Tesis sobre el tema "Decadal prediction"

This study assesses the monthly precipitation of CMIP5 decadal experiment over Brisbane River catchment for a spatial resolution of 0.050 and then predicts the monthly precipitation for decadal timescale through a Bidirectional LSTM and Machine Learning Algorithms using GCMs and observed data. To use GCM data in this future prediction, investigations were carried out for a suitable spatial interpolation method, a better simulation period, model drifts, and drift correction alternatives based on different skill tests.

En Afrique de l’Ouest, les précipitations durant la période d’été ont connu, au cours du XXe siècle, de fortes modulations aux échelles de temps décennales. On note en particulier une période très humide vers les années 60 suivie d'un épisode de sécheresse durant les années 80 et une reprise des pluies au cours des années 2000. Ces modulations ont des conséquences socio-économiques régionales majeures. Considérées comme l’un des plus grands signaux du changement climatique en Afrique, elles ont fait l'objet de nombreuses études afin de comprendre leur(s) origine(s) et d’anticiper les prochains changements. Ces modulations ont longtemps été associées à la variabilité interne du système climatique et notamment à un rôle de la température de surface de l’océan Atlantique. De récentes études montrent cependant une contribution notable des forçages externes notamment les gaz à effet de serre et les aérosols anthropiques en particulier au cours de la seconde moitié du XXe siècle, soit via les températures de surface de l’océan Atlantique soit via un forçage radiatif direct. Dans tous les cas, le lien avec les températures de surface de l’océan offrent des perspectives de prévisibilité de ces modulations des précipitations. Les prévisions climatiques décennales ont été développées afin d’exploiter cette source de prévisibilité. Elles sont potentiellement très importantes pour la planification économique et structurelle en Afrique de l’Ouest. Dans cette thèse, nous proposons d’attribuer, en premier lieu, les modulations décennales des précipitations au Sahel sur la période temporelle de 1901-2014. En second lieu, nous proposons d’évaluer la prévision des précipitations en Afrique de l’Ouest à l’échelle de temps décennale sur la période 1968-2012. Toute cette étude est réalisée par le diagnostic de simulations climatiques réalisées à partir de modèles contribuant à la phase 6 du projet d’intercomparaison des modèles couplés. Étant donné que les modèles couplés sous-estiment communément le maximum de précipitation au Sahel durant la période d’été, nous avons commencé par proposer un domaine adaptatif des pluies au Sahel. Nos résultats de la première partie de cette thèse montrent que les forçages externes en particulier les aérosols anthropiques contribuent significativement à la chronologie des modulations décennales des précipitations sahéliennes. Ces aérosols modulent en effet la température de surface de l'océan qui se traduit en modulations de précipitation au Sahel par les déplacements de la zone de convergence intertropicale et de la dépression thermique saharienne. La seconde partie de cette thèse s'intéresse à la prévisibilité des précipitations à l’échelle de temps décennale en Afrique de l’Ouest et en particulier au Sahel. Nos résultats montrent que les précipitations sont prévisibles au Sahel dans 5⁄8 des modèles étudiés à une échéance temporelle entre 1 et 10 ans. Cependant, seuls 3 de ces 5 modèles simulent un signal décennal combiné à une variance totale réaliste à ces échéances. La moyenne multi-modèle est généralement plus prévisible que la majorité des modèles individuels. L’initialisation des variables climatiques permet d’améliorer les scores de prévision des précipitations au Sahel pour la moyenne multimodèle et 3⁄5 des modèles montrant une prévisibilité et la fiabilité des prévisions pour un seul modèle. Cet impact de l’initialisation relativement réduit corrobore l’importance des forçages externes déduits de la première étude. En conclusion, nos résultats suggèrent que les effets anthropiques sur le climat deviennent des facteurs majeurs pour expliquer les modulations décennales des précipitations moyennes de mousson au Sahel. En perspective, ces modulations sont associées, au cours du XXe siècle, à des pluies extrêmes qui sont devenues fréquentes de nos jours. Il serait intéressant de se demander si ces pluies extrêmes sont prévisibles en Afrique de l’Ouest aux échelles de temps décennales
In West Africa, rainfall during the summer period (July, August and September) has shown strong modulations on decadal time scales during the 20th century. In particular, there was a very wet period in the 1960s followed by a drought in the 1980s and a recovery of rainfall in the 2000s. These modulations have major regional socio-economic consequences. Considered as one of the most important signals of climate change in Africa, they have been the subject of numerous studies in order to understand their origin(s) and to anticipate future changes. These modulations have long been associated with the internal variability of the climate system and in particular with the role of the surface temperature of the Atlantic Ocean. However, recent studies show a significant contribution of external forcings, notably greenhouse gases and anthropogenic aerosols, in particular during the second half of the 20th century, either via Atlantic Ocean surface temperatures or via direct radiative forcing. In all cases, the link with ocean surface temperatures offers prospects for predicting these precipitation modulations. Decadal climate forecasts, i.e. with a 5-30 year time horizon, have been developed to exploit this source of predictability. They are potentially very important for economic and structural planning in West Africa. In this thesis, we propose to attribute, first, the decadal modulations of Sahel rainfall over the time period of 1901-2014. Secondly, we propose to evaluate the prediction of rainfall in West Africa on the decadal time scale over the period 1968-2012. All this study is carried out through the diagnosis of climate simulations carried out from models contributing to the phase 6 of the coupled models intercomparison project. Since coupled models commonly underestimate the maximum rainfall in the Sahel region during the summer period, we first proposed an adaptive rainfall domain for the Sahel. Our results from the first part of this thesis show that external forcings, in particular anthropogenic aerosols, contribute significantly to the timing of decadal modulations of Sahel rainfall. These aerosols indeed modulate the ocean surface temperature which is translated into precipitation modulations in the Sahel by the displacement of the intertropical convergence zone and the Saharan Heat Low. The second part of this thesis focuses on the predictability of precipitation on decadal time scales in West Africa and in particular in the Sahel. Our results show that rainfall is predictable in the Sahel in 5⁄8 of the models studied at lead times between 1 and 10 years. However, only 3 of these 5 models simulate a decadal signal combined with a realistic total variance (signal-to-noise ratio close to 1) at these lead times. The multi-model mean is generally more predictable than the majority of individual models. Initialization of climate variables improves Sahel rainfall prediction skills for the multi-model mean and 3⁄5 of the models showing predictability and forecast reliability for a single model. This relatively small impact corroborates the importance of external forcings inferred from the first part of this thesis. In conclusion, our results suggest that anthropogenic effects on climate become major factors in explaining decadal modulations of mean monsoon rainfall in the Sahel. In perspective, these modulations are associated, during the 20th century, with extreme rainfall events that have become frequent nowadays. It would be interesting to ask whether these extreme events are predictable in West Africa on decadal time scales

Decadal climate predictions have the main feature of being initialized, hence lying midway between initialized seasonal forecasts and forced multi-decadal projections. The North Atlantic is among the few places where decadal variations are considered potentially predictable with an added value of the initialization due to the Atlantic Meridional Overturning Circulation (AMOC), which exhibits slow multi-annual fluctuations. A correct representation of this process is fundamental to skillfully predict climate variability in the Northern Hemisphere at these timescales. In this thesis, AMOC predictability is investigated in the CMCC-CM2-SR5 (CMCC Coupled Model v2 in standard resolution) decadal system. The ability of the model to forecast the AMOC is evaluated in both a deterministic and probabilistic way, comparing a set of hindcasts initialized between 1960 and 2018 with observations, ocean reconstructions, and a non-initialized historical simulation. Special attention is devoted to the analysis of AMOC biases. Indeed, it is documented that predictions suffer from initial shocks and tend to drift towards the model's equilibrium state. We find that the potential predictability of the system is high up to a ten-year forecast range, but this is not reflected in the AMOC transport forecast skill, which undergoes a sudden reduction after the first year. An interesting finding is that the drift of the model is start-date dependent: we leverage on this feature to propose a new post-processing approach for the drift adjustment, different from the usual one in which drifts are treated as stationary. The experimented approach significantly increases the forecast skill. Furthermore, we identify a reduction of convection in the Labrador Sea, a feature that previous studies linked with the model drift of the AMOC. Further research with an increased ensemble size of both initialized and historical simulations and with a multi-model set is envisaged.

Le remodelage ventriculaire gauche est une complication fréquente des patients ayantprésenté un syndrome coronarien aigu, pouvant conduire à terme à une situation d’insuffisancecardiaque. Il est donc important de connaître les facteurs associés à la survenue d’un remodelageventriculaire afin de dépister plus précocement les patients à plus haut risque d’insuffisance cardiaqueet ainsi optimiser leur prise en charge. Ce travail comprend deux axes. Le premier porte sur larecherche de nouveaux paramètres d’imagerie associés à la survenue du remodelage. Nous avonsdans un premier temps réalisé une revue de la littérature concernant la définition du remodelageventriculaire gauche en imagerie par résonance magnétique. Puis, nous avons conduit deux étudesayant pour but de rechercher une association entre (i) le strain atrial gauche et, (ii) le gradient depression intraventriculaire gauche diastolique, évalués en échocardiographie 24-48 heures après lesyndrome coronarien aigu et le remodelage ventriculaire gauche au cours du suivi. Le second axe portesur les biomarqueurs associés au remodelage ventriculaire post-infarctus. Nous avons réalisé une revuede la littérature au sujet des biomarqueurs qui, dosés lors de l’hospitalisation initiale, sont associés àl’existence d’un remodelage lors du suivi. Nous avons ensuite étudié la valeur prédictrice de deuxbiomarqueurs (la néprilysine et le coenzyme Q10) pour la survenue d’un remodelage ventriculairegauche
Left ventricular remodeling is a common complication in patients following acutemyocardial infarction and may lead to heart failure. Some baseline parameters are associated withremodeling at follow-up, allowing to better discriminate patients with an increased risk of heart failureto optimize therapeutics. This work has two axes, focused on imaging and biological parametersassociated with left ventricular remodeling, respectively. First, we reviewed past studies that definedremodeling using cardiac magnetic resonance imaging. Then, we studied the association betweensome echocardiographic parameters (left atrial strain and diastolic intraventricular pressure gradient)and left ventricular remodeling after ST-elevation myocardial infarction. In the other axis, wereviewed biomarkers that have been associated with left ventricular remodeling in prior studies. Then,we investigated the association between neprilysin and coenzyme Q10 levels and left ventricularremodeling in STEMI patients

碩士
國立中央大學
大氣科學學系
105
In this study, we use monthly data from the multi-model ensemble (MME) of Coupled Model Intercomparison Project Phase 5 (CMIP5) decadal prediction experiments to assess interannual prediction skills for several atmospheric and oceanic variables in Tropics (30°S-30°N). First, we applied pattern stability analyses to extract persistent empirical orthogonal functions (EOFs) from observations-based data as reference spatial patterns. By projecting CMIP5 MME predictions to the extracted EOFs, then we compared these associated time series to assess the MME prediction skills. Finally, we applied linear regression and rank histogram to calibrate the associated time series of MME predictions. In the meantime, this study also evaluates the grid-point scale prediction capability in Tropics by EOF reconstructed fields. Pattern stability analyses of the observations-based data indicated that at least 4 persistent EOFs can be found in each examined variable field. The first EOF (EOF1) mainly corresponds to the mean state of the given field, while the second EOF and beyond correspond to more and more localized spatial structures. Except for the third EOF (EOF3) of sea surface temperature (SST) field that has close relation to the El Nino Southern Oscillation (ENSO), most of our efforts focused on the study of interannual prediction skill associated with EOF1. Results indicated that, except for near surface air temperature (SAT) and SST fields, most variable fields did not have any interannual prediction skill. Furthermore, the apparent prediction skill that SAT and SST fields possessed may largely come from the warming trend observed in the last half of the 20th century. As for the ENSO related prediction skill, the EOF3 related time series showed certain prediction skill. This skill may be related to the capability of climate models to better synchronize with ENSO evolution through the adoption of yearly initialization procedure. Additionally, the results of the calibrated MME predicted time series showed that both linear regression and rank histogram calibration methods could effectively reduce the prediction errors and the MME uncertainty. Furthermore, the use of EOF reconstruction reduced MME prediction errors on extensive continent and coastal regions.

Introdução: Doentes previamente submetidos a craniectomia são geralmente submetidos ao procedimento de cranioplastia para restabelecer a estrutura do crânio, melhorar a proteção cerebral, a perfusão vascular e dinâmica do LCR. Apesar de serem frequentemente considerados como procedimentos minor, estudos de cranioplastias relatam frequentemente a ocorrência de complicações significativas. Este estudo visa identificar fatores preditivos de complicações pós-cirúrgicas da cranioplastia. Métodos: Realizámos um estudo retrospetivo unicêntrico num hospital terciário, analisando todos os pacientes submetidos a cranioplastia (CP) após craniectomia (CE) entre 2008 e 2019. A informação demográfica dos doentes, a história médica passada, as características da CE e da CP foram obtidas através dos registos hospitalares. "Complicações de CP" foram definidas como qualquer evento sintomático que tenha levado ou não a uma reintervenção cirúrgica, tais como: hemorragia intracerebral sintomática, hemorragia extradural ou subdural, hidrocefalia, infeção ou reabsorção óssea. Resultados: Foram incluídos trezentos e quatro doentes no estudo. Após revisão dos critérios de inclusão, 168 doentes foram considerados para análise (139 adultos, com uma idade média de 47.6±12.68 anos, e 29 pacientes pediátricos com uma idade média de 11.8±5.62 anos), com uma ligeira predominância de doentes do sexo masculino. A taxa de complicações total foi de 26,2%. Utilizando um modelo de regressão binomial, e controlando para idade e sexo, identificámos os seguintes fatores preditivos de complicação de CP: história de coagulopatia primária (risco 14,3 vezes superior, p=0,034); punção ventricular intra-operatória (risco 7,9 vezes superior, p=0,009) e relato de rutura da camada dural (risco 2,8 vezes superior, p=0,033). Estar a residir no domicílio na altura da cranioplastia apresentou-se como um fator protetor. Conclusão: Os procedimentos de cranioplastia merecem um enfoque contínuo na prevenção de complicações. A análise realizada demonstra que, para estes pacientes, as perturbações da coagulação, a necessidade de punção ventricular intra-operatória e a violação do limite dural intra-operatório durante o procedimento foram associadas a complicações.
Introduction: Patients previously submitted to craniectomy generally undergo cranioplasty procedures to reestablish cosmesis and benefit cerebral protection, perfusion and CSF dynamics. Frequently regarded as minor procedures, cranioplasty series repeatedly report major complication rates. This study aims to identify predictive factors of post-operative cranioplasty complications. Methods: We performed a retrospective single center study in a tertiary hospital, analyzing all patients submitted to cranioplasty (CP) following craniectomy (CE) between 2008 and 2019. Patients' demographic information, medical history, CE and CP characteristics were retrieved from hospital records. "CP Complications" were considered any symptomatic event that led or not to surgical re-intervention such as symptomatic intracerebral hemorrhage, extradural or subdural hemorrhage, hydrocephalus, infection or bone resorption. Results: Three-hundred and four patients were admitted to the study. After inclusion criteria were revised, 168 patients were considered for analysis (139 adult, with a mean age of 47.6±12.68 years-old, and 29 pediatric with a mean age of 11.8±5.62 years-old), with a slight male predominance. The total complication rate was 26.2%. Using a binomial regression model, and controlling for age and gender, we identified as predictive factors for CP complication: history of primary coagulopathy (14.3 fold risk increase, p=0,034); intra-operative ventricular puncture (7.9 fold risk increase, p=0,009) and reported breach of the dural layer (2.8 fold risk increase, p=0,033). Living at home before cranioplasty was identified as a protective factor. Conclusion: Cranioplasty procedures merit a continuous focus on complication avoidance. The analysis demonstrates that for these patients, coagulation disturbances, the need for intraoperative ventricular puncture and intraoperative dural limit violation during the procedure were associated with cranioplasty complications.

Introdução: Doentes previamente submetidos a craniectomia são geralmente submetidos ao procedimento de cranioplastia para restabelecer a estrutura do crânio, melhorar a proteção cerebral, a perfusão vascular e dinâmica do LCR. Apesar de serem frequentemente considerados como procedimentos minor, estudos de cranioplastias relatam frequentemente a ocorrência de complicações significativas. Este estudo visa identificar fatores preditivos de complicações pós-cirúrgicas da cranioplastia. Métodos: Realizámos um estudo retrospetivo unicêntrico num hospital terciário, analisando todos os pacientes submetidos a cranioplastia (CP) após craniectomia (CE) entre 2008 e 2019. A informação demográfica dos doentes, a história médica passada, as características da CE e da CP foram obtidas através dos registos hospitalares. "Complicações de CP" foram definidas como qualquer evento sintomático que tenha levado ou não a uma reintervenção cirúrgica, tais como: hemorragia intracerebral sintomática, hemorragia extradural ou subdural, hidrocefalia, infeção ou reabsorção óssea. Resultados: Foram incluídos trezentos e quatro doentes no estudo. Após revisão dos critérios de inclusão, 168 doentes foram considerados para análise (139 adultos, com uma idade média de 47.6±12.68 anos, e 29 pacientes pediátricos com uma idade média de 11.8±5.62 anos), com uma ligeira predominância de doentes do sexo masculino. A taxa de complicações total foi de 26,2%. Utilizando um modelo de regressão binomial, e controlando para idade e sexo, identificámos os seguintes fatores preditivos de complicação de CP: história de coagulopatia primária (risco 14,3 vezes superior, p=0,034); punção ventricular intra-operatória (risco 7,9 vezes superior, p=0,009) e relato de rutura da camada dural (risco 2,8 vezes superior, p=0,033). Estar a residir no domicílio na altura da cranioplastia apresentou-se como um fator protetor. Conclusão: Os procedimentos de cranioplastia merecem um enfoque contínuo na prevenção de complicações. A análise realizada demonstra que, para estes pacientes, as perturbações da coagulação, a necessidade de punção ventricular intra-operatória e a violação do limite dural intra-operatório durante o procedimento foram associadas a complicações.
Introduction: Patients previously submitted to craniectomy generally undergo cranioplasty procedures to reestablish cosmesis and benefit cerebral protection, perfusion and CSF dynamics. Frequently regarded as minor procedures, cranioplasty series repeatedly report major complication rates. This study aims to identify predictive factors of post-operative cranioplasty complications. Methods: We performed a retrospective single center study in a tertiary hospital, analyzing all patients submitted to cranioplasty (CP) following craniectomy (CE) between 2008 and 2019. Patients' demographic information, medical history, CE and CP characteristics were retrieved from hospital records. "CP Complications" were considered any symptomatic event that led or not to surgical re-intervention such as symptomatic intracerebral hemorrhage, extradural or subdural hemorrhage, hydrocephalus, infection or bone resorption. Results: Three-hundred and four patients were admitted to the study. After inclusion criteria were revised, 168 patients were considered for analysis (139 adult, with a mean age of 47.6±12.68 years-old, and 29 pediatric with a mean age of 11.8±5.62 years-old), with a slight male predominance. The total complication rate was 26.2%. Using a binomial regression model, and controlling for age and gender, we identified as predictive factors for CP complication: history of primary coagulopathy (14.3 fold risk increase, p=0,034); intra-operative ventricular puncture (7.9 fold risk increase, p=0,009) and reported breach of the dural layer (2.8 fold risk increase, p=0,033). Living at home before cranioplasty was identified as a protective factor. Conclusion: Cranioplasty procedures merit a continuous focus on complication avoidance. The analysis demonstrates that for these patients, coagulation disturbances, the need for intraoperative ventricular puncture and intraoperative dural limit violation during the procedure were associated with cranioplasty complications.