Dans le contexte du changement climatique, il est nécessaire d'affiner les informations relatives à l'évolution du climat dans un pays, susceptible d'être négativement impacté par le réchauffement global, comme le Maroc. En effet, les différentes études de projections futures, dont celles du GIEC, sont basées majoritairement sur les sorties de modèles climatiques à faible résolution qui ne permettent pas d'aborder les échelles régionales et locales. La première partie de ce travail concerne l'étude des tendances et évolutions observées au niveau du climat du Maroc à travers un certain nombre d'indices climatiques. Outre la forte variabilité interannuelle des précipitations et l'augmentation de la fréquence des sécheresses depuis le début des années 1980, la distribution des précipitations du Maroc a bien connu un changement au cours de la période 1961-2008. Ce changement, consistant en une évolution vers des conditions plus sèches, a coïncidé avec une augmentation de la température moyenne en toutes saisons. L'évaluation des changements futurs est réalisée tout d'abord à l'aide d'une descente d'échelle dynamique effectuée avec le modèle ARPEGE-Climat dans sa version à résolution variable. L'examen des capacités du modèle, dont la résolution est de l'ordre de 50km sur le Maroc, a montré son aptitude à simuler correctement la circulation de grande échelle ainsi que la variabilité interannuelle des précipitations marocaines en dépit d'une sous-estimation de leur quantité. A l'horizon 2021-2050, une baisse des cumuls généralisée à tout le pays concernerait la saison d'hiver (DJF). Si on se limite à la zone située à l'ouest des montagnes de l'Atlas, la baisse concernerait la partie la plus pluvieuse de l'année (ONDJFM). Cette baisse serait accompagnée d'une diminution du nombre de jours humides et du nombre d'événements de fortes précipitations ainsi que d'une augmentation de la persistance temporelle de la sécheresse. Ce changement de la distribution des précipitations coïnciderait avec un réchauffement qui se manifesterait à la fois aux échelles saisonnière et annuelle. Les sorties d'une dizaine de modèles régionaux de climat (MRC) du projet FP6-ENSEMBLES, sont utilisées pour balayer une partie de la marge des incertitudes relatives aux changements climatiques et notamment celles inhérentes à la modélisation. Les changements futurs issus de ces MRC, compatibles dans l'ensemble avec ceux issus d'ARPEGE-Climat, optent dans la plupart des cas pour une réduction des cumuls pluviométriques de l'hiver accompagnée généralement d'une baisse du nombre d'événements de fortes précipitations et d'une augmentation du nombre maximal de jours consécutifs secs. L'examen, réalisé à la fois à l'aide du modèle ARPEGE-Climat et des MRC d'ENSEMBLES, de la possibilité d'utilisation d'une méthode de réduction d'échelle statistique basée sur les régimes de temps de l'Atlantique nord pour la déduction des changements futurs des précipitations locales, montre la déficience de ce type d'approche dans le cas de la pluviométrie hivernale marocaine. La méthode de correction par quantiles étendues aux régimes de temps de l'Atlantique nord et appliquée aux sorties d'ARPEGE-Climat confirme le signe des changements issus de ce modèles, malgré un effet légèrement modérateur de leurs amplitudes. L'évaluation des impacts potentiels sur l'hydrologie à l'aide du modèle hydrologique GR2M et des scénarios climatiques d'ARPEGE-Climat, montre une future réduction des débits mensuels au niveau du bassin versant de la Moulouya du fait de la concomitance de précipitations moins abondantes et d'une évapotranspiration potentielle accrue par l'augmentation de température. Enfin, une descente d'échelle dynamique réalisée à l'aide du modèle à aire limitée ALADIN-Climat à très haute résolution (12km) sur la moitié nord du pays permet de confirmer dans l'ensemble les projections issues d'ARPEGE-Climat à la fois en termes de moyennes et d'extrêmes. / In the context of climate change, it is important to improve climate information concerning countries that may be negatively impacted by global warming such as Morocco. Indeed, various studies of future projections, including IPCC ones, are mainly based on the outputs of low resolution climate models that do not allow accessing the regional and local scales. The first part of this work focuses on the study and analysis of observed climate evolution and trends in Morocco through a set of climate indices. Moroccan rainfall is characterized by a high interannual variability and more frequent droughts have occurred since the early 1980s. Furthermore, a clear change is shown in the distribution of precipitation during the period 1961-2008. It consists in a shift towards warmer and drier conditions. The assessment of future climate changes is done, firstly, using a variable resolution version of the global GCM ARPEGE-Climat with high resolution over Morocco (50km). The examination of this version capability shows the ability of the model to well reproduce the large scale circulation as well as the interannual variability of Moroccan rainfall despite an underestimation of its amount. A reduction of winter rainfall over the whole country is projected by the model for 2021-2050. In the region located west of the Atlas Mountains, the reduction could concern the wettest part of the year (ONDJFM). The changes in rainfall characteristics may also occur through a decrease in the number of wet days and the number of heavy precipitation events and by more persistent droughts. Furthermore, an increase of mean temperature is projected at annual and seasonal scales. The outputs of ten RCMs of the FP6-ENSEMBLES (ENSEMBLES) project are used to assess the uncertainties associated to future climate change. The changes issued from ARPEGE-Climat are in the range covered by the ten RCMs. Most of the models agreed on a reduction of winter precipitation associated with a decrease in the number of heavy precipitation events and an increase in the number of maximum consecutive dry days. The evaluation of a statistical downscaling approach that uses large scale fields such as North Atlantic weather regimes to construct local scenarios of future climate change shows the deficiency of this approach in the case of Moroccan winter precipitation. This result is obtained by both ARPEGE-Climat and the ENSEMBLES RCMs. The quantile-quantile correction method extended to weather regimes and applied to the outputs of ARPEGE-Climat confirms the sign of the changes despite a slight reduction of their amplitudes. The assessment of potential impacts on hydrology done using the hydrological model GR2M and the climate scenarios issued from ARPEGE-Climat shows a future reduction of the Moulouya watershed discharges. This is due to the combination of a rainfall decrease and an enhanced potential evapotranspiration induced by increasing temperature. Finally, a dynamical downscaling achieved using the limited area model ALADINClimat with very high resolution (12km) on the northern half of the country allows a further assessment of future climate changes and related uncertainties. The projections issued from ARPEGE-Climat are generally confirmed both in terms of average and of extremes

No description available.

Les précipitations sont un phénomène dont la variabilité s'étend sur une très large gamme d'échelles : de l'échelle millimétrique de la goutte d'eau (échelle microphysique) à l'échelle des circulations atmosphériques globales (échelle synoptique). Il n'existe pas de système unique capable de fournir des observations des précipitations couvrant toutes ces échelles. Les observations satellite sont celles qui actuellement résolvent le plus efficacement les grandes échelles spatiales et temporelles : de la méso-échelle à l'échelle synoptique. Dans cette thèse, nous explorons en zone tropicale les capacités des satellites à résoudre les échelles spatiales de l'ordre de 100km, jusqu'aux échelles kilométriques ; et les échelles temporelles comprises entre 24 heures et 15 minutes (afin de résoudre le cycle diurne). L'approche retenue est physico-statistique. Si les grandes échelles peuvent être résolues par des approches déterministes combinant les mesures de multiples instruments spatiaux, plusieurs facteurs limitent la pertinence des approches déterministes à fine échelle : - Les limites instrumentales en terme de résolution spatiale. - Le nombre d'instruments en orbite qui limite la fréquence d'échantillonnage des mesures. - La nature dynamique de la variabilité fine échelle. En particulier, aux fines échelles, c'est la difficulté à parfaitement localiser les structures précipitantes qui entraine les erreurs d'estimation les plus importantes. L'approche physico-statistique est ici synonyme de déterministe (pour les grandes échelles) et probabiliste (pour les fines échelles). Le premier objectif de cette thèse est de déterminer précisément la limite des échelles qui peuvent être résolues de façon déterministe. L'approche physico-statistique de l'estimation des intensités de précipitation est implémentée dans cette thèse à partir d'une méthode multicapteur déterministe pré-existante : l'algorithme TAPEER, développé dans le cadre de la mission Megha-Tropiques, qui fournit une estimation du cumul pluviométrique journalier à une résolution de 1°. C'est la génération d'ensembles désagrégés par une méthode stochastique multi-échelle qui a été retenue ici. Les ensembles sont contraints par une information fine échelle : un masque de détection des aires précipitantes dérivé des images infrarouge metosat-SG à une résolution de 3km (et avec une image toutes les 15 minutes). La génération d'ensemble permet de caractériser l'incertitude sur l'estimation à travers la dispersion des réalisations de l'ensemble. Chaque réalisation de l'ensemble est générée de façon à reproduire le plus fidèlement possible les propriétés statistiques (distribution de fréquence des intensités, autocorrélation spatiale et temporelle) des véritables champs de précipitation. Ces champs et cette technique ont un apport pour les applications hydrologiques, par exemple pour améliorer le ruissellement lié aux précipitations intenses dans les modèles. Considérer plusieurs réalisations permet de plus d'étudier la propagation des incertitudes à travers un modèle. / Rainfall variability involves a wide range of scales: from the millimeter-scale associated with microphysics to the synoptic scale of the global atmospheric circulation. No existing observation system is able to cover all these scales by itself. Satellite-based observation systems are currently the most efficient systems to resolve the large spatial and temporal scales: from mesoscale meteorology to the synoptic scale. This thesis is dedicated to the exploration of satellites ability to resolve spatial scales from 100km to 2km and temporal scales from 24h to 15 min (in order to resolve the diurnal cycle). The chosen approach is both physical and statistical (or deterministic and probabilistic). The idea is that the deterministic approach can resolve the large scales, but several factors limit its relevance when dealing with fine scales: -The limited resolution of the instruments. -The number of orbiting instruments that limits temporal sampling. -The dynamic nature of fine scale variability. At fines scales, most of the errors in rainfall estimation from satellite comes from not perfectly localizing the precipitating cells. The first objective of this thesis is to identify precisely the lowest limit in scale where the deterministic approach is appropriate. The implementation of the physical-statistical approach relies on an existing multisensor estimate of daily precipitation at a 1° resolution: the TAPEER algorithm developed as part of the Megha-Tropiques mission. The chosen method is a hybrid physical disaggregation and stochastic downscaling via a multiscale representation. The result is an ensemble of high-resolution probable realizations of the rain intensity field. The ensemble is constrained by a high resolution rain detection mask derived from meteosat-SG infrared images at 3km resolution (one image every 15 minutes). The uncertainty associated with the final estimation is handled through the ensemble dispersion. Every realization is generated so that its statistical properties (frequency distribution of the intensities, autocorrelation function) mimic those of the true rain field. The generated fields and the proposed technique contribute to hydrological applications for instance by improving the runoff associated to high precipitation rates in models. Using several realizations is a way to study uncertainty propagation through a model.

Land ice is losing mass to the worlds oceans at an accelerated rate. The worlds glaciers contain much less ice than the ice sheets but contribute equally to eustatic sea level rise and are expected to continue to do so over the coming centuries if global temperatures continue to rise. It is therefore important to characterize the mass balance of these glaciers and its relationship to climate trends and variability. In the Canadian High Arctic, analysis of long-term surface mass balance records shows a shift to more negative mass balances after 1987 and is coincident with a change in the mean location of the July circumpolar vortex, a mid-troposphere cyclonic feature known to have a strong influence on Arctic summer climate. Since 1987 the occurrence of July vortices centered in the Eastern Hemisphere have increased significantly. This change is associated with an increased frequency of tropospheric ridging over the Canadian High Arctic, higher surface air temperatures, and more negative glacier mass balance. However, regional scale mass balance modeling is needed to determine whether or not the long-term mass balance measurements in this region accurately reflect the mass balance of the entire Canadian High Arctic. The Canadian High Arctic is characterized by high relief and complex terrain that result in steep horizontal gradients in surface mass balance, which can only be resolved if models are run at high spatial resolutions. For such runs, models often require input fields such as air temperature that are derived by downscaling of output from climate models or reanalyses. Downscaling is often performed using a specified relationship between temperature and elevation (a lapse rate). Although a constant lapse rate is often assumed, this is not well justified by observations. To improve upon this assumption, near-surface temperature lapse rates during the summer ablation season were derived from surface measurements on 4 Arctic glaciers. Near-surface lapse rates vary systematically with free-air temperatures and are less steep than the free-air lapse rates that have often been used in mass balance modeling. Available observations were used to derive a new variable temperature downscaling method based on temperature dependent daily lapse rates. This method was implemented in a temperature index mass balance model, and results were compared with those derived from a constant linear lapse rate. Compared with other approaches, model estimates of surface mass balance fit observations much better when variable, temperature dependent lapse rates are used. To better account for glacier-climate feedbacks within mass balance models, more physically explicit representations of snow and ice processes must be used. Since absorption of shortwave radiation is often the single largest source of energy for melt, one of the most important parameters to model correctly is surface albedo. To move beyond the limitations of empirical snow and ice albedo parameterizations often used in surface mass balance models, a computationally simple, theoretically-based parameterization for snow and ice albedo was developed. Unlike previous parameterizations, it provides a single set of equations for the estimation of both snow and ice albedo. The parameterization also produces accurate results for a much wider range of snow, ice, and atmospheric conditions.

Water resource planning and management practices in the southeastern United States may be vulnerable to climate change. This vulnerability has not been quantified, and decision makers, although generally concerned, are unable to appreciate the extent of the possible impact of climate change nor formulate and adopt mitigating management strategies. Thus, this dissertation aims to fulfill this need by generating decision worthy data and information using an integrated climate change assessment framework. To begin this work, we develop a new joint variable spatial downscaling technique for statistically downscaling gridded climatic variables to generate high-resolution, gridded datasets for regional watershed modeling and assessment. The approach differs from previous statistical downscaling methods in that multiple climatic variables are downscaled simultaneously and consistently to produce realistic climate projections. In the bias correction step, JVSD uses a differencing process to create stationary joint cumulative frequency statistics of the variables being downscaled. The functional relationship between these statistics and those of the historical observation period is subsequently used to remove GCM bias. The original variables are recovered through summation of bias corrected differenced sequences. In the spatial disaggregation step, JVSD uses a historical analogue approach, with historical analogues identified simultaneously for all atmospheric fields and over all areas of the basin under study. In the second component of the integrated assessment framework, we develop a data-driven, downward hydrological watershed model for transforming the climate variables obtained from the downscaling procedures to hydrological variables. The watershed model includes several water balance elements with nonlinear storage-release functions. The release functions and parameters are data driven and estimated using a recursive identification methodology suitable for multiple, inter-linked modeling components. The model evolves from larger spatial/temporal scales down to smaller spatial/temporal scales with increasing model structure complexity. For ungauged or poorly-gauged watersheds, we developed and applied regionalization hydrologic models based on stepwise regressions to relate the parameters of the hydrological models to observed watershed responses at specific scales. Finally, we present the climate change assessment results for six river basins in the southeastern United States. The historical (baseline) assessment is based on climatic data for the period 1901 through 2009. The future assessment consists of running the assessment models under all IPCC A1B and A2 climate scenarios for the period from 2000 through 2099. The climate assessment includes temperature, precipitation, and potential evapotranspiration; the hydrology assessment includes primary hydrologic variables (i.e., soil moisture, evapotranspiration, and runoff) for each watershed.

Using the regional climate model WRF, and the NCEP-NCAR Reanalysis Project data asboundary and initial conditions, regional precipitation was estimated by means of thedynamical downscaling technique for two selected periods, January 2000 and September2007. These months show very particular climatic characteristics of the precipitationregimen in Central America, like dry (wet) conditions in the Pacific (Caribbean) coast of theCentral American isthmus, in January, and wet (dry) conditions, respectively in each coast,during September. Four-nested-domains, each grids of resolution of 90 km (d01), 30 km(d02), 10 km (d03), and 3.3 km (d04), were configured over this region. The runs werereinitialized each 5 days with 6 hours of spin-up time for adjustment of the model. A total of8 experiments (4 per month) were tested in order to study: a) two important CumulusParameterization Schemes (CPS), Kain-Fritsch (KF) and Grell-Devenyi (GD); and b) thephysical interaction between nested domains (one- and two-way nesting), during eachsimulated month.January 2000 results showed that the modeled precipitation is in agreement withobservations, and also captured the mean climate features of rainfall concerning magnitude,and spatial distribution, like the particular precipitation contrast between the Pacific and theCaribbean coast.Outputs from September 2007 revealed significant differences when a visual comparison ismade to the spatial distribution of each coarse domain (d01, d02, and d03) with theirrespective domain in each experiment. However, the inner grids (d04) in all theexperiments, showed a similar spatial distribution and magnitude estimation, mainly inthose runs using one-way nesting configuration. Furthermore, the results for this mothdiffer substantially with observations, and the latter could be related with associateddeficiencies in the boundary condition that do not reproduce well the transition periodsfrom warm to cold El Niño episodes.Moreover, in all the experiments, the KF scheme calculated more precipitation than the GDscheme and it is associated to the ability of the GD scheme to reproduce spotty but intenserainfall, and apparently, this scheme is reluctant to activate, frequently yielding little or norain. However, when rainfall does develop, it is very intense.Also, the time series do not replicate specific precipitation events, thus, the 5-daysintegration period used in this study, is not enough to reproduce short-period precipitationevents.Finally, physical interaction issues between the nested domains are reflected indiscontinuities in the precipitation field, which have been associated to mass fieldadjustment in the CPS. / Nederbörden i Central Amerika har uppskattats med dynamisk nedskalning för två utvaldaperioder, januari 2000 och september 2007. Global återanalysdata från NCEP-NCARsåteranalysprojekt har använts som randdata och initialdata till den regionalaklimatmodellen WRF. De studerade månaderna uppvisar stora variationer inederbördsmönster, t ex lite (mycket) nederbörd under januari och mycket (lite) nederbördunder september för kustområdena längs Stilla havet (Karibiska havet). Fyra nästladedomäner över Central Amerika har använts med en upplösning på 90 km (d01), 30 km (d02),10 km (d03) och 3,3 km (d04). Simuleringarna initialiserades var 5:e dag och de första 6timmarna efter varje initialisering används för modellens anpassning till initialtillståndet.Totalt 8 experiment genomfördes (4 för varje månad) för att studera: (a) två olika sätt attparameterisera konvektion i Cumulusmoln (CPS), Kain-Fritsch (KF) och Grell-Devenyi (GD)och (b) den fysikaliska interaktionen mellan de nästlade domänerna (en- respektive tvåvägsnästlade scheman).För januari 2000 var det god överensstämmelse mellan modellerad och observeradnederbörd. Modellen beskriver väl såväl mängden nederbörd som den rumsligafördelningen, t ex den stora kontrasten mellan kustområdena längs Stilla havet och Karibiskahavet.För september 2007 uppvisar den modellerade nederbörden stora skillnader i de olikaexperimenten för de yttre domänerna (d01, d02, d03). För den inre domänen (d04) ärresultaten från de olika experimenten betydligt mer lika, särskilt för experimenten medenvägs nästlade scheman. Vidare skiljer sig den modellerade nederbörden väsentligt frånobserverad nederbörd under september 2007. Detta kan förklaras med felaktiga randdatapå grund av problemet i återanalys data att reproducera perioder med övergång från varmtill kall El Niño. I alla experiment gav KF mer nederbörd än GD, det kan förklaras med att GDbättre reproducerar kortvarig, intensiv nederbörd. Det finns en viss tröghet innannederbörden i GD aktiveras, vilket innebär större frekvens av lite eller ingen nederbörd. Närnederbörden väl utvecklas blir den dock intensiv. WRF-modellen klarar inte av att återgespecifika nederbördshändelser för de genomförda experimenten, vilket betyder att 5-dagarär för lång simuleringstid för att kunna reproducera specifika händelser. Slutligen,interaktion mellan de nästlade domänerna skapar diskontinuiteter i nederbördsmöns.

No description available.

Climate change is altering hydrological processes with varying degrees in various regions of the world. This research work investigates the possible impacts of climate change on water resource and Hydropower production potential in central and southern Africa. The Congo, Zambezi and Kwanza, Shire, Kafue and Kabompo basins that lie in central and southern Africa are used as case studies. The review of climate change impact studies shows that there are few studies on impacts of climate change on hydropower production. Most of these studies were carried out in Europe and north America and very few in Asia, south America and Africa. The few studies indicate that southern Africa would experience reduction in precipitation and runoff, consequently reductions in hydropower production. There are no standard methods of assessing the resulting impacts. Two approaches were used to assess the impacts of climate change on water resources and hydropower. One approach is lumping changes on country or regional level and use the mean climate changes on mean annual flows as the basis for regional changes in hydropower production. This is done to get an overall picture of the changes on global and regional level. The second approach is a detailed assessment process in which downscaling, hydrological modelling and hydropower simulations are carried out. The possible future climate scenarios for the region of central and southern Africa depicted that some areas where precipitation are likely to have increases while other, precipitation will reduce. The region northern Zambia and southern Congo showed increases while the northern Congo basin showed reductions. Further south in southern African region, there is a tendency of decreases in precipitation. To the west, in Angola, inland showed increases while towards the coast highlighted some decreases in precipitation. On a global scale, hydropower is likely to experience slight changes (0.08%) due to climate change by 2050. Africa is projected for a slight decrease (0.05%), Asia with an increase of 0.27%, Europe a reduction up to 0.16% while America is projected to have an increase of 0.05%. In the eastern African region, it was shown that hydropower production is likely to increase by 0.59%, the central with 0.22% and the western with a 0.03%. The southern, and northern African regions were projected to have reductions of 0.83% and 0.48% respectively. The basins with increases in flow projections have a slight increase on hydropower production but not proportional to the increase in precipitation. The basins with decreases had even high change as the reduction was further increased by evaporation losses. The hydropower production potential of most of southern African basins is likely to decrease in the future due to the impact of climate change while the central African region shows an increasing trend. The hydropower system in these regions will be affected consequently. The hydropower production changes will vary from basin to basin in these regions. The Zambezi, Kafue and Shire river basins have negative changes while the Congo, Kwanza and Kabompo river basins have positive changes. The hydropower production potential in the Zambezi basin decreases by 9 - 34%. The hydropower production potential in the Kafue basin decreases by 8 - 34% and the Shire basin decreases by 7 - 14 %. The southern region will become drier with shorter rainy seasons. The central region will become wetter with increased runoff. The hydropower production potential in the Congo basin reduces slightly and then increases by 4% by the end of the century. The hydropower production potential in the Kwanza basin decreases by 3% and then increases by 10% towards the end of the century and the Kabompo basin production increases by 6 - 18%. It can be concluded that in the central African region hydropower production will, in general, increase while the southern African region, hydropower production will decrease. In summary, the analysis has shown that the southern African region is expected to experience decreases in rainfall and increases in temperature. This will result in reduced runoff. However the northern part of southern Africa is expected to remain relatively the same with slight increase, moving northwards towards the central African region where mainly increases have been registered. The southern African region is likely to experience reductions up to 5 - 20% while the central African region is likely to experience an increase in runoff in the range of 1 - 5%. Lack of data was observed as a critical limiting factor in modelling in the central and southern Africa region. The designs, plans and operations based on poor hydrological data severely compromise performance and decrease efficiency of systems. Climate change is expected to change these risks. The normal extrapolations of historical data will be less reliable as the past will become an increasingly poor predictor of the future. Better (observed) data is recommended in future assessments and if not better tools and methods for data collection/ should be used. Future designs, plans and operations should include and aspect of climate change, if the region is to benefit from the climate change impacts.

Les étiages extrêmes entraînent souvent des conséquences importantes sur de multiples secteurs socio-économiques. Les récentes études liées au changement climatique semblent indiquer que ces événements risquent de devenir plus sévères et plus fréquents au cours des prochaines décennies. Malheureusement, le peu de données hydrométéorologiques disponibles avant les années 1970 ne permet pas de remettre dans un contexte historique les derniers événements observés ni ceux projetés par les études d'impact. Ces travaux de thèse s'attachent ainsi à améliorer l'état des connaissances sur les étiages extrêmes historiques ayant touché le territoire français depuis la fin du XIXe siècle grâce à une reconstruction hydrométéorologique. Ils proposent aussi un nouveau cadre méthodologique pour l'étude de ces extrêmes.Des informations sur la situation synoptique atmosphérique depuis la fin du XIXe siècle ont récemment été mises à disposition de la communauté via des réanalyses globales étendues comme la Twentieth Century Reanalysis (20CR). Ces travaux introduisent la méthode SCOPE qui permet de reconstruire la météorologie locale sur l'ensemble de la France à partir de la réanalyse 20CR. Elle produit un ensemble de 25 chroniques météorologiques spatialement homogènes de précipitations, température et évapotranspiration sur la période 1871-2012 et sur une grille de 64 km² recouvrant la France. Ces séries constituent le jeu de données SCOPE Climate, qui est ensuite utilisé comme forçage d'un modèle hydrologique sur un large échantillon de plus de 600 bassins versants français faiblement anthropisés. Un ensemble de 25 reconstructions de débits journaliers, appelé SCOPE Hydro, est ainsi produit entre 1871 et 2012. SCOPE Climate et SCOPE Hydro ont tous deux montré de bonnes performances en comparaison de données indépendantes de leur construction.Des événements spatio-temporels d'étiage extrême sont finalement identifiés et caractérisés à partir des chroniques de débit de SCOPE Hydro. Ces événements sont tout d'abord définis localement comme une période où le débit se trouve en-dessous d'un seuil mixte, combinaison d'un seuil fixe et d'un seuil variable avec la saison. Un regroupement spatial des événements locaux au sein du même événement spatio-temporel d'étiage extrême est effectué à l'échelle de la France, suivant une méthode spécialement mise au point dans le cadre de ces travaux. A l'issue de ces étapes, un événement peut être étudié localement ou à l'échelle nationale, grâce à des caractéristiques d'étendue, de durée ou de sévérité. Cette méthodologie permet d'identifier des événements exceptionnels d'étiage extrême anciens et peu connus (1878, 1893, 1942-1949), ou relativement récents mais peu documentés (1972, 1978, 1985), en plus de ceux connus (1921, 1976, 1989-1990, 2011). L'étude de l'évolution de ces événements sur 140 ans montre une plus grande proportion du territoire français touchée par des étiages extrêmes après les années 1940. Une comparaison des événements reconstruits à des sources documentaires faisant mention des sécheresses passées montre une bonne cohérence entre reconstructions et réalité.Ce travail de thèse contribue aux questions scientifiques d'actualité, notamment dans le cadre de la variabilité climatique et du changement climatique d'origine anthropique et de leurs conséquences hydrologiques. Ces travaux proposent deux méthodes innovantes sur la descente d'échelle statistique multivariée et l'identification spatio-temporelle des événements d'étiage extrême. Ils ont par ailleurs produit deux jeux de données hydroclimatiques ensemblistes de référence utilisables pour tout type d'étude climatique et hydrologique requérant une profondeur historique importante. / Extreme low-flow events have significant consequences on numerous socio-economic fields. Hydroclimate projections for the 21st century suggest an increase in low-flow severity and frequency. Nevertheless, projected events as well as recent observed events can hardly be put into a sufficiently long historical perspective due to the lack of hydrometeorological data before the 1970s. This work proposes to improve the knowledge on past extreme low-flow events having affected France since the end of the 19th century thanks to a hydrometeorological reconstruction. It also provides a new methodological framework to study these extreme events.Information on the atmospheric synoptic situation since the end of the 19th century have recently been released to the scientific community through extended global reanalyses like the Twentieth Century Reanalysis (20CR). This work introduces the SCOPE method that provides local meteorological reconstructions on the entire France derived from the 20CR reanalysis. SCOPE produces a 25-member ensemble of spatially coherent meteorological series of daily precipitation, temperature and evapotranspiration over the 1871-2012 period and on a 64 km² grid covering France. This dataset, called SCOPE Climate, is then used as forcings to run continuous hydrological simulations over more than 600 near-natural French catchments leading to a 25-member ensemble of daily streamflow time series -- called SCOPE Hydro -- between 1871 and 2012. SCOPE Climate and SCOPE Hydro show a relatively high skill during validation experiments against independent data.Spatio-temporal extreme low-flow events are finally identified and characterised from SCOPE Hydro reconstructed series. The events are first locally identified based on deficit characteristics under a novel combination of a fixed threshold and a daily variable threshold. A spatial matching procedure at the scale of France is developed in order to spatially assemble local extreme events into the same spatio-temporal event. After these steps, an event can either be studied at the local or national scale through its spatial extent, duration or severity characteristics. This work identified past and little known exceptional extreme events (1878, 1893, 1942-1949) or recent but poorly documented events (1972, 1978, 1985) besides well-known events (1921, 1976, 1989-1990, 2011). The evolution of these events since 1871 shows that a greater proportion of the French territory is affected by extreme low-flow events since the 1940s. A good coherence is found between reconstructed events and documentary sources on historical droughts.This work contributes to timely scientific issues, especially within the context of climate change and its hydrological impacts. This work proposes two innovative methods on multivariate statistical downscaling and spatio-temporal identification of extreme low-flow events. It also produced two reference hydroclimatic datasets that may be used in any study requiring long hydrometeorological series.

Le Bilan de Masse de Surface (BMS, c'est-à-dire les précipitations de neige auxquelles est retranchée l'ablation par sublimation, ruissellement ou érosion) de la calotte polaire Antarctique représente une contribution majeure et encore mal connue à l'évolution actuelle du niveau des mers. Le stockage d'eau douce par accumulation de neige sur la calotte posée est supposé s'intensifier au cours du 21eme siècle, modérant l'élévation du niveau des mers et modifiant la dynamique glaciaire. Les trois-quarts du bilan de masse de surface Antarctique sont concentrés au dessous de 2000 m d'altitude alors que cette zone ne représente que 40% de la surface de la calotte posée. Les précipitations orographiques sont une contribution majeure à l'accumulation dans cette région, il est donc crucial d'estimer précisément ce terme. La modélisation de ce processus est fortement dépendant de la résolution des modèles, car les pentes de la calotte influencent l'intensité des précipitations orographiques. La sublimation et la fonte de la neige sont eux aussi fortement dépendant de l'élévation. Bien qu'ils contribuent actuellement peu au bilan de masse de surface de l'Antarctique, ils sont susceptibles de subir des changements importants au cours des prochains siècles. Les modèles atmosphériques de climat, globaux ou régionaux, ne peuvent pas atteindre une résolution allant au delà de 40 km sur l'Antarctique pour des simulations à l'échelle du siècle du fait de coûts de calcul importants. A ces résolutions, la topographie des zones côtières Antarctique n'est pas correctement représentée. C'est pourquoi nous avons développé le modèle de régionalisation SMHiL (Surface Mass balance High-resolution downscaLing) qui permet d'estimer les composantes du bilan de masse de surface Antarctique à haute résolution (~15 km) à partir de champs atmosphériques de plus grande échelle. Nous calculons l'effet de la topographie fine sur les précipitations orographiques et sur les processus de couche limite menant à la sublimation, la fonte et le regel. SMHiL est validé pour la période actuelle à partir d'un jeu de données inédit constitué de plus de 2700 observations de qualité contrôlée. Cependant, les observations représentatives du BMS de la zone côtière Antarctique y sont sous-représentées. Dans ce contexte, nous montrons que la ligne de balise mise en place par l'observatoire GLACIOCLIM-SAMBA en bordure de calotte constitue une référence pour estimer les performances des modèles. Enfin, nous utilisons SMHiL à l'aval du modèle de circulation générale LMDZ4 pour évaluer les variations de BMS au cours du 21eme et du 22eme siècles. Le BMS à haute résolution est significativement différent de celui de LMDZ4 et est plus proche du BMS observé pour la période actuelle. Les résultats suggèrent que les précédentes estimations d'augmentation du BMS au cours du prochain siècle étaient sous-estimées de près de 30% par LMDZ4. Les changements de BMS à faible élévation résulteront d'une compétition entre l'augmentation d'accumulation de neige et de ruissellement. SMHiL est un outil destiné à être appliqué à l'aval d'autres modèles de climat, globaux ou régionaux, pour une meilleure estimation des variations futures du niveau des mers. / The Antarctic Surface Mass Balance (SMB, i.e. the snow accumulation from which we subtract ablation by sublimation, run-off or erosion) is a major yet badly known contribution to changes in the present-day sea level. Water storage by snow accumulation on the Antarctic continent is expected to increase in the 21st century, which would moderate the rise in sea level and impact the ice dynamic response of the ice sheet. Three-quarters of the Antarctic SMB are concentrated below 2000 m above sea level whereas this area represents only 40% of the grounded ice sheet area. Orographic precipitation is a major contributor to snow accumulation in this region, which is why a better estimation of this term is important. The representation of this process by models depends to a great extent on the resolution of the model, since precipitation amounts depend on the ice sheet slopes. Sublimation and snowmelt also depend on elevation, and although they are presently minor contributors to the Antarctic SMB, their role is expected to become more important in the coming centuries. Global and regional atmospheric climate models are unable to achieve a 40-km resolution over Antarctica at a century time scale, due to their computing cost. At this resolution, the Antarctic coastal area is still badly represented. This is why we developed the downscaling model SMHiL (Surface mass balance high-resolution downscaling) to estimate the Antarctic SMB components at a high resolution (~15 km) from large-scale atmospheric forcings. We computed the impact of the high-resolution topography on orographic precipitation amounts and the boundary layer processes that lead to sublimation, melting and refreezing. SMHiL has been validated for the present period with a dataset composed of more than 2700 quality-controlled observations. However, very few of these observations are representative of the Antarctic coastal area. In this context, we show that the GLACIOCLIM-SAMBA stake lines located on the ice sheet coast-to-plateau area is an appropriate reference to evaluate model performance. Finally, we used SMHiL to estimate the SMB changes during the 21st and 22nd centuries, by downscaling the atmospheric global climate model LMDZ4. The high-resolution SMB is significantly different from the SMB given by LMDZ4 and is closer to the observed one for the present period. Our results suggest that previous studies using the LMDZ4 models underestimated the future increase in SMB in Antarctica by about 30%. Future changes in the Antarctic SMB at low elevations will result from the conflict between higher snow accumulation and runoff. The downscaling model is a powerful tool that can be applied to climate models for a better assessment of a future rise in sea level.