Search

The floods that struck lower Tagus valley in February 1979 correspond to the most intense floods in this river and affected the largest number of people in a river flow event in Portugal, during the last 150 years. In fact, the vast area affected impacted significantly circa 10k people in the lower Tagus sector (and an additional 7k in other regions of Portugal), including thousands of people evacuated or made homeless. In this context, the present study focuses on an in depth analysis of this event from a hydrodynamic perspective by means of Iber+ numerical model and on developing strategies to mitigate the flood episodes that occur in the lower section of the Tagus river using the outstating floods of February 1979 as benchmark. In this sense, several large dams and reservoirs are located along the Tagus river basin, and for that, dam operating strategies were developed and analyzed to take advantage of these infrastructures to minimize the effect of floods. Overall, the numerical results indicate a good agreement with water marks and some descriptions of the 1979 flood event, which demonstrates the model capability to evaluate floods in the area under study. Regarding flood mitigation, obtained results indicate that the frequency of floods can be reduced with the proposed strategies. In addition, hydraulic simulations corroborated an important decrease in water depth and velocity for the most extreme flood events, and also a certain reduction of flood extension was detected. This confirms the effectiveness of the proposed strategies to help in reducing flood impact in lower Tagus valley through the efficient functioning of dams.

Número especial: Advances in machine learning for natural hazards risk assessment [Abstract:] Reservoirs play a key role in many human societies due to their capability to manage water resources. In addition to their role in water supply and hydropower production, their ability to retain water and control the flow makes them a valuable asset for flood mitigation. This is a key function, since extreme events have increased in the last few decades as a result of climate change, and therefore, the application of mechanisms capable of mitigating flood damage will be key in the coming decades. Having a good estimation of the outflow of a reservoir can be an advantage for water management or early warning systems. When historical data are available, data-driven models have been proven a useful tool for different hydrological applications. In this sense, this study analyzes the efficiency of different machine learning techniques to predict reservoir outflow, namely multivariate linear regression (MLR) and three artificial neural networks: multilayer perceptron (MLP), nonlinear autoregressive exogenous (NARX) and long short-term memory (LSTM). These techniques were applied to forecast the outflow of eight water reservoirs of different characteristics located in the Miño River (northwest of Spain). In general, the results obtained showed that the proposed models provided a good estimation of the outflow of the reservoirs, improving the results obtained with classical approaches such as to consider reservoir outflow equal to that of the previous day. Among the different machine learning techniques analyzed, the NARX approach was the option that provided the best estimations on average. FEDER; 0034_RISC_ML_6_E Xunta de Galicia; ED431C 2021/44 Xunta de Galicia; ED481B-2021-108 Universidade de Vigo; 0000 131H TAL 641

The expansion of marine renewable power is a major alternative for the reduction of greenhouse gases emissions. In Europe, however, the high penetration of offshore wind brings intermittency and power variability into the existing power grid. Offshore solar photovoltaic power is another technological alternative under consideration in the plans for decarbonization. However, future variations in wind, air temperature or solar radiation due to climate change will have a great impact on both renewable energy resources. In this context, this study focusses on the offshore energy assessment off the coast of Western Iberia, a European region encompassing Portugal and the Northwestern part of Spain. Making use of a vast source of data from 35 simulations of a research project called CORDEX, this study investigates the complementarity of offshore wind and solar energy sources with the aim of improving the energy supply stability of this region up to 2040. The most pessimistic greenhouse gases emission scenario (RCP8.5) is considered. The 35 simulations are validated by comparing wind speed at 10m, air temperature at 2m and solar radiation values with data from the ERA5 reanalysis database. Although the offshore wind energy resource has proven to be higher than solar photovoltaic resource at annual scale, both renewable resources showed significant spatiotemporal energy variability throughout the western Iberian Peninsula. When both renewable resources are combined, the stability of the energy resource increased considerably throughout the year. The proposed wind and solar combination scheme is assessed by a performance classification method called Delphi, considering stability, resource, risk, and economic factors. The total index classification increases when resource stability is improved by considering hybrid offshore wind-photovoltaic solar energy production, especially along the nearshore waters.

Financiado para publicación en acceso aberto: Universidade da Coruña/CISUG [Abstract:] Modelling hydrological processes with fully distributed models based on the shallow water equations implies a high computational cost, which often limits the resolution of the computational mesh. Therefore, in practice, modellers need to find a compromise between spatial resolution, numerical accuracy and computational cost. Moreover, this balance is probably related to the accuracy and resolution of the underlying Digital Terrain Model (DTM). In this work, it is studied the effect of the DTM resolution and the size of the computational mesh on the results and on the runtime of a hydrological model based on the 2D shallow water equations. Seven rainfall events in four different basins have been modelled using 3 DTMs and 3 different mesh resolutions. The results obtained highlight the relevance of the vertical accuracy versus the horizontal resolution of the DTMs. Furthermore, it has been observed that mesh resolutions greater than 25 m, together with LiDAR-based DTMs with horizontal resolution greater than 25 m, provide comparable outflow hydrographs. Xunta de Galicia; ED481B-2021-108 Xunta de Galicia; ED431C 2021/44 Xunta de Galicia; ED431C 2018/56 Fondo Europeo de Desarrollo Regional (FEDER); 0034_RISC_ML_6_E

[Abstract:] River floods, which are one of the most dangerous natural hazards worldwide, have increased in intensity and frequency in recent decades as a result of climate change, and the future scenario is expected to be even worse. Therefore, their knowledge, predictability, and mitigation represent a key challenge for the scientific community in the coming decades, especially in those local areas that are most vulnerable to these extreme events. In this sense, a multiscale analysis is essential to obtain detailed maps of the future evolution of floods. In the multiscale analysis, the historical and future precipitation data from the CORDEX (Coordinated Regional Downscaling Experiment) project are used as input in a hydrological model (HEC-HMS) which, in turn, feeds a 2D hydraulic model (Iber+). This integration allows knowing the projected future changes in the flow pattern of the river, as well as analyzing the impact of floods in vulnerable areas through the flood hazard maps obtained with hydraulic simulations. The multiscale analysis is applied to the case of the Miño-Sil basin (NW Spain), specifically to the city of Ourense. The results show a delay in the flood season and an increase in the frequency and intensity of extreme river flows in the Miño-Sil basin, which will cause more situations of flooding in many areas frequented by pedestrians and in important infrastructure of the city of Ourense. In addition, an increase in water depths associated with future floods was also detected, confirming the trend for future floods to be not only more frequent but also more intense. Detailed maps of the future evolution of floods also provide key information to decision-makers to take effective measures in advance in those areas most vulnerable to flooding in the coming decades. Although the methodology presented is applied to a particular area, its strength lies in the fact that its implementation in other basins and cities is simple, also taking into account that all the models used are freely accessible. Xunta de Galicia; ED431C 2021/44 FEDER; 0034_RISC_ML_6_E Xunta de Galicia; ED481B-2021-108

The DualSPHysics model is used to reproduce the physical model tests carried out in the large-scale wave flume at CEDEX, in Madrid (Spain), focused on wave overtopping of vertical coastal structure. This work presents the first attempt to apply Smoothed Particle Hydrodynamics (SPH) models to simulate long duration events (670 waves, corresponding to 25min time simulation) where the cumulative overtopping volume is measured with waves interacting against a vertical wall, that represents a typical defence designed for harbours and marinas in the Mediterranean Sea. A sensitivity analysis on some model parameters, the initial inter-particle distance, the smoothing length and the piston transfer function (by means of a gain factor), has been carried out for model validation. The two first parameters are related to the numerical resolution and the particle interaction distance in SPH, respectively. The gain factor is applied to amplify the magnitude of the piston displacement in order to compensate the wave decay, dependent on the wave steepness. A good agreement between the experimental and numerical cumulative overtopping volume and relative overtopping rate is achieved with errors lower than 1%. Once the model is validated and calibrated, it has been applied to different wave time series with the same energy spectrum in order to investigate the sensitivity of the overtopping response to different wave grouping, obtaining acceptable agreement with the forecasts provided by the European Overtopping manual.

Financiado para publicación en acceso aberto: Universidade de Vigo/CISUG Climate change is causing the global sea surface temperature (SST) to rise. However, warming rates are lower in coastal areas affected by the upwelling that buffers the SST warming. The influence of the Canary upwelling system on coastal SST throughout the 21st century was analyzed taking advantage of the high spatial resolution of the Global Climate Models (GCMs) from the CMIP6 project to capture upwelling features. Future SST changes were evaluated both on the coast and in their oceanic counterparts under the 5–8.5 socioeconomic pathway scenario. Results showed a latitudinal displacement of the upwelling system to the north by the end of the century. Middle and north latitudes showed a strengthening in the buffering effect of upwelling with a greater influence on the differences between coast and ocean. Oppositely, the southernmost latitudes showed a weakening causing a homogenization of the SST between coast and ocean. Xunta de Galicia | Ref. ED431C 2017/64