No abstract

Wave and wind forces from tropical cyclones are one of the main design parameters of coastal and offshore infrastructure in tropical areas. The estimation of ocean waves parameters in the design of structures in tropical areas is difficult due to the complexity of wind fields associated with tropical cyclones. The use of numerical wave models, forced with parametric wind fields, is a common practice within the climatic characterization of extreme events. However, there is currently no consensus on the selection of parametric models for wave prediction due to the lack of a rigorous assessment of different models. In this study, six well-known parametric wind models were tested, compared, and applied in the Gulf of Mexico and the Caribbean Sea. Therefore, the evaluation and comparison of the resulting wind and wave fields are presented, showing that a particular model may best represent a specific event, but, when dealing with a large number of events, the choice of a particular parametric wind model or a combination of them does not guarantee greater accuracy.

We investigate the storm impact associated with historical events in the northern Yucatan Peninsula. The study area is prone to coastal flooding due to both its geographical location and low-lying areas. Extreme events associated with tropical cyclones and Central American cold surge (CACS; locally known as Nortes) are ubiquitous in this region, and coastal development in the study area has exacerbated the erosion of the sand beach-dune system. This study aims to assess the impact on the northern coast of Yucatan associated with different types of storms and to investigate the role of the dune in its spatial variability. Nearshore hydrodynamics, associated with hurricanes (Gilbert: 14 September 1988; Isidore: 22 September 2002) and energetic Nortes (Norte A: 12 March 1993; Norte B: 25 December 2004), were computed using a numerical model. The beach and dune characteristics were extracted from a LIDAR flight with a spatial resolution of 1 m conducted in 2011. Furthermore, the extreme water levels and the spatiotemporal variability of the storm-impact regime (swash, collision, overwash, or inundation), along a 41.5 km stretch of coast, were derived using both runup parametrizations and the modeling results. On the one hand, the predominant storm impact regimes for Hurricanes Gilbert and Isidore were inundation and overwash, respectively. The flood that propagated from east to west in the northern Yucatan was due to westerly-directed hurricane tracks. On the other hand, for the Norte events, the predominant impact regimes were collision and overwash for Nortes A and B, respectively. This difference in the impact regime between Norte events can be ascribed to tidal differences. Moreover, during the passages of Nortes A and B, the flood was propagated from west to east in the northern Yucatan, consistent with cold-front paths. The results suggest that the western part of the study area presented a stronger impact regime due to the dune degradation caused by coastal infrastructure and settlements established in those areas. This work highlights the important role of sand dunes in providing natural coastal protection during Norte events.

No abstract

No abstract

This study applies three different methods to assess the flood risk and damage from the strongest high-pressure cold front (locally known as ‘Norte’) event in terms of the residual tide from 30 years (1979–2008) of data for Progreso, Yucatan. The most important difference between the three methods is the estimation of flood vulnerability for Progreso. The first method, proposed by Mexico’s National Center for the Prevention of Disasters (CENAPRED) and used by the Mexican government is based mostly on economic asset (household goods) values and flood impacts. The second (CENAPREDv2) and third (FRI) methods are proposals for assessing risk that include 17 socioeconomic indicators. The former includes economic asset values, as is the case for CENAPRED, while the latter does not. The main results of this study show that the modeled ‘Norte’ event flooded 25% of Progreso’s city blocks, with an estimated economic flood risk of $USD 16,266 (CENAPRED) and $USD 223,779 (CENAPREDv2), and flood damage of $USD 48,848 and $USD 671,918, respectively. When calculating flood risk (FRI) and flood damage (FRI_FD) without monetary terms, the risk categories along the back-barrier behind Progreso varied spatially from ‘very low’ to ‘high’, while areas along the coastal side presented a ‘low’ and ‘very low’ risk. These categories increased for the flood damage because the exceedance probability of the flood was not considered as it was for flood risk in the three methodologies. Therefore, flood damage provides the losses caused by a given flood event without considering how probable that loss may be. In conclusion, this study proposes that the selection of the applied method depends on the main objectives and specific interests when assessing flood risk. For instance, if economic damage is the main concern, then the CENAPRED method should be used as it identifies where the larger economic impacts could occur; when a socioeconomic approach is needed then the FRI should be applied, but if both economic damage and socioeconomic aspects are needed, the CENAPREDv2 is recommended. Besides considering economic aspects, the FRI method also includes social variables that can help to map the most vulnerable population in terms of mobility, education, communication access and others. Therefore, the proposed FRI method is very relevant for disaster risk managers and other stakeholders interested in disaster risk reduction.

Despite the low occurrence of tropical cyclones at the archipelago of San Andres, Providencia, and Santa Catalina (Colombia), Hurricane Iota in 2020 made evident the area vulnerability to tropical cyclones as major hazards by obliterating 56.4 % of housing, partially destroying the remaining houses in Providencia. We investigated the hurricane storm surge inundation in the archipelago by forcing hydrodynamic models with synthetic tropical cyclones and hypothetical hurricanes. The storm surge from synthetic events allowed identifying the strongest surges using the probability distribution, enabling the generation of hurricane storm surge flood maps for 100 and 500 year return periods. This analysis suggested that the east of San Andres and Providencia are the more likely areas to be flooded from hurricanes storm surges. The hypothetical events were used to force the hydrodynamic model to create worst-case flood scenario maps, useful for contingency and development planning. Additionally, Hurricane Iota flood levels were calculated using 2D and 1D models. The 2D model included storm surge (SS), SS with astronomical tides (AT), and SS with AT and wave setup (WS), resulting in a total flooded area (percentage related to Providencia’s total area) of 67.05 ha (3.25 %), 65.23 ha (3.16 %), and 76.68 ha (3.68%), respectively. While Hurricane Iota occurred during low tide, the WS contributed 14.93 % (11.45 ha) of the total flooded area in Providencia. The 1D approximation showed that during the storm peak in the eastern of the island, the contribution of AT, SS, and wave runup to the maximum sea water level was −3.01%, 46.36%, and 56.55 %, respectively. This finding provides evidence of the water level underestimation in insular environments when modeling SS without wave contributions. The maximum SS derived from Iota was 1.25 m at the east of Providencia, which according to this study has an associated return period of 3,234 years. The methodology proposed in this study can be applied to other coastal zones and may include the effect of climate change on hurricane storm surges and sea-level rise. Results from this study are useful for emergency managers, government, coastal communities, and policymakers as civil protection measures.

Debido a la posición geográfica del archipiélago de San Andrés, Providencia y Santa Catalina (SPSC), esta es la parte del territorio colombiano más expuesto a ser inundado por marea de tormenta generada por ciclones tropicales (CT). Entre los peligros asociados a los CT, las inundaciones suelen ocasionar los mayores daños. En este estudio se evalúa un escenario extremo de inundación por marea de tormenta asociada a CT para establecer las áreas susceptibles a la inundación en este archipiélago. Dada la escasez de CT históricos en este archipiélago, se generó una base de datos de CT hipotéticos, los cuales son un conjunto de eventos con una velocidad de viento constante (95.17 m/s), una velocidad de traslación constante de 5.87 m/s y un radio máximo de viento constante de 56.3 km, para trayectorias con cinco direcciones de aproximación al área de interés. Se evaluaron siete trayectorias paralelas para cada dirección, separadas por 6 km. La misma metodología fue utilizada para la isla de San Andrés, y separadamente para Providencia y Santa Catalina, se usaron diferentes eventos debido a la distancia entre ellas (90km). El modelo hidrodinámico se forzó con campos de viento y presión generados a partir de la base de datos de CT hipotéticos para determinar la marea de tormenta e inundación por CT en el archipiélago de SPSC. Los resultados obtenidos incluyen la envolvente de las envolventes individuales de la distribución espacial del espejo de agua (nivel del agua referido al nivel del terreno) de cada evento, reportando el escenario de inundación más conservador generado por marea de tormenta de CT en el archipiélago de SPSC. Las áreas propensas a inundación en la isla de San Andrés están ubicadas al Este de la misma, principalmente donde se encuentran los puertos de abrigo y en la parte Norte donde se encuentran los principales asentamientos humanos. Para la isla de Providencia estas áreas se encuentran al Este, en regiones circunvecinas al aeropuerto y al Norte, a lo largo del canal que la separa de la isla de Santa Catalina. Para esta última isla, las áreas propensas a inundación se encuentran al Sureste. Bajo este escenario de inundación el porcentaje de área afectada sería de 13.39%, 4.24%, y 4.43% para San Andrés, Providencia y Santa Catalina, respectivamente.