· · · Institutional Repository

Fort Bay harbour is a small harbour, located at the South of the island of Saba. This island is part of the Netherlands Antilles and it is situated in the Caribbean Sea where hurricanes may occur several times a year. Since its construction in 1972, Fort Bay harbour has experienced several hurricanes. Unfortunately some of these hurricanes were responsible for damage to the harbour structures (mostly to the breakwaters). The damages caused by the most recent hurricane Lenny are still present. To improve this situation, local authorities of Saba have commissioned CEC (subsidiary company of Witteveen+ Bos) to produce a plan focused on a full restoration of the original harbour facilities in combination with an economically feasible protection against hurricane conditions. A basic element in this restoration plan is to identify what has caused the apparent damage to the Fort Bay harbour structures. The first part of the present study is meant to contribute to such an identification. From an overview of the history of Fort Bay harbour, an inventory is made of possible reasons why the harbour structures have not been able to withstand the conditions they have been exposed to. This analysis has a qualitative character. One of the items that have come forward in this analysis is a worsening of atmospheric and hydraulic conditions in the area around Saba. Looking at the damage caused by hurricanes that have occurred in the past few decades, there is a possibility that such adverse conditions show an increase in both frequency and intensity. Verification of this potential increase is the subject of the second (and also the last) part of the present study.

Het eiland Bonaire, onderdeel van de benedenwindse eilanden van de Nederlandse Antillen, Hgt op de rand van de Caribbische hurricanegordel. Het risico bestaat dan ook dat het eiland getroffen wordt door een hurricane. Aangezien de westkust van het eiland weinig hoger dan het waterniveau ligt, is er in zo'n geval weinig bescherming tegen de doordringing van golven op de kust. Om de mate van golfaanval en inundatie te bepalen zijn in diverse voorstudies de waterstandsverhoging en golfhoogte voor de kust van Bonaire bepaald met computermodellen. Hieruit is gebleken dat er een significante golfhoogte van ongeveer 3 m te verwachten is met een maximale waterstandsverhoging van 0.5 m. Voor de overgang van de situatie voor het breken van de golf naar de brekerzone en golfoverslag zijn nog experimenten gedaan, maar de verdere doordringing op het land zijn echter geen computermodellen of theorieen voor handen. Om toch deze situatie te kunnen analyseren is er gekozen voor het gebruik van een schaalmodel. Allereerst is een tweetal prototype situaties gedefinieerd die als representatief kunnen worden gezien voor de westkust van Bonaire. Deze situaties verschillen op het punt dat er een situatie met strandmuur en een situatie zonder strandmuur is. Deze prototype situaties zijn vervolgens op schaal1 :15 nagebouwd in een golfgoot, waarbij ook de golfhoogten en waterstanden zijn omgeschaald. Er is vervolgens een meetprogramma samengesteld waarbij variaties zijn aangebracht in de waterstand, de golfsteilheid van de inkomende golven en de ruwheid van het achterland. Na verwerking van de meetgegevens zijn deze met behulp van spectraalanalyse en dimensie-analyse geanalyseerd om de relaties te vinden tussen de in het model gevarieerde parameters en de mate van golfdoordringing en setup op het land. Uit deze analyses zijn vooral de relaties tussen de waterstand en de golfdoordringing of setup naar voren gekomen Vervolgens zijn de gegevens uit de modelsituatie omgeschaald naar de prototype situatie en is er gekeken naar de impactbelasting die de golven kunnen veroorzaken tegen de gebouwen aan de kust teneinde een zone te kunnen bepalen waarbuiten geen schade op zal treden aan deze gebouwen. De golfhoogten op het land bleken maximaal 0.8 m in de eerste meters te worden voor de situatie waar een strandmuur aanwezig is, bij een waterstand van ongeveer 0.5 m, die over een lange afstand landinwaarts aanwezig blijft. Aan de hand van controle berekeningen voor betonconstructies is gebleken dat het risico van schade aan de wanden van de gebouwen niet zo zeer aanwezig is. Eventuele semi-permanente bebouwing als eettentjes of terrasoverkappingen zouden in gevaar kunnen komen, maar zijn indien daarop ontworpen ook te verwijderen voor een hurricane. Ramen en deuren op de begane grond behoeven wellicht extra bescherming. Door de gebouwen boven het niveau van de te verwachten waterstand op het land te bouwen neemt de kans op schade door overstroming ook behoorlijk af.

On the morning of August 29, 2005, Hurricane Katrina struck New Orleans. A storm of Katrinas strength and intensity is expected to cause major flooding but the main portion of the destruction was caused by its exposure to failures at almost all levels of responsibility. In Katrinas wake, a Grand Plan for Louisiana is introduced by the Louisiana State University. In this thesis, the confluence of the Mississippi River Gulf Outlet (MRGO) and Gulf Intracoastal Waterway (GIWW) is analyzed. It is recommended to implement two flood protection structures in combination with a straight levee alignment between them. This option has the best high water safety and most positive ecological influence. It is concluded that closing the MRGO is the favorable option for the future development. The GIWW is highly navigated and should remain open A vertical lifting gate is chosen as optimal flood protection structure. Its hoisting system is complex but well known, providing a reliable closure system. It can also safely discharge excess water, which represents a main functional requirement. Quantifying time dependent processes leads to conclude that the optimal combination is a gate height of +30 ft (10.6 m, MSL) and leakage width of 0.05 m at the sides and 0.15 m at the bottom. This has the benefit that no perfect closure is needed during a storm event, which reduces the need for maintenance and increases the reliability of the closing process. In the design of the steel lifting gate reference is made to the Hartel Canal Barrier, a proven protection scheme in the Netherlands. The total mass of the gate amounts 1060 tonnes. The main contributors are the bottom and center lens-shaped barrier sections. The front plate forms the largest single contributor at 20% of the total mass of the gate. The upper lens-shaped barrier section requires only 40% of the steel volume needed for the leading lens-shaped sections and results in a weight reduction of 126 tonnes. Major problems due to the lifting gate vibration in vertical direction are not expected as the system is positively damped and the excitation frequency is sufficiently small compared to the first Eigen frequency.

Master project report. Havana is protected from the sea’s severity by a seawall which is called the Malecón. The study area of this report is section 4 of the Malecón which more or less matches the coast of the district ‘Centro Habana’. Since the construction of the Malecón the hinterland has frequently been flooded due to storms and hurricanes in the Gulf of Mexico and on the Atlantic Ocean. Recent inundations were caused by hurricanes Michelle (2001) and Wilma (2005). Among the damaged buildings was the Almejeiras hospital, one of Cuba’s most important. The inundations are mainly caused by waves overtopping the seawall. In the present situation the seawall is 4.3 m above MSL and during design conditions 274 litres of water will be discharged per metre over the seawall on time average. Other contributing mechanisms are rainfall and wave penetration into the drainage system, which are in the order of 10 l/s/m. Because the overtopping is the most important cause the main objective of the report is to find feasible alternatives to drastically decrease the overtopping. A solution has to be integrated with the drainage systems to prevent build up of the water mass in the hinterland. First research was done on the causes. With the findings a list of conditions and a list of demands for the solution was formed. Directly after, all sorts of solutions were considered and weighed in a Multi Criteria Analysis, resulting in a berm in front of the seawall and a submerged detached breakwater as the best alternatives. These have been investigated further. Both the berm and breakwater are considered feasible. The berm must be made of rubble with a nominal diameter for the armour layer of 1.0 m. At the crest (MSL + 2 m) the berm is 6 m wide and the bottom ends about 30 m seaward of the wall. Unfavourable aspects of the berm are that it is situated above MSL and that its height makes the permeability performance questionable. The breakwater should be a rubble mound breakwater. A monolithic breakwater proved unfavourable due to the required width and the high transmission. The rubble breakwater is about 90 m off shore. The armour layer consists of 1.5 m rocks, the crest, situated at MSL, is 10 m wide and the height is 7.3 m. The breakwater is more expensive because it requires more material. It matches the list of conditions and demands better and therefore its value is also much higher than the berm’s. The secondary objective of the report is finding hydraulic boundary conditions which are generated by hurricanes. To reach this objective first a literature study on hurricanes was done to gain knowledge of wind generation and wave spectrum generation by wind fields. For this analytical and empirical approaches were investigated. An analysis was done on hurricanes that have past Cuba in the past century. This resulted in parameters that affect the sea severity. The most important parameters are the fetch and the forward speed of the hurricane. From the study on the parameters two worst paths of hurricanes for the Havana coast have been derived.

Student project report, in cooperation with Smith-Warner International Ltd. (SWIL), Kingston, Jamaica. At this moment the shipping channels in Kingston Harbour, Jamaica, slowly accrete. When the harbour is expanded, the local and global sediment transport is likely to change. During this project it is investigated whether these changes are significant and if they will have a negative influence on the Kingston Harbour area. Also the increase of flood risk for the area surrounding Hunts Bay is investigated. This investigation is done by modeling the hydrodynamics of the Kingston Harbour area with MIKE21 and Delft3D, where after both modeling programs are compared to each other. For the input data for the models, research has been done concerning the boundary conditions. This data is gathered from several projects done in the past about other areas in the harbour and fieldwork in Hunts Bay. During the year, most of the wind comes from the east and south-east direction. There are also two mayor streams which debouch into Hunts Bay, namely the Sandy Gully and the Rio Cobre. Since there is only discharge known about the Rio Cobre (daily values from 1985 to 2010), only the Rio Cobre is taken into account. The maximum measured value was 563 m3/s (during hurricane IVAN) and the average value is about 12 m3/s. For the sediment input data some fieldwork is done in Hunts Bay to gather information about the type of soil. From this it is concluded that it is silt, which is confirmed after a lab research of the sediment. However these accurate soil properties couldn’t be implemented into the models due to the lack of time. During the fieldwork also a bathymetric survey was done, which showed that Hunts Bay is sedimented compared to the previously used bathymetric data, gathered from admiralty charts in 2000. Calibration of both models is done by comparing it with the measured water level and flow velocities underneath the Causeway Bridge. Since this is the only point where data was available for, the calibration kept global, and should be improved in the future. The modeling showed that most of the sediment transport into the shipping channel is caused by the high discharge of the Rio Cobre. Ivan showed the most extreme sedimentation and the biggest change due to the expansion. In the present situation the shipping channel is gradually silting, with two areas where the siltation is concentrated. With the first phase expansion these ‘mountainous’ areas will be much more concentrated. However it can be concluded that the changes in the sediment transport due to the first phase expansion are not significant and will not lead to more problems than there are without this expansion. For this problem a sediment trap is proposed. At first it was placed just eastward of the Causeway Bridge, but this didn’t solve the problem and it would be in the way for the phase two expansion. Therefore a sand trap is designed in Hunts Bay, just westward of the Causeway Bridge. This location is really effective, since it stores the sediment from the rivers. This solution prevents the shipping channel to silt. Again, since the lack of reference data, on the size of the pit nothing can be said.

This report provides a comparison of the Dutch Delta Works, New Orleans and the Ike Dike concept. The Ike Dike is a coastal barrier that, when completed, would protect the Houston-Galveston region including Galveston Bay from hurricane storm surge.