Developing a spatial risk profile: assessing building vulnerability to extreme coastal inundation hazard

Williams, Lauren Lyn (2020-12)

Thesis (DPhil)--Stellenbosch University, 2020.

Thesis

ENGLISH ABSTRACT: Coastal zones are dynamic spaces where human activities and infrastructure interface directly with natural forces, particularly extreme weather events such as storm surges. Coastal inundation is regarded as one of the most dangerous and destructive natural hazards, and while there are many studies to analyse these events, few provide assessment techniques relevant to the local context. This research aimed at developing a spatial risk profile for building vulnerability to coastal inundation hazard. GIS was determined to be the most appropriate technology as more sophisticated technologies such as hydrodynamic modelling were found to be limited to specialists,‘data-hungry’ and computationally expensive. An improved GIS based enhanced Bathtub Model (eBTM)was thus developed, which is more appropriate to the local coastal inundation context than the widely used simple Bathtub Model (sBTM). The advantage of the eBTM is that incorporates beach slope and surface roughness and that it instils hydrological connectivity to the coast through embedded cost-distance models. The use of such models thus allows for simplistic hydrodynamic processes such as the water distribution through urban infrastructure to be simulated, the output of which also includes the potential water depth relative to the input elevation model. The model was packaged into a user-friendly Graphical User Interface (GUI) tool and the modelled outputs were further tested and validated against observed data, which supported its applicability. Strand and Fish Hoek (Cape Town, South Africa )were selected as the study sites for which the inundation levels for three independent scenarios were determined by combining the xtreme sea level for a 1-in-100 year storm and two sea level rise scenarios. The risk assessment component used the eBTM to generate inundation hazard maps for the three defined scenarios and to identify affected buildings. Both the hazard and hazard exposure scores were directly dependent on the eBTM outputs in terms of the inundation limits and water depths respectively. The physical building vulnerability indicators were developed through stakeholder engagements. The assessment was undertaken through the use of desktop technologies and on site building inspections. Weighted indicators were used to determine the vulnerability scores for each individual building. Determining the spatial risk profile was based on the scores from the preceding work to determine the risk status of each individual building in each study site. Again, weightings were applied to highlight the importance of components and to reduce the weight of less robust input factors. Overall, in Fish Hoek, one building was found to be at high risk of experiencing coastal inundation based on the given scenarios and three are at low risk. In Strand the spatial risk profile showed that 71 buildings are at low risk, 86 at moderate risk and 35 at high risk. The identified high risk buildings can thus be further assessed in terms of methods to reduce their vulnerability and/or hazard exposure.This study departs from existing regional risk assessment approaches and presents an assessment mechanism that allows the risk components (i.e. hazard, hazard exposure and vulnerability) to be assessed individually, at a locally relevant scale and through their individual assessment frameworks.The eBTM tool and assessment techniques were developed to be transferable to other areas. Furthermore, the eBTM tool enhances the accessibility of GIS based techniques for undertaking localised coastal risk assessments. The maps produced for the individual risk components can thus be used for knowledge transfer, while the final risk maps can be used to inform the management response required relevant to the local context.

AFRIKAANSE OPSOMMING: Kusgebiede is dinamiese ruimtes waar menslike aktiwiteite en infrastruktuur direk met natuurkragte in aanraking kom, veral uiterste weersomstandighede soos stormstuwings. Kusinundasie word beskou as een van die gevaarlikste en vernietigendste natuurlike gevare, en hoewel daar baie studies is wat hierdie aspekte geanaliseer het, is daar min wat assesseringstegnieke binne die plaaslike konteks verskaf. Hierdie navorsing het ten doel gehad om 'n ruimtelike risikoprofiel met betrekking tot die kwesbaarheid vir kusinundasie te ontwikkel. Daar is vasgestel dat GIS die geskikste tegnologie was, aangesien meer gevorderde tegnologieë soos hidrodinamiese modellering beperk is tot spesialiste, 'datahonger' is en rekenaarmatig duur is. 'n Verbeterde GIS-gebaseerde verbeterde badmodel (eBTM) is dus ontwikkel, wat meer geskik is binne die plaaslike kusinundasiekonteks as die eenvoudige badmodel (sBTM) wat algemeen gebuik word. Die voordeel van die eBTM is dat dit die helling van die strand en die oppervlakte-grofheid inkorporeer en dat dit hidrologiese konneksie aan die kus deur middel van ingeboude koste-afstandsmodelle vestig. Die gebruik van sulke modelle maak dit dus moontlik om simboliese hidrodinamiese prosesse soos die verspreiding van water deur stedelike infrastruktuur te simuleer, waarvan die uitset ook die potensiële waterdiepte relatief tot die insethoogte-model insluit. Die model is verpak in 'n gebruikersvriendelike grafiese gebruikers koppelvlak (GUI) instrument en die gemodelleerde uitsette is verder getoets en geverifieer teen waargeneemde data, wat die toepaslikheid daarvan ondersteun. Strand en Vishoek (Kaapstad, Suid-Afrika) is gekies as die studieterreine waarvoor die inundasievlakke vir drie onafhanklike scenarios bepaal is deur die ekstreme seevlak te kombineer vir 'n 1 in 100 jaar storm, asook twee seevlak scenarios. Die risikobepalingskomponent het die eBTM gebruik om kaarte vir inundasie gevaar vir die drie gedefinieerde scenarios te genereer en om geboue wat geaffekteer is, te identifiseer. Beide die tellings vir gevaar en blootstelling aan gevaar was direk afhanklik van die eBTM-uitsette met betrekking tot die inundasie limiete en waterdieptes onderskeidelik. Die aanwysers vir die kwesbaarheid van die fisiese gebou is ontwikkel deur gesprekke met belanghebbendes. Die beoordeling is onderneem deur gebruik te maak van lessenaar tegnologieë en ter plaatse-inspeksies. Geweegde aanwysers is gebruik om die kwesbaarheidstellings vir elke individuele gebou te bepaal. Die bepaling van die ruimtelike risikoprofiel was gebaseer op die tellings van die voorafgaande werk om die risikostatus van elke individuele gebou op elke studieplek te bepaal. Weereens is skalering toegepas om die belangrikheid van komponente te beklemtoon en om die gewig van minder robuuste insetfaktore te verminder. In Vishoek het dit in die algemeen geblyk dat een gebou 'n groot risiko het om kusoorvloei te ervaar op grond van die gegewe scenarios, en drie met 'n lae risiko. In Strand het die ruimtelike risikoprofiel getoon dat 71 geboue 'n lae risiko het, 86 ‘n matige risiko en 35 'n hoë risiko. Die geïdentifiseerde hoërisiko-geboue kan dus verder beoordeel word ten op sigte van metodes om hul kwesbaarheid en / of gevaarblootstelling te verminder. Hierdie studie wyk af van bestaande streeksrisiko-assesseringsbenaderings en bied 'n assesseringsmeganisme aan wat toelaat dat die risikokomponente (d.w.s. gevaar, blootstelling aan gevaar en kwesbaarheid) individueel beoordeel kan word, op 'n plaaslik relevante skaal en deur hul individuele assesseringsraamwerke. Die eBTM-instrument en assesseringstegnieke is ontwikkel om na ander gebiede oorgedra te word. Verder verbeter die eBTM-instrument die toeganklikheid van GIS-gebaseerde tegnieke vir die uitvoering van gelokaliseerde assesserings ten opsigte van kus risiko. Die kaarte wat vir die individuele risiko komponente vervaardig is, kan dus gebruik word vir kennisoordrag, terwyl die finale risikokaarte gebruik kan word om die bestuurs reaksie binne die plaaslike konteks in te lig.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/109099
This item appears in the following collections: