abstract: This study uses the Weather Research and Forecasting (WRF) model to simulate and predict the changes in local climate attributed to the urbanization for five desert cities. The simulations are performed in the fashion of climate downscaling, constrained by the surface boundary conditions generated from high resolution land-use maps. For each city, the land-use maps of 1985 and 2010 from Landsat satellite observation, and a projected land-use map for 2030, are used to represent the past, present, and future. An additional set of simulations for Las Vegas, the largest of the five cities, uses the NLCD 1992 and 2006 land-use maps and an idealized historical land-use map with no urban coverage for 1900. The study finds that urbanization in Las Vegas produces a classic urban heat island (UHI) at night but a minor cooling during the day. A further analysis of the surface energy balance shows that the decrease in surface Albedo and increase effective emissivity play an important role in shaping the local climate change over urban areas. The emerging urban structures slow down the diurnal wind circulation over the city due to an increased effective surface roughness. This leads to a secondary modification of temperature due to the interaction between the mechanical and thermodynamic effects of urbanization. The simulations for the five desert cities for 1985 and 2010 further confirm a common pattern of the climatic effect of urbanization with significant nighttime warming and moderate daytime cooling. This effect is confined to the urban area and is not sensitive to the size of the city or the detail of land cover in the surrounding areas. The pattern of nighttime warming and daytime cooling remains robust in the simulations for the future climate of the five cities using the projected 2030 land-use maps. Inter-city differences among the five urban areas are discussed. / Dissertation/Thesis / Doctoral Dissertation Aerospace Engineering 2015

Étudier les précipitations et leur lien avec la circulation atmosphérique augmente notre connaissance de leurs caracteristiques et aide à anticiper leur comportement futur. Des méthodes de déscente d'échelle sont développées pour fournir des informations météorologiques locales et importantes pour l'hydrologie à partir des informations issues des réanalyses ou des projections globales du climat. La méthode SANDHY (Stepwise ANalogue Downscaling method for HYdrology) est étendue à l'ensemble de la France métropolitaine en optimisant les domaines pour le prédicteur géopotentiel pour les 608 zones climatiquement homogènes en France en utilisant un algorithme qui permet de prendre en compte l'équifinalité. Une grande diversité des domaines pour le prédicteur géopotentiel a été trouvée. Trois voies pour augmenter la cohérence spatiale et diminuer l'espace des paramètres sont explorés : prendre en compte les domaines optimisés pour des zones voisines, rassembler des zones en utilisant des algorithmes d'aggregation et utiliser un preditant moins asymétrique pendant l'optimisation. Utiliser de l'information issues de zones voisines permet de compenser certaines limitations de l'algorithme d'optimisation. Une méthode de vérification spatiale (SAL) est ici adaptée pour les précipitations probabilistes simulées par SANDHY. Des mesures de performance derivées de cette version probabiliste du SAL sont ensuite utilisées pour évaluer différentes stratégies de déscente d'échelle concernant la cohérence spatiale à l'échelle d'un bassin versant. Les domains optimisés localement pour le prédicteur géopotentiel permettent de mieux localiser les précipitations dans le bassin tandis que des domains uniformes sur tout le bassin apportent une structure des précipitations plus réaliste. Les simulations de débit pour le bassin de la Durance sont le plus sensible à la localisation des précipitations ce qui souligne l'interêt d'une optimisation locale des domaines des prédicteurs. / Studying past and present day precipitation and its link to large scale circulation increases our understanding of precipitation characteristics and helps to anticipate their future behaviour. Downscaling techniques are being developed to bridge the gap between large-scale climate information from global reanalyses or GCM global projections and local meteorological information relevant for hydrology. The stepwise analogue downscaling method for hydrology (SANDHY) is extended to the whole mainland of France by optimising the geopotential predictor domains for 608 zones covering France using a multiple growing rectangular domain algorithm that allows to take equifinality into account. A high diversity of predictor domains has been found. To increase the spatial coherence three ways are explored to reduce the parameter space: assessing the skill for predictor domains found for other zones, form groups of zones using cluster algorithms and using a less skewed predictand variable during optimisation. Using information from neighbouring zones allows to counterbalance in part limitations of the optimisation algorithm. A feature based spatial verification method (SAL) is adapted for probabilistic precipitation simulation as provided by SANDHY. Skill scores derived from the probabilistic SAL are used to assess different strategies for spatially coherent precipitation downscaling at catchment scale. Locally optimised predictor domains lead to a better localisation of precipitation in the catchment and higher local skill while uniform predictor domains for the whole catchment lead to a more realistic spatial structure of the simulated precipitation. Streamflow simulations for the Durance catchment (Southern Alps) are most sensitive to the realistic localisation of precipitation which highlights the interest of locally optimising predictor domains.STAR

abstract: This study performs numerical modeling for the climate of semi-arid regions by running a high-resolution atmospheric model constrained by large-scale climatic boundary conditions, a practice commonly called climate downscaling. These investigations focus especially on precipitation and temperature, quantities that are critical to life in semi-arid regions. Using the Weather Research and Forecast (WRF) model, a non-hydrostatic geophysical fluid dynamical model with a full suite of physical parameterization, a series of numerical sensitivity experiments are conducted to test how the intensity and spatial/temporal distribution of precipitation change with grid resolution, time step size, the resolution of lower boundary topography and surface characteristics. Two regions, Arizona in U.S. and Aral Sea region in Central Asia, are chosen as the test-beds for the numerical experiments: The former for its complex terrain and the latter for the dramatic man-made changes in its lower boundary conditions (the shrinkage of Aral Sea). Sensitivity tests show that the parameterization schemes for rainfall are not resolution-independent, thus a refinement of resolution is no guarantee of a better result. But, simulations (at all resolutions) do capture the inter-annual variability of rainfall over Arizona. Nevertheless, temperature is simulated more accurately with refinement in resolution. Results show that both seasonal mean rainfall and frequency of extreme rainfall events increase with resolution. For Aral Sea, sensitivity tests indicate that while the shrinkage of Aral Sea has a dramatic impact on the precipitation over the confine of (former) Aral Sea itself, its effect on the precipitation over greater Central Asia is not necessarily greater than the inter-annual variability induced by the lateral boundary conditions in the model and large scale warming in the region. The numerical simulations in the study are cross validated with observations to address the realism of the regional climate model. The findings of this sensitivity study are useful for water resource management in semi-arid regions. Such high spatio-temporal resolution gridded-data can be used as an input for hydrological models for regions such as Arizona with complex terrain and sparse observations. Results from simulations of Aral Sea region are expected to contribute to ecosystems management for Central Asia. / Dissertation/Thesis / Ph.D. Aerospace Engineering 2012

Ce travail de thèse poursuit l’objectif d’une simulation numérique multiéchelle d’un assemblage de réacteur à neutrons rapides. Malgré la puissance croissante des ordinateurs, la CFD fine complète d’un tel système demeure extrêmement coûteuse dans un contexte de recherche et développement. Nous proposons alors, une fois déterminé le comportement thermohydraulique moyen de l’assemblage, de reconstruire localement l’information aux fines échelles, l’ensemble de la démarche requérant un temps de calcul bien moindre qu’une simulation de la totalité de la structure. La description à l’échelle moyenne est obtenue soit par le formalisme de prise de moyenne volumique en milieu poreux, soit via une approche alternative historiquement développée pour les assemblages de RNR-Na. Elle fournit des informations utilisées comme contraintes d’un sous-problème de raffinement d’échelle, par l’intermédiaire d’une technique de pénalisation des équations de conservation locales. Ce sous-problème exploite le caractère périodique de la structure en s’appuyant sur des conditions aux limites de périodicité des champs recherchés ou de leur déviation spatiale. Après validation des méthodologies sur des applications modèles, nous entreprenons leur mise en oeuvre sur des configurations « industrielles » qui démontrent la viabilité de cette approche multiéchelle. / The present work is devoted to a multiscale numerical simulation of an assembly of fast neutron reactor. In spite of the rapid growth of the computer power, the fine complete CFD of a such system remains out of reach in a context of research and development. After the determination of the thermalhydraulic behaviour of the assembly at the macroscopic scale, we propose to carry out a local reconstruction of the fine scale information. The complete approach will require a much lower CPU time than the CFD of the entire structure. The macroscale description is obtained using either the volume averaging formalism in porous media, or an alternative modeling historically developed for the study of fast neutron reactor assemblies. It provides some information used as constraint of a downscaling problem, through a penalization technique of the local conservation equations. This problem lean on the periodic nature of the structure by integrating periodic boundary conditions for the required microscale fields or their spatial deviation. After validating the methodologies on some model applications, we undertake to perform them on “industrial” configurations which demonstrate the viability of this multiscale approach.

The worldwide challenges to achieve cost effective protection against global warming impacts and to acquire reliable decision making tools continually force new developments in the area of climate change research. Climate change impacts projections involve several steps: emission scenarios generation, Global Circulation Models and Regional Climate Models (GCM/RCM) runs, downscaling, impact model running, analysis of results and decision making. Unfortunately, GCM/RCMs outputs are often biased and need to be processed before being fed into impact models. This thesis describes the effort carried out to alleviate the burden of downscaling coarse hydro-climatology data outputs from GCM/RCM and making results readily available for climate change impact analysis for specific regions, particularly in the African continent. GCM/RCM outputs are highly unreliable at the sub-grid scale to be used for region specific impact analysis (Wilby, Hay, & Leavesly, 1999). Furthermore, raw GCM/RCM outputs are often downscaled under the premises that the latter offer very coarse spatial resolution. The Internet is a common resource for users of climate change data to access relevant information. Web-based interfaces offer users the capability to retrieve such data. This thesis involves the development of a new web-portal, which addresses the demand for climate change data at the daily scale. It is a user-friendly interactive web-based interface with multiple functionalities including: capacity to process information, capacity to search, sort, retrieve and filter data and download features. Six climate variables are considered in this project: precipitation, maximum temperature, minimum temperature, wind speed, relative humidity and solar radiation. The aforementioned climate variables have been downscaled to specific geographical locations and results have been made available at a fine temporal resolution – the daily scale. The data portal currently hosts climate change data for nine stations in western Africa: Agadez, Brini N’Konni, Gaya, Maine Soroa, Maradi Airport, Niamey Airport, Tahoua, Tillabery and Zinder Airport. The above mentioned climate stations are all located in Niger. Nonetheless, the project aims to expand and cover further ground in Africa. Quantile - Quantile downscaling, also known as Quantile-Quantile mapping, matching or transformation is a statistical procedure used in this project to downscale raw GCM/RCM outputs. GCM/RCM outputs from the AMMA-Ensemble sets under the SRES A1B scenario were used as raw data.

Predicted future precipitation is downscaled and used to drive a hydrologic model to assess future erosion potential in a semi-alluvial clay-bed watercourse, Watts Creek. The 21 km2 watershed is predominantly urban, with overall impervious cover of 22%, and the remaining land use split between agricultural and forested areas. Continuous simulations for the open water year, excluding spring freshet (April 1st to October 31st) were performed using the SWMHYMO (Stormwater Management Hydrologic Model) lumped hydrologic modelling platform. A shear stress exceedance and stream power erosion routine was added to the platform to calculate erosion potential. To account for uncertainty in the collected data, nine different observed discharge data sets were used to calibrate the model, each leading to a distinct set of calibrated parameter values. The difference between the observed data sets lies in the choice of rating curves and the collection period. The 2041-2080 precipitation outputs of the fourth version of the Canadian Regional Climate Model (CanRCM4) ran under Representative Concentration Pathways (RCPs) 4.5 and 8.5 at the MacDonald Cartier International Airport were downscaled using quantile matching and then used as input to the hydrologic model. For each set of calibrated parameters, a cumulative effective work index (CWI) based on the reach-averaged shear stress was calculated for Watts Creek during the open water year using both the historic (1968 2007) and projected future (2041-2080) flows, using a bed material critical shear stress for entrainment of 3.7 Pa. Results suggest an increase of 75% (resp. 139%) under RCP4.5 (resp. RCP8.5) in CWI compared to historic conditions for the average measured bed strength. The work index increase is driven by an increased occurrence of above-threshold events, and more importantly by the increased frequency of large events. The predicted flow regime under climate change would significantly alter the erosion potential and stability of Watts Creek. A channel adjustment sensitivity analysis, which balances future erosion potential with historic potential, was implemented and indicated that the channel could widen in the future from the current bankfull width of 6.1 m to 8.2 m for RCP4.5 and 10.2 m for RCP8.5. Specific morphological behaviour should be investigated in more detail, particularly to assess if the governing erosion mechanism is seasonally dependent, perhaps incising during spring freshet and widening when the bed is vegetated in the summer.

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第20688号 / 工博第4385号 / 新制||工||1681(附属図書館) / 京都大学大学院工学研究科都市社会工学専攻 / (主査)教授 小池 克明, 教授 三ｹ田 均, 准教授 須崎 純一 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

No description available.

No description available.

No description available.