<p>A global climate change may have large impacts on water resources on regional and global scales. General circulation models (GCMs) are the most used tools to evaluate climate-change scenarios on a global scale. They are, however, insufficiently describing the effects at the local scale. This thesis evaluates different approaches of statistical downscaling of precipitation from large-scale circulation variables, both concerning the method performance and the optimum choice of predictor variables. </p><p>The analogue downscaling method (AM) was found to work well as “benchmark” method in comparison to more complicated methods. AM was implemented using principal component analysis (PCA) and Teweles-Wobus Scores (TWS). Statistical properties of daily and monthly precipitation on a catchment in south-central Sweden, as well as daily precipitation in three catchments in China were acceptably downscaled.</p><p>A regression method conditioning a weather generator (SDSM) as well as a fuzzy-rule based circulation-pattern classification method conditioning a stochastical precipitation model (MOFRBC) gave good results when applied on Swedish and Chinese catchments. Statistical downscaling with MOFRBC from GMC (HADAM3P) output improved the statistical properties as well as the intra-annual variation of precipitation.</p><p>The studies show that temporal and areal settings of the predictor are important factors concerning the success of precipitation modelling. The MOFRCB and SDSM are generally performing better than the AM, and the best choice of method is depending on the purpose of the study. MOFRBC applied on output from a GCM future scenario indicates that the large-scale circulation will not be significantly affected. Adding humidity flux as predictor indicated an increased intensity both in extreme events and daily amounts in central and northern Sweden.</p>

A global climate change may have large impacts on water resources on regional and global scales. General circulation models (GCMs) are the most used tools to evaluate climate-change scenarios on a global scale. They are, however, insufficiently describing the effects at the local scale. This thesis evaluates different approaches of statistical downscaling of precipitation from large-scale circulation variables, both concerning the method performance and the optimum choice of predictor variables. The analogue downscaling method (AM) was found to work well as “benchmark” method in comparison to more complicated methods. AM was implemented using principal component analysis (PCA) and Teweles-Wobus Scores (TWS). Statistical properties of daily and monthly precipitation on a catchment in south-central Sweden, as well as daily precipitation in three catchments in China were acceptably downscaled. A regression method conditioning a weather generator (SDSM) as well as a fuzzy-rule based circulation-pattern classification method conditioning a stochastical precipitation model (MOFRBC) gave good results when applied on Swedish and Chinese catchments. Statistical downscaling with MOFRBC from GMC (HADAM3P) output improved the statistical properties as well as the intra-annual variation of precipitation. The studies show that temporal and areal settings of the predictor are important factors concerning the success of precipitation modelling. The MOFRCB and SDSM are generally performing better than the AM, and the best choice of method is depending on the purpose of the study. MOFRBC applied on output from a GCM future scenario indicates that the large-scale circulation will not be significantly affected. Adding humidity flux as predictor indicated an increased intensity both in extreme events and daily amounts in central and northern Sweden.

Made available in DSpace on 2017-07-20T13:41:56Z (GMT). No. of bitstreams: 1 GABRIELA LEITE NEVES.pdf: 2850568 bytes, checksum: d212a8a17a32a986fc2b0200c7ee256e (MD5) Previous issue date: 2015-02-09 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Climate change is a matter of research, since there are still many questions about its causes and consequences. The effects of climate change are diverse, and can reach from the biodiversity of the planet to the economic, social and environmental sectors. Thus the knowledge of climate variables, their changes and impacts on natural resources is very important. Many researchers already had observed that water, a natural resource of great value, may be disturbed due to climate change. Thus the study of water availability related to climate change becomes necessary. The aim of this study was to analyze the impact of possible climate changes on water availability in the state of Paraná, Brazil. It were used daily data of precipitation, and maximum and minimum air temperature of 28 locations, with series of data from 1980 to 2009. By means of the PGECLIMA_R tool daily weather data were simulated for the period 2010 to 2099, using the climate scenarios projected in the IPCC AR4-, IPCC AR5- and trend analysis of historical data from 1980 to 2009. For the creation of the latter scenario, there was a parallel study to obtain the changes in temperature for each location, which were then projected in the simulations. In order to quantify the water availability under the proposed conditions, we estimated the water balance for the current and simulated weather data using the methodology of Thornthwaite and Matter (1955). With interpolation tools, the results of changes in temperature and hydric indices obtained by the water balance were mapped for the entire state of Paraná. The hydric indices were then compared in different scenarios and analyzed periods. The results show that the state of Paraná has a great climate variability across regions. The trend analysis of air temperature pointed from reductions to increases in this variable and these were responsible for different impacts on water availability. In general, all scenarios were responsible for reductions in water availability, but the northern region is more susceptible to these reductions. The trend scenario showed reductions in water availability for most regions, but there was also increase in some specific regions of Parana State. / As modificações do clima é assunto de pesquisa, já que ainda existem muitos questionamentos sobre suas causas e consequências. Os efeitos das mudanças climáticas são diversos, e podem atingir desde a biodiversidade do planeta até os setores econômicos, sociais e ambientais. Sendo assim o conhecimento das variáveis climáticas, de suas alterações e seus impactos nos recursos naturais é de suma importância. Muitos pesquisadores já observaram que a água, um recurso natural de grande valor, poderá sofrer perturbações decorrentes das mudanças climáticas. Assim o estudo da disponibilidade hídrica frente às mudanças climáticas se torna necessário. O objetivo desse trabalho foi analisar o impacto de possíveis mudanças climáticas na disponibilidade hídrica no estado do Paraná. Utilizou-se dados diários de precipitação, temperatura máxima e temperatura mínima de 28 localidades, que correspondem a uma série de dados de 1980 a 2009. Por meio da ferramenta PGECLIMA_R, realizaram-se simulações de dados climáticos diários, para o período de 2010 até 2099, considerando os cenários climáticos projetados no IPCC-AR4, IPCC-AR5 e na análise de tendência dos dados históricos no período de 1980 a 2009. Para a criação deste último cenário, realizou-se um estudo paralelo em que se obteve as alterações na temperatura para cada localidade, as quais foram então projetadas nas simulações. A fim de quantificar a disponibilidade hídrica nas condições propostas, estimou-se os balanço hídricos para os dados climáticos atuais e para os simulados, pela metodologia de Thornthwaite e Matter (1955). Com ferramentas de interpolação, os valores das alterações na temperatura e dos índices hídricos obtidos pelo balanço hídrico foram espacializados para todo o estado do Paraná. Os índices hídricos então foram comparados nos diferentes cenários e períodos analisados. Os resultados apontam que o estado do Paraná apresenta uma grande variabilidade climática entre suas regiões. A análise de tendência da temperatura do ar apontou desde reduções a aumentos nessa variável e essas alterações foram responsáveis também por diferentes impactos na disponibilidade hídrica. De forma geral, todos os cenários foram responsáveis por reduções na disponibilidade hídrica, porém a região norte foi aquela que mostrou ser mais suscetível a essas reduções. O cenário de tendência apontou reduções na disponibilidade hídrica para a maioria das regiões, porém houve aumento pontual da mesma em algumas regiões especificas do Estado do Paraná.

Tesis por compendio / [ES] La mayoría de los estudios científicos pronostican un incremento en la frecuencia y magnitud de los episodios de precipitaciones extremas como consecuencia de los efectos del cambio climático. Además, se espera que en un plazo de 50 años el 80% de la población mundial viva en zonas propensas a inundaciones. Este incremento en la peligrosidad, vulnerabilidad y en la exposición al peligro de lluvias intensas supone un aumento significativo en el riesgo de inundaciones, ya de por si elevado, que manifiesta la urgente necesidad de tomar medidas encaminadas a reducir la vulnerabilidad y desarrollar metodologías capaces de estimar con la mayor precisión posible la magnitud y la probabilidad de ocurrencia de estos posibles fenómenos extremos. En esta última dirección va dirigida la presente tesis doctoral, presentando una nueva metodología basada en el uso de generadores meteorológicos estocásticos para la estimación de la frecuencia de avenidas extremas tanto en escenarios de clima actual como de cambio climático. Más allá del paradigma de la tormenta de diseño y de los estudios tradicionales de análisis de frecuencia de inundaciones, la metodología propuesta en esta tesis se basa en la simulación sintética continua: generador meteorológico estocástico + modelo hidrológico espacialmente distribuido. El uso de generadores meteorológicos estocásticos para el análisis de frecuencia de inundaciones es una práctica cada vez más común dentro de la comunidad hidrológica. Sin embargo, es necesario disponer de observaciones largas y completas para obtener estimaciones de cuantiles confiables para altos períodos de retorno. La novedad que introduce la metodología propuesta se basa en la integración de estudios regionales de precipitación máxima en la implementación del generador meteorológico, lo que reduce considerablemente la incertidumbre en las estimaciones de cuantiles (especialmente aquellos asociados a eventos de baja frecuencia) debida a los usualmente cortos y escasos registros hidrometeorológicos de los que se dispone hasta la fecha. Esta tesis se presenta como un compendio de cinco publicaciones: tres de ellas ya publicadas y dos en proceso de revisión en revistas indexadas en el Journal Citation Report. Estos documentos narran la progresión de la metodología a lo largo de diversas etapas hasta llegar al enfoque final. Inicialmente concebida para el clima actual a escala diaria, la metodología fue posteriormente adaptada a escala subdiaria y finalmente desarrollada para su aplicación en escenarios de clima futuro. A lo largo de este proceso, se abordaron estudios de incertidumbre asociados a la cantidad de información que involucran tanto las estimaciones de cuantiles de precipitación como de inundación. Las metodologías se han implementado en dos casos de estudio: Rambla de la Viuda (Castellón) y; la cuenca del río Segura, cuyos resultados han evidenciado la solidez y eficacia de las metodologías. En el ámbito de la modelización meteorológica, los resultados han sido consistentes y satisfactorios, demostrando la capacidad de la metodología para representar con precisión las complejidades de los patrones climáticos. Asimismo, en el ámbito hidrológico, la metodología ha exhibido una eficaz capacidad para representar y simular los procesos relacionados con el ciclo del agua, ofreciendo resultados coherentes y satisfactorios en la estimación de caudales y eventos de inundación tanto en clima actual como en clima futuro. Esta consistencia en la robustez de la metodología, tanto en la modelización meteorológica como hidrológica, respalda su aplicabilidad y confiabilidad en entornos y condiciones climáticas diversas. / [CA] La majoria dels estudis científics pronostiquen un increment de la freqüència i la magnitud dels episodis de precipitacions extremes a conseqüència dels efectes del canvi climàtic. A més, s'espera que en un termini de 50 anys el 80% de la població mundial habite en zones propenses a inundacions. Aquest acreixement en la perillositat, vulnerabilitat i exposició al perill de pluges intenses suposa un augment significatiu del risc d'inundacions, ja de per si elevat, que manifesta la urgent necessitat de prendre mesures encaminades a reduir la vulnerabilitat i desenvolupar metodologies que permeten estimar amb la major precisió possible la magnitud i la probabilitat d'ocurrència d'aquests possibles fenòmens extrems. En aquesta última direcció va dirigida la present tesi doctoral, que presenta una nova metodologia basada en l'ús de generadors meteorològics estocàstics per a l'estimació de la freqüència d'avingudes extremes, tant en escenaris de clima actual com de canvi climàtic. Més enllà del paradigma de la tempesta de disseny i dels estudis tradicionals d'anàlisis de freqüència d'inundacions, la metodologia proposada en aquesta tesi es basa en la simulació sintètica contínua: generador meteorològic estocàstic + model hidrològic espacialment distribuït. L'ús de generadors meteorològics estocàstics per a l'anàlisi de freqüència d'inundacions és una pràctica cada vegada més comuna dins de la comunitat hidrològica. No obstant això, és necessari disposar d'observacions llargues i completes per a obtindre estimacions de quantils de confiança per a alts períodes de retorn. La novetat que introdueix la metodologia proposada es basa en la integració d'estudis regionals de precipitació màxima en la implementació del generador meteorològic, la qual cosa redueix considerablement la incertesa en les estimacions de quantils (especialment d'aquells associats a esdeveniments de baixa freqüència) a causa dels usualment curts i escassos registres hidrometeorològics dels quals es disposa fins a la data. Auqesta tesi es presenta com un compendi de cinc publicacions: tres d'elles en ja publicades i dos en procés de revisió en revistes indexades en el Journal Citation Report. Aquests documents narren la progressió de la metodologia al llarg de diverses etapes fins arribar a l'enfocament final. Inicialment concebuda per al clima actual a escala diària, la metodologia va ser posteriorment adaptada a escala subdiària i finalment desenvolupada per a la seua aplicació en escenaris de clima futur. Al llarg d'aquest procés, es van abordar estudis d'incertesa associats a la quantitat d'informació que involucren tant les estimacions de quantils de precipitació com d'inundació. Les metodologies s'han implementat en dos casos d'estudi: Rambla de la Vídua (Castelló) i; la conca del riu Segura, els resultats del qual han evidenciat la solidesa i eficàcia de les metodologies. En l'àmbit de la modelització meteorològica, els resultats han estat consistents i satisfactoris, ja que han demostrat la capacitat de la metodologia per a representar amb precisió les complexitats dels patrons climàtics. Així mateix, en l'àmbit hidrològic, la metodologia ha exhibit una eficaç capacitat per a representar i simular els processos relacionats amb el cicle de l'aigua, i ens ha oferit resultats coherents i satisfactoris en l'estimació de cabals i esdeveniments d'inundació tant en clima actual com en clima futur. Aquesta consistència en la robustesa de la metodologia, tant en la modelització meteorològica com en la hidrològica, recolza la seua aplicabilitat i confiabilitat en entorns i condicions climàtiques diverses. / [EN] Most scientific studies predict an increase in the frequency and magnitude of extreme precipitation events as a consequence of climate change effects. Furthermore, it is expected that within 50 years, 80% of the global population will reside in flood-prone areas. This heightened risk, vulnerability, and exposure to intense rainfall hazards signify a significant rise in the flood risk, already elevated. It underscores the urgent need to implement measures to reduce vulnerability and develop methodologies capable of accurately estimating the magnitude and probability of occurrence of these potential extreme events. This doctoral thesis is directed towards this objective, presenting a new methodology based on the use of stochastic weather generators for estimating the frequency of extreme floods in both current and climate change scenarios. Beyond the Design Storm paradigm and traditional Flood Frequency Analysis studies, the methodology proposed in this thesis relies on continuous synthetic simulation: stochastic weather generator + spatially distributed hydrological model. The use of stochastic weather generators for Flood Frequency Analysis is becoming increasingly common within the hydrological community. However, reliable quantile estimates for high return periods require long and complete observations. The innovation introduced by the proposed methodology lies in the integration of regional studies of maximum precipitation into the implementation of the weather generator, significantly reducing uncertainty in quantile estimates (especially those associated with low-frequency events) due to the typically short and limited hydrometeorological records available to date. This thesis is presented as a compilation of five publications: three already published and two under review in journals indexed in the Journal Citation Report. These documents narrate the progression of the methodology through various stages to its final approach. Initially conceived for the current climate at a daily scale, the methodology was later adapted to a subdaily scale and ultimately developed for application in future climate scenarios. Throughout this process, uncertainty studies were addressed concerning the amount of information involving both precipitation and flood quantile estimates. The methodologies have been implemented in two case studies: Rambla de la Viuda (Castellón) and the Segura River basin, with results demonstrating the robustness and effectiveness of the methodologies. In the field of meteorological modeling, the results have been consistent and satisfactory, showcasing the methodology's ability to accurately represent the complexities of climate patterns. Likewise, in the hydrological domain, the methodology has exhibited effective capabilities in representing and simulating processes related to the water cycle, offering coherent and satisfactory results in the estimation of flows and flood events in both current and future climates. This consistency in the robustness of the methodology, both in meteorological and hydrological modeling, supports its applicability and reliability in diverse environmental and climatic conditions. / This work was supported by the Spanish Ministry of Science and Innovation through the research projects TETISCHANGE (RTI2018-093717-B-100) and TETISPREDICT (PID2022-141631OB-I00). Funding for the Open Access charge has been provided by Universitat Politècnica de València / Beneyto Ibáñez, C. (2024). On the Use of Weather Generators for the Estimation of Low-Frequency Floods under a Changing Climate [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/205179 / Compendio

The UK’s decarbonisation strategy to increasingly electrify heating and transport will change the demand requirement on the electricity system. Additionally, under a climate change future, it is projected that the decarbonised grid will need to be able to operate under higher average temperatures in the UK, increasing the need for comfort cooling during summer and leading to additional electricity demand. These new demands will result in greater variation between minimum and peak demand as well as a significant increase in overall demand. Concurrently, supply-side decarbonisation programmes may lead to more intermittent renewables such as wind, PV, tidal and wave, elevating variability in electricity generation. Coupled with the anticipated higher variation in demand this brings on several challenges in operating the electricity grid. In order to characterise these challenges this research develops a bespoke electricity dispatch model which builds on hourly models of demand and generation. The hourly demand profiles are based on a high electrification of heating, transport and cooling coupled with future temperatures premised on the UKCP09 high emission scenario climate projections. The demand profiles show a significant increase in peak demand by 2050 reaching 194 GW, mainly due to summer cooling loads which contribute 70% of the demand. The cumulative CO2 emissions budgets of the GB power sector that are consistent with avoiding global climate change to 2°C are used to develop two low carbon generation scenarios distinguished by the amount of intermittent renewable generation technologies. The dispatch model tests the capability of generation scenarios with the use of hourly generation models in meeting future demand profiles out to 2050.The outputs from dispatch model indicate that there are shortages and excesses of generation relative to demand from 2030 onwards. The variability analysis outlines low and high generation periods from intermittent technologies along with the pace at which intermittent generation increases or decreases within an hour. The characterisation of variability analysis reveals the type of reserve capacity or smart solutions that are required to maintain the security of electricity supply. The solutions that could address the challenges quantified from the model outputs in operating a decarbonised GB electricity grid are explored using expert interviews. The analysis of the stakeholder interviews suggests smart grid solutions that include technologies as well as changes in operational procedures in order to enhance the operational resilience of the grid. Active Network Management through monitoring and control, demand management, storage systems and interconnectors are proposed to address challenges arising from varying demand and generation variability.