The integrated web-based flood risk management system for Dhaka that projects risk and damage under climate change, and help with planning.
This proposal uses the city of Dhaka, the capital of Bangladesh, as a test case to design a flood risk management system and build it on a digital interface that can be used in a simple fashion by anybody. Bangladesh is positioned at the junction where three rivers join. The country and the capital city has a long history of flooding. They could use this tool to their benefit in many ways. The tool will allow the government agencies, planners, financial industry, residents and the other stakeholders identify the locations riskiest to flooding (both high hazard exposure and high vulnerability), and evaluate the cost of a potential mitigation project against the cost of flooding damage. All the functions of address searching, data visualization, interactive mapping and online calculation will be done through a user-friendly interface.
Map 1 Bangladesh flood zone
Map 2 Regional Rivers surrounding Dhaka City – Institute for Water Management
Climate.IQ has been focusing on flood and climate change impact modeling, big data analytics, visualization, and software building for two years. Climate.IQ® develops location-based climate risk models and maps vulnerable property and infrastructure on a simple to use interface. The team has developed a beta product that allows New York City to understand the level of its risk to flooding and hurricanes on a property by property basis. Climate.IQ has mapped over 800,000 properties in the city and categorized each one to deliver a risk score, flood depth and duration now and in the future based on scientific modeling. It has developed a method and an interface to show a user the estimated cost of rehabilitation or replacement for each property based on the flood level and likely damage to occur. To read more, please visit www.climate-iq.com
What actions do you propose?
We propose to build on the methods already developed for New York City by Climate.IQ and focus on Old Dhaka as a test area. The City of Dhaka is projected to have a population of 22m by 2050.
1. The city inhabits a vast low-land area in the deltas of the Ganges, Brahmaputra, and Meghna Basin making it a high risk flood zone.
2. Urban flooding from intense monsoon rainfall is a recurring phenomenon that adversely affects citizens and livelihoods to the extreme.
3. Projections by the IPCC and the World Meteorological Organization suggest that much more erratic monsoon rainfall could happen in the future.
4. The World Bank (WB), specialists and government agencies of Bangladesh have completed some thorough studies that include hydraulic modeling, potential physical damage to property & infrastructure and specific engineering upgrades to the drainage systems. So, the problem is well understood.
5. The city of Dhaka, with support from WB specialists have quantified the number of people that could be harmed against the future monsoon weather that could occur in the future out 25, 50 and 100 years which is helpful.
What is lacking is a coordinated flood information and response tool that Climate.IQ can design and build.
1. The city does not have a comprehensive early warning system that needs to include evacuation procedures and locations, rescue steps and locations and first aid locations; Climate.IQ can develops this; why? Because its tool is able to collect real-time data and install in one place to enable a user to access information that is updated hourly, daily, weekly with results reported by quarter.
2. Prevention requires spatial planning, flood control, preparedness, insurance requirements to measure & price the premiums for flood insurance, and lastly evacuation plans. These components are explained as a requirement for a flood risk plan and some information has been gathered and analyzed in detail. However, we see no method that integrates all the data that can be used by stakeholders on a comprehensive, web-based platform. Prevention tools exist; however, they do not contain all the relevant information that can also alert different authorities and civil society in a simple way.
Criteria for Large City Flood Risk Management and Measurement – How Solved?
An integrated flood management design takes into consideration the entire risk system and the interaction of each risk including the uncertainties. The method and the information generated can support an early warning process and also provide the necessary details on a city’s assets and overall vulnerability of its physical infrastructure and its population. It is an all-encompassing solution that can save lives and livelihoods, if completed in a comprehensive, web-enabled way. It can be used to respond to flood risk, hazards and vulnerability. It needs to take into account climate change and land use change. A traditional defensive flood risk “pathway” approach is an engineering solution commonly adopted that relies on analyzing an extreme meteorological event; It is more straightforward and includes, source, (rainfall, wind, wave, sea-level rise); Pathway (river catchment), Receptor (People, property) and negative consequences (loss of life, economic damage and pollution).
What technical aspects are involved with an Integrated Flood Risk Management System Design?
The components of an integrated flood management system are risk analysis, risk evaluation and risk reduction; solving the problem, lies in the actual way that relevant information is collected and presented to all stakeholders that takes full advantage of current technology, data sources and models that incorporate well-advanced background information. It is now widely accepted as the state-of-art methodology that also can include mitigation and adaption. The Thames Estuary for London has adopted this method and it is currently under development, which is very exciting.
The 2 main aspects to an early warning flood management system are further explained in Figure 1 below. Each aspect is described in terms of what it means and how Climate.IQ achieves it.
Figure 1. Basic framework of flood risk management (Schanze 2009) CIQ Table – MS
Flood Risk Analysis
This provides information on previous, current and future flood risks involving 2 tasks; hazard and vulnerability analysis = flood risk maps. Climate.IQ has generated these maps for NYC at 30m resolution and the methods employed can be translated to Dhaka.
The first step of the analysis is to conduct a flood hazard exposure assessment at a refined resolution (3 m - 100 m depending on the resolution of the remote sensing data). This is a spatial and statistical assessment. Exposure is defined as the likelihood of risks and stressors associated with or exacerbated by climate change or other hazards that could have repercussions for a community. In the case of flooding, Climate.IQ examines the hazard by different causes, including riverine flooding, coastal flooding and urban flooding. We apply hydraulic and hydrodynamic models to simulate flow characteristics in the investigated area by balancing the input and output variables of precipitation evaporation and discharges. We also consider spatial high resolution information about the terrain. The models have already been designed and applied to NYC area, which are briefly discussed below. We would need to obtain the data for Dhaka.
For each cause, the analysis follows similar steps:
1. Data input design as the drive for the modeling (i.e. precipitation, wind, river level rise, tidal level, etc.)
2. Model the hazard outcomes (i.e. water depth, flooding days, inundation area) to each calculation grid and validate the results based on historical analyses of previous flooding-driven damage.
3. Project extremes in weather data of different return periods based on the climate change projection as input to feed in the model for anticipatory flooding results in the future.
4. Conduct simulations to generate an ensemble of scenarios.
5. Design the exposure score.
The team models the water depth at each modeling grid for historical flooding events. By inputting historical hourly precipitation data into the precipitation-runoff model, Climate.IQ delineates the floodplain inundation area and the water depth. Then, we consider the downscaled future precipitation data generated from climate models, which is available at CMIP5 site. This data will be resampled and fed into our validated model to project future flooding events and water depth. We will fit a plausible probability distribution curve to the data and run simulations. Eventually, the probability and water depth of flooding in each model grid cell for future years can be estimated to account for the impact of climate change.
To consider the coastal flooding risk exacerbated by future sea level rise, we take the range of sea level rise magnitudes for 2050 and 2100 available from the stations along the coast. Then the historical water level data will be obtained from stations and estimated for a particular recurrence interval. All the station data will be layered with grid units and interpolated along the coastal line. A plausible probability distribution will be inferred for the sea level rise range based on the triangular function. We then use a Monte Carlo procedure to sample from this distribution and add the resulting sea level values to the water levels.
These water level data will then be used as the boundary forcing in the flooding model to model two-dimensional coastal floodplain inundation area and inundation depth. This yields an ensemble of simulations for a given recurrence interval event, where each member represents a different possible scenario of sea level rise by 2100. The flood extent in each simulation will be weighted and combined to estimate the probability of flooding in each model grid cell due to sea level rise.
We propose a set of indicators that allow the modeling of vulnerability in a data-scarce environment, such as Dhaka. Data variability is critically important for the spatial multi-criteria needs. Available data for Dhaka has been identified to incorporate as vulnerability indicators and includes; points of interest (i.e. amenities, offices, shops, and tourist attractions that are vital for local income generation and livelihood), population distribution, housing and poverty. The more populated and less wealthy the area, the more vulnerable they are to the risk of flooding.
Risky area mapping
Vulnerability data and hazard data is layered together to generate the most risky areas exposed to the danger of flooding. These locations can be prioritized during planning and decision making in the next step, which is risk evaluation and reduction. Climate.IQ maps all the information that will be useful in several ways:
o Raising awareness among people at risk and decision makers.
o Providing information for land-use planning and urban development, investment planning and priority setting.
o Helping to assess the feasibility of structural and non-structural flood control measures.
o Serving as a base for deriving flood insurance premiums.
o Allowing disaster managers to prepare for emergency situations.
Risk Evaluation and Reduction
These aspects of the system are designed to conduct a cost-benefit analysis at the stage of evaluating different risk mitigation options. The ALARP (Figure 2) method illustrates three different scenarios: unacceptable risk region where cost of the risk exceeds benefits in every possible way, the tolerable risk regions where the risk is tolerable if benefits of mitigation are not significantly greater than costs, broadly acceptable region where it is necessary to maintain assurance that risk remain at this level. Therefore, it is important to fully consider the costs (both the of the proposed risk reduction project and of the flooding damage) plus the benefits of the risk reduction project.
Figure 2. Level of risk and ALARP (Floodsite Consortium 2009)
Calculate the cost from flooding damage:
To derive the cost at each grid cell, Climate.IQ uses the land cover-damage analysis method. The land cover spatial data is derived from an automatic classification of Landsat Thematic Mapper data and is available for Dhaka. The land cover data is used to set values for the financial damage that would occur through inundation of that particular cell. These friction and land use values for each land use type are summarized in Figure 2 and demonstrates that the water depth data calculated from the hazard exposure analysis coupled with the land cover data will be used to calculate the damage.
The end output will be the estimated flooding damage cost of different return periods at each grid cell.
Calculate the cost and the benefit of a flood risk mitigation project:
The U.S. Federal Emergency Management Agency (FEMA) has designed a benefit-cost analysis (BCA) data template that could be applied to understand the potential cost and benefit of a proposed risk mitigation project. A user could propose a potential structural or non-structural mitigation project and provide all the quantitative information such as the project useful life (PUL), base year of costs, value of services, after mitigation loss of function and damages after mitigation. All these inputs will be plugged into the system with a detailed algorithm to calculate the cost and benefit of the project. Basically, if the benefits outweigh the costs of the project and the potential costs from flooding events (i.e. damage of the buildings, roads and utilities), then it can be recommended to be implemented.
Figure 3. Flooding hazard and vulnerability analysis against population density in Dhaka.
Figure 4. Damage and cost-benefit analysis of potential risk reduction project.
Who will take these actions?
The majority of the design-build work will be completed by the Climate.IQ team with input from knowledgeable parties and experts.
The primary, initial service that we will provide is the technical framework design of a flood risk management system for 1-3 areas of Dhaka (TBD); to this end, we propose that the focus is on the basic elements necessary in Phase1. Once the project is delivered, we will reach out to the stakeholders of the flooding issue in Dhaka and work with them to see the use of the tool and modify it after receiving field feedback.
Where will these actions be taken?
We would propose to develop the technical framework design from NYC, where we are located. However, engaging with experts in the city of Dhaka to review the design, collect data and receive feedback during later phases will entail ideally, hiring a data-analysis team member that is based in Dhaka. For testing, we would also wish for much of this to be completed face to face with users in Dhaka. Thus, approximately 70% of the team will be in NYC and the rest in Dhaka.
What are other key benefits?
· A comprehensive web platform for an integrated flood risk management system, is the key benefit and application to be produced. It is a helpful tool for its government, citizens and businesses.
· Another benefit under the category of “Prevention” is for banking and the insurance sector to use the platform to gauge and estimate flood damage in $s (Bangladesh Taka) as the metric that provides the baseline for numerous services that need to be provided.
· Early Warning involves predictive analysis that needs to include evacuation procedures, rescue steps and first aid locations. It is combination of real-time data and alerts that can be incorporated into the platform.
· The system shows the areas of the city that have been most affected. With a platform such as proposed, the city have a handle on the number of people displaced and the costs involved to better make evacuation plans.
What are the proposal’s costs?
The technical framework, data collection, development of the algorithms, testing and building the platform for Dhaka could be 8-10 months - allow $300k, that includes expenses and will be based on a final scope of work to be decided against the realities of data availability and time involved to integrate. An allowance of $250K is allocated towards additional expenses related to funding agreements, administration and extras related to an entity, such as the WB for approval and administration. Please note that costs are only estimated and can be adjusted based on a comprehensive understanding of the relationship that Climate.IQ will have under a MOU with the City of Dhaka and its funding sources.
Climate.IQ (CIQ) Team subsidized costs - allow $150K (Note that CIQ is prepared to subsidize the team costs so that they align with labor costs established under the WB and the city of Dhaka for purchasing the services from a private company.)
Storage & Software costs – allow $5K
Local and academic specialists - fees & expenses – allow $100K
Testing by Climate.IQ – allow $20K
Contingency for travel and Misc. expenses – allow $30K
An unknown misc. expense for funding allowances under a multilateral agency and or a private hybrid source for administration and expert oversight – allow $350K (Please note that Climate.IQ is a for-profit, private company and its solution could be backed by private sources that in turn, changes the funding dynamics based on the value-add it can provide its users and for what.)
Note: A full description of all costs against a timeline will be provided at MOU stage and the funding sources and budget can be negotiated dependent upon an agreement on the scope of work to be implemented.
MOU on Scope of Work between the City of Dhaka, Funding Sources for work underway (WB and private sources, such as a partnership with a re-insurance company and or a bank) and Climate.IQ along with preliminary budgets for each phase to be agreed. Allow 3 months.
Detailed Technical Framework Design on data and models, plus building the system working against a baseline of all work achieved to-date, with gaps identified for a product – allow 2 months
Feedback on technical design from modelling experts (academia and WB), Dhaka Institute of Water modeling, and engineering Dhaka-based firms such as (ConDev) with input from the Danish Hydrology Institute (DHI) and others, to be determined – allow 1 months
Data Collection based on agreement to access all sources and in the case whereby the data requires further identification and work to develop – allow 2 months. Note that data collection required, but not available could stall the project for a few months.
Model development – allow 2 months in tandem with data
Feedback and Verification – allow 1 months
Testing with Users – allow 2 months
Climate.IQ has developed a proposal for the U.S. Federal Government, the Dept. of Defense (DoD) how extreme weather events impact energy security.
Climate.IQ has completed a further related proposal for the U.S. Federal Government, under the Small Business Innovation Program (SBIR) and the National Science Foundation (NSF).
Climate.IQ submitted a proposal in 2016 under the Climate CoLab Adaption Competition, titled “Vulnerability Assessment and Adaptation Strategies for Property and Community”. Climate.IQ won in the category of popular choice in the overall competition.
The City of Miami is considering using CIimate.IQ’s predictive analysis product after a demonstration whereby it realized that what the company was generating for New York City could provide their city with preparedness criteria for identifying specific vulnerable properties.
Analysis and Evaluation of the Flood Risk Management Practices in Selected Megacities; A thesis in cooperation with UNDP GRIP by Meiling Li. University of Dresden 2013
Acceptable levels of risk and the ALARP principle, Source: Adapted from FLOODsite; Integrated Flood Risk Analysis and Management Methodologies ‘Language of Risk.’ March 2005
Cities and Flooding: A Guide to Integrated Urban Flood Risk Management for the 21st Century; Abhas K Jha | Robin Bloch Jessica Lamond. World Bank – GFDRR
Urban Flooding of Greater Dhaka in a Changing Climate Building Local Resilience to Disaster Risk; Susmita Dasgupta, Asif Zaman, Subhendu Roy, Mainul Huq, Sarwar Jahan, and Ainun Nishat; The World Bank Group 2012
Zaman, Asif M. 2014. Urban Flooding of Greater Dhaka Area in a Changing Climate: Vulnerability, Adaptation and Potential Costs. Background report prepared for this study, Bangladesh Climate Change Resilience Fund and Institute of Water Modelling, Dhaka
DWASA (Dhaka Water Supply and Sewerage Authority). 2014. “Preparation/Updating of Storm water Drainage Master Plan for Dhaka City.” Prepared by Jurutera Perunding Zaaba Sdn. Bhd. (Malaysia), Farhat Consulting Engineers and Architects Limited, and SARM Associates Limited for the World Bank–funded Dhaka Water Supply and Sanitation Project, Dhaka.
IWM (Institute of Water Modelling). 2012. Water Supply Master Plan for Dhaka City. Dhaka: Institute of Water Modelling.
Sugiyama, M. 2012. “Study on Future Climate Scenarios for Asian Cities.” Prepared for the World Bank.
UNESCO (United Nations Educational, Scientific and Cultural Organization). 2012. “Flood Vulnerability Indices.”
Huq, Mainul. 2014. Cost of Adaptation to a Changing Climate. Background report. World Bank, Washington, DC.
Research paper; After the Bangladesh Flood Action Plan: Looking to the future; Hugh Brammer, ENVIRONMENTAL HAZARDS 9 (2010)www.southasia.ox.ac.uk
Flood Action Plan of Bangladesh: Megan Visk, Casey Leslie, Kristian Nikolic, Gary Barkauski warpo.gov.bd/fap/ May 5, 2014.
Management of Urban Floods in Dhaka City; article Global Water Partnership of SE Asia, September 2012
Early warning Early action: © International Federation of Red Cross and Red Crescent Societies www.ifrc.org/Global/Publications/disasters/ew-ea-2008.pdf
Global early warning systems for natural hazards: systematic and people-centred; BY REID BASHER* UN-ISDR Platform for the Promotion of Early Warning (PPEW), Goerresstrasse 30, 53113 Bonn, Germany
Understanding the Economics of Flood Risk Reduction; KATE HAWLEY, MARCUS MOENCH AND LEA SABBAG; Institute for Social and Environmental Transition-International Boulder, CO USA
Chen, A. S., Hammond, M. J., Djordjević, S., Butler, D., Khan, D. M., & Veerbeek, W. (2016). From hazard to impact: flood damage assessment tools for mega cities. Natural Hazards, 82(2), 857-890. doi:10.1007/s11069-016-2223-2
US, Federal Emergency Management Agency. (2009). Final BCA Reference Guide. Washington, DC: FEMA.