The design and development of Geoinformation Flood Simulation Program 1 (GFSP-1) flood hydrodynamic model
Ugonna C. Nkwunonwo
Flood risk management (FRM) is the science that is motivated by the inexorability of flooding. Following this notion, the idea of "living with floods rather than fighting them" was contrived - all credits to the United Nations International Strategy on Disaster Reduction (UNISDR). Interestingly, much knowledge has evolved constantly over the years in this area of scientific enquiry. Previous topical and thematic debates are no sooner proposed than they are superseded. This demonstrates the unparalleled significance of meeting the challenges of flooding within the human societies. Despite the critical objectives of FRM and the much achievement recorded over the years, the level of awareness of flooding in especially the developing countries (DCs) has been considerably poor. The issue of data paucity to address flooding on the basis of best practices still lingers. More scientific procedures, such as flood modelling is lacking. This is the critical foundation for the research which is being described here.
This monograph describes the design and development of a novel flood model, GFSP-1, (Geoinformation Flood Simulation Program-1) designed to meet the challenges of flood modelling in the DCs and data poor urban localities. This is the key aspect of a PhD research conducted at the university of Portsmouth, United Kingdom. GFSP-1 combined two conceptual parts - Cellular Automata (CA) and Semi-Implicit Finite Difference Scheme (SIFDS), and required only a 2-m horizontal resolution airborne Light Detection and Ranging Digital Elevation Model (LiDAR DEM), Manning's friction coefficient, and a rainfall intensity value to simulate urban flood hydrodynamics. This is a significant contribution to the science of flood modelling, whist GFSP-1 attempts to complement existing flood models and thus addressed some key limitation such as requirement of elaborate datasets, limited model external calibration, copyright restrictions and model extensive computation cost which often prescribes high end computers.
GFSP-1 was tested and validated in Portsmouth, United Kingdom, using a severe flooding event that occurred on September 15th 2000. Simulation of various spatial and temporal scenarios for the July 11th 2011 flooding in Lagos Nigeria was also carried out. These events were chosen since map of hotspots of surface water flooding and social media-based information, especially photographic images of the events, were available to enable a rigorous validation of the new model. In both of the test cases, GFSP-1 simulated flooding at locations similar to those depicted by the map of hotspots of surface water flooding in Portsmouth, and identified during the reconnaissance survey in Lagos. Simulated maximum flood water depths from ten sampled locations in Portsmouth and six in Lagos compared well with estimated maximum flood water depths. The Pearson correlation coefficient (r) between model predictions and estimated values is 0.986 for Portsmouth, and 0.968 for Lagos. This indicates optimal performance for the new model in terms of reconstructing the characteristics of urban flooding. Additionally, the plots of water depth vs. time which produce a smooth curve throughout the simulation, and the short time spent in the simulation show that the model's outputs are unconditionally stable, and inexpensive from a computational point of view. These are major issues of considerations in flood modelling research.
The challenges of flooding in the DCs will continue unabated unless significant improvements are made on current flood risk policy and management efforts. This will necessitate evolving new measures, by which the urgent needs to protect human lives and economic infrastructure in the DCs outweigh considerations for uncertainty and standardisation in FRM. These new measures will consider the critical understanding of the dynamics of flooding, and the factors that influence increasing vulnerability to the hazard in the DCs. While such understanding is underpinned by provision of data and mapping of flood hazard and risk, considering climate change scenarios, how to maximise the potentials within presently available datasets in the DCs should be explored as a major research opportunity. The research presented in this monograph explores this opportunity, and, through its objectives and findings, provides flood hazard underpinnings, as well as makes significant contributions to knowledge in the area of ameliorating the impacts of flooding in the DCs and data poor urban centers. It is fundamental to innovative FRM policy and practice within these areas, as well as to strengthen existing flood risk adaptation efforts.
In preparing this monograph, which presents a key aspect of the doctoral research conducted at the university of Portsmouth United Kingdom, the author feels strongly overwhelmed by a litany of assistance and kind gestures from all and sundry. Of course the thesis which reports a whole spectrum of the doctoral research contains a full section of acknowledgements for agencies, organisations, and countless individuals, most of whom are friends, families and academics of global reputation, to whom the doctoral thesis in particular, and by extension this monograph owe their existence. However, there has never been any overindulgence in repeating 'Thank you'.
Therefore, I would like to acknowledge again the doctoral supervisory team - Dr. Malcolm Whitworth, Dr. Brian Baily, and Dr. Robert Inkpen. The Tertiary Education Trust Fund (TETFUND) deserves a special acknowledgement for funding the doctoral study. Professor Vincenzo Casulli, whose seminal work on Semi Implicit Finite Difference Scheme (SIFDS) has been very useful to the completion of the flood modelling aspect of the doctoral study is acknowledged. I thank the two Vice-chancellors, Professors Bartho Okolo and Benjamin Ozumba, of the University of Nigeria Nsukka (UNN). The former approved my application for study leave with pay while the latter granted extensions on the original approval. I thank all my colleagues at the department of Geoinformatics and Surveying UNEC, for their camaraderie and cooperation. Thanks to Dr. Elijah Ebinne, the current HOD, whose permission and candid suggestion led to the actualization of the monograph.
Whilst it is hard to accommodate an endless list of acknowledgements within a lone page, a quick "THANK YOU" to all whose jokes, personal interest and valuable hints were instrumental to the writing of this monograph. While this effort could not have been possible without the stunning support of many, any shortcomings, mistakes or omissions are all mine.
To read further you can click on Download link.