissant une statistique de l'activité des ondes ISKw durant l'évolution des événements El Niño de type Central Pacifique. Nos résultats montrent que l'onde ISKw subit une forte dissipation dans le Pacifique Est, qui est interprétée comme provenant de la dispersion des ondes lorsqu'elles rencontrent le front zonal de la stratification dans l'Est du Pacifique (i.e. la pente de la thermocline d'Ouest en Est). Une réflexion partielle de l'onde ISKw en onde de Rossby équatoriale de près de 120°W est également identifiée, ce qui peut expliquer le confinement dans le Pacifique central des anomalies de température de surface associées aux événements El Niño de type Central Pacifique. Nous suggérons que la fréquence accrue au cours des dernières années des événements CP El Niño peut être associée à l'état froid - de type La Niña - observé dans le Pacifique Equatorial depuis les années 90 et les changements dans la variabilité saisonnière de la profondeur de la thermocline depuis les années 2000. / The El Niño phenomenon is the dominant mode of climate variability at interannual timescales in the tropical Pacific. It modifies drastically the regional climate in surrounding countries, including Peru for which the socio-economical impacts can be dramatic. Understanding and predicting El Niño remains a top-priority issue for the climatic community. Large progress in our understanding of El Niño and in our ability to predict it has been made since the 80s thanks to the improvement of the observing system of the tropical Pacific (TOGA program and emergence of the satellite era). At the end of the Twentieth century, whereas new theories were proposed and tested, progress in numerical modeling and data assimilation led to the idea that El Niño could be predicted with at least 2 or 3 seasons in advance. The observations since the beginning of the 21st century have wiped out such expectation: A new type of El Niño, known as the Central Pacific El Niño (CP El Niño) or Modoki El Niño has put the community in front of a new challenge. This thesis is a contribution to the current international effort to understand the dynamics of this new type of El Niño in order to propose mechanisms explaining its increased occurrence in recent decades. More specifically, the objective of the thesis is to study the role of the oceanic equatorial waves in the dynamic and thermodynamic along the equatorial Pacific Ocean, focusing on the CP El Niño. This thesis first takes a close look at the first CP El Niño of the 21st century of this type, i.e. the 2002/03 El Niño, based on an Oceanic General Circulation Model. Then it documents the characteristics of the IntraSeasonal Kelvin waves (ISKws) over the period 1990-2011, providing a statistics on the ISKws activity during the evolution of CP El Niño events. We find that the ISKw experiences a sharp dissipation in the eastern Pacific that is interpreted as resulting from the scattering of energy associated to the zonal contrast in stratification (i.e. sloping thermocline from west to east). Partial reflection of the ISKw as Rossby waves near 120°W is also identified, which may explain the confinement of CP El Niño warming in the central Pacific. We suggest that the increased occurrence of CP El Niño in recent years may be associated to the La Niña-like state since the 90s and changes in the seasonality of the thermocline since the 2000s.

<p>The general perception today is that the Shashi-Limpopo Basin in southern Africa is hot and dry and not conducive to human habitation. Today there is no doubt that the Shashi-Limpopo Basin has been home to many communities throughout the pre-historical period. A study of the changing ecological conditions in the Mateke Hills and the Shashi-Limpopo Valley as well as historical and present day land-usage offers an alternative explanation of how prehistoric communities could have interacted with this changing landscape. The archaeological record, historical sources and recent land-use patterns show that settlement location has always been orientated towards the rivers and circumscribed environments. The mosaic of floodplains, wetlands, drylands and circumscribed zones provided the ideal ecological setting for the development of socio-political complexity in southern Africa. The resilience of these semi arid savanna regions together with human innovation and local knowledge ensured that societies continued to derive subsistence even in the face of seasonal variability in rainfall and even climate change.</p>

The general perception today is that the Shashi-Limpopo Basin in southern Africa is hot and dry and not conducive to human habitation. Today there is no doubt that the Shashi-Limpopo Basin has been home to many communities throughout the pre-historical period. A study of the changing ecological conditions in the Mateke Hills and the Shashi-Limpopo Valley as well as historical and present day land-usage offers an alternative explanation of how prehistoric communities could have interacted with this changing landscape. The archaeological record, historical sources and recent land-use patterns show that settlement location has always been orientated towards the rivers and circumscribed environments. The mosaic of floodplains, wetlands, drylands and circumscribed zones provided the ideal ecological setting for the development of socio-political complexity in southern Africa. The resilience of these semi arid savanna regions together with human innovation and local knowledge ensured that societies continued to derive subsistence even in the face of seasonal variability in rainfall and even climate change.