Challenge Fund round 1

Phase 1 (fifteen projects)

Title Team Description
Piloting crowdsourcing enhancement for multi-hazard mobile applications American Red Cross

Access to preliminary data from hazard impacts improves operational decision making, saving critical time in response efforts. Throughout south-east Asia, however, these real-time data are often lacking in regions where disaster risk is highest. To help address this gap, the American Red Cross's Global Disaster Preparedness Center (GDPC) developed a crowdsourcing enhancement for a popular mobile application to support data collection in the region. By self-reporting hazard impacts in real time, citizens can now help ensure that vital data is communicated directly to decision makers to better enable rapid, well-informed response. This project focused on adding enhanced features to pre-existing Red Cross multi-hazard mobile applications to enable smartphone users to assist in quickly and accurately crowdsourcing information about current hazard impacts.

3-Sided Cube, a software development partner, built an algorithm that is triggered by affirmative responses regarding disaster observations, which then sends push notifications to other users within a pre-defined nearby area. The data received is then heat-mapped to visually represent reported impacts, and forwarded to relevant disaster management officials.

During the pilot phase, nearly 14 000 devices were registered and more than 800 responses made, showing strong data collection during power outage, wind damage and flooding scenarios.

Multilingual films for resilience to risks from volcanic hazards University of Bristol

In spite of their catastrophic potential, most volcanoes erupt infrequently, and many have had no eruptions in living memory. Communities near such volcanoes can have increased vulnerability due to their inexperience of volcanic hazards, lack of preparedness and limited knowledge of volcanic hazards and risk.

Teaming with scientists and communication experts, a team from the University of Bristol set out to produce multilingual and multiplatform films to increase resilience to risks from volcanic hazards. With about one tenth of the world's population exposed to volcanic hazards, these films are intended for a global audience.

An initial database of existing, informative films on volcanic risk was produced to better understand how and why certain films became successful, and where improvements could be made. The team also engaged with outreach officers and colleagues involved in the communication of risk to understand how to produce the most effective films.

A modular approach was adopted, comprised of short, individual films designed to provide simple messages, animations, and dramatic imagery to capture people's attention and enable the uptake of key information. These modular films can also be combined to create custom sequences for particular volcanic settings.

Web map services to improve real-time flood data in Africa Dartmouth Flood Observatory

Important flood information is often missing or incomplete in many regions of Africa that experience frequent heavy rainfall events, even while Earth observation satellites are providing useful information.

To help address this shortfall, the Dartmouth Flood Observatory (DFO) created a set of digital products with accurate quantitative maps of long-term flood risks.

The DFO measures, at daily intervals, surface water extent by utilising sensors aboard satellites launched by NASA and the European and Japanese space agencies. This technique of measuring flood extents has been successfully implemented over 15 years, but accessing this data by end users can be cumbersome.

Significant progress of geospatial software now makes it possible to access and visualise geospatial data through traditional web browsers. For this project, the DFO utilised these tools by implementing web-map services through which datasets are easily accessible to decision makers and local planners.

Several organisations active in Africa have shown strong interest:

  • Disaster Risk Reduction Centre
  • Peace Parks Foundation (PPF)
  • Regional Centre for Mapping of Resources for Development (RCMRD)
  • Southern African Development Community (SADC)
  • World Food Program (WFP)
  • World Wide Fund for Nature/World Wildlife Fund (WWF)

These organisations and others can now access Africa-related flood data through a web browser, or ingest data in local GIS systems.

Participatory terrain data and modelling Deltares

As cities across the globe become more vulnerable to devastating floods due to climate change, sea-level rise, soil subsidence, and increased development in flood-prone areas, there is a global need for better information on flood risk and response measures amidst considerable uncertainty. New technologies can simulate water flowing in street-level detail, allowing decision makers to assess which buildings and assets are at risk and what interventions will reduce risk.

In many regions of the world, however, terrain data and exposure data are not accurate enough for these breakthroughs in risk identification techniques. Deltares worked to combine global data with participatory mapping to create high-resolution digital terrain models accurate enough for flood-risk modelling at the street level.

This unprecedented level of precision, available in an open-access format, will allow decision makers to better incorporate flood risk into urban planning as well as efforts to build resilience to climate and disaster risk.

Using Twitter data to map flood risk FloodTags

Thousands of messages per hour are shared on Twitter during flood events. These include early, on-the-ground observations accessible in real-time.

After witnessing an Indonesian NGO manually transcribe tweeted data onto a map to inform disaster-response actions, a team at FloodTags decided to create a better solution to efficiently and effectively capturing risk information generated on social media, to help improve preparedness and response actions.

The Philippines, both disaster-prone and especially active on social media, was selected as a highly relevant context to develop the initiative. Working with local actors like the Philippine Red Cross (PRC), the team combined natural language processing and flood modelling to improve risk information via social media activity, presenting the information in standardised maps, tables, and graphs to effectively support response and preparedness procedures.

Extensive scoping research helped determine how the tool's design could best support PRC's existing preparedness and response measures. To incorporate the expertise of the PRC, the project developed a tool called 'The Relevancer'. Using a supervised machine learning approach, the PRC was able to exclude irrelevant material and add specific labels to Twitter content. The tool then generates and applies new filters to the incoming Twitter feed, ensuring PRC receives information customised to their needs.

Flying sensors for local, ultra-high-resolution flood-risk identification FutureWater

Mozambique's major rivers flood almost annually, with devastating impacts for neighbouring communities. Regulating small dykes and levées is crucial in managing this flood risk, yet critical information needed by groundwater managers is often lacking, including assessments of vulnerable flood mitigation infrastructure. To help local water managers better detect and identify flood risk through improved monitoring of dykes, a team from FutureWater adapted a low-cost, high-resolution Flying Sensor toolkit to Mozambique's unique geographical context. The team's use of aerial photographs allows water managers to anticipate floods by developing an appropriate response plan, take immediate action in case of flooding, and assist in recovery efforts during a flood event.

Local water managers in Beira and Xai-Xai were trained on how to fly the sensor, collect and assess appropriate risk data, and later employ the information in flood management plans. At first, the flight operations were planned on a laptop in a software that required extensive training, however, through the course of the project, the missions were moved to operate on a user-friendly tablet that would better visualise the data. The Flying Sensor toolkits have been handed over to local decision makers, and continue to employed in water management plans.

Local risk knowledge to strengthen community resilience Global Network of Civil Society Organizations for Disaster Reduction

Targeted risk information is critical to enabling effective policies that reduce risk and make communities safer. Through its innovative Frontline program, the Global Network of Civil Society Organizations for Disaster Reduction (GNDR) set out to collect local data on risk and resilience in 14 countries. These data were used to create a global, open source database to allow stakeholders to access and analyze important risk information for effective policy making.

However, an extensive consultation with local civil society organisations (CSOs) revealed that users needed e-learning modules to learn how to better use resilience data to identify priority problems on the ground and to advocate for disaster risk management policies with their governments.

To develop the content and examples within the new tools, GNDR teams worked in close collaboration with 35 local CSOs from 21 countries. After initial testing, a capacity-building workshop was held to strengthen both CSOs' understanding of the e-learning platform, and to enable them to educate other stakeholders on the platform. Within two months, 14 CSOs were planning workshops in their own countries to bring together other members of civil society to use e-learning to better identify local resilience issues and plan evidence-based advocacy initiatives at the national level.

Building resilience to drought in the Sahel region through early risk identification IBIMET

Early-warning systems (EWS) are a fundamental component in preventing and managing food crises in the Sahel. While there are a number of tools and models that allow for near-real-time monitoring of the crop season, those in the Sahel region are often outdated and obsolete, hindering yield gains and putting populations at greater risk of food shortages.

In an effort to address this, the Institute of Biometeorology of the National Research Council (IBIMET-CNR) co-designed and co-developed an open-source crop risk zone (CRZ) monitoring system called 4Crop, along with the national meteorological services of Mali and Niger.

The tool, which uses open-source meteorological satellite datasets, offers decision makers and interested stakeholders a long-term and sustainable solution in accessing risk information. Based on the CRZ agro-meteorological model, the newly built open-source web application allows national EWS to better detect seeding failures, and to monitor state of crops throughout their growth cycle.

The team submitted questionnaires to the targeted users of Niger and Mali meteorological services in order to understand their user experience with current tools and guide the design of the new web infrastructure. The team then tested wireframes and prototypes with specified users, and shared mockups of the web application.

Online operational natural disaster risk assessment (OONDRA) iMMAP in partnership with Kartoza

Developing accurate impact scenarios is essential for disaster preparedness, mitigation, and well-planned disaster response. Running a successful impact scenario, however, requires hazard data, such as water runoff and other flood-specific information, and exposure data, such as population density. A risk model can then predict the location and extent of potential impacts, providing an evidence base to inform targeted risk-reduction policies.

There are several available tools for performing these kinds of impact scenarios, such as the InaSAFE platform, but finding adequate input data is still a challenge for many organisations. A team at iMMAP, along with its partner Kartoza, set out to develop a platform to facilitate online risk assessment, in order to improve access to hazard and exposure data.

An investigation into user needs on the ground was undertaken in Dar es Salaam with various partnering organisations. The team found that, in order to perform disaster scenario planning, organisations first had to obtain local data. The iMMAP/Kartoza team set out to build a platform, tools, and methodology to enable users to more easily find hazard and exposure online data, and bring this information directly into InaSAFE to create fast and accurate impact scenarios.

Real-time urban flood risk data via cellphone network analysis Institut de Recherche pour le Développement

Extreme weather events are increasing in frequency and intensity due to climate change, with floods now impacting typically drought-prone regions like the Sahel. Fast population growth, unplanned urbanisation and the lack of adequate waste water and drainage networks worsen flood risks for millions.

The ability to monitor rainfall with sufficient precision is vital to anticipate these risks, as well as to provide information and tools for the authorities to make timely, potentially life-saving decisions. In high-income countries, rainfall and urban/rural flood monitoring is provided by weather and water agencies with well-developed observation systems like radar. However, this expensive and high-maintenance equipment is often out of reach for low-income countries.

To provide an alternative, innovative, and affordable solution, a team at the Institut de Recherche pour le Développement tested an urban flood protection system in Burkina Faso that measures rainfall via the loss of cellular network signals.

In Africa, where the mobile phone network is dense and growing, this method offers an unprecedented potential for accurate and real-time rainfall monitoring, especially in urban areas where the network is extremely dense. With this new technique, a hydrological model was developed for Ouagadougou city to provide a flood early-warning system for decision makers.

Open-source mobile weather stations for flood resilience International Water Management Institute

Relying only on manual precipitation measurements to manage reservoirs, the Irrigation Department of Sri Lanka was often unprepared for flood waves, resulting in rapid releases, breaches, and increased flood damages. The International Water Management Institute (IWMI) was asked by the Irrigation Department in January 2015 to provide a solution for improving flood resilience.

The department needed data on hourly rainfall intensity so it could prepare reservoirs to receive and store intense rainfall, but traditional weather stations were too expensive for this application. To provide an effective and affordable solution, IWMI developed a low-cost mobile weather station that transmits data through SMS, using open-source hardware and software.

Five stations are being piloted in the catchment of the Nachchaduwa reservoir in Anuradhapura. The operation and maintenance of the stations are being studied to identify avenues for technical and policy development to better support flood risk reduction policies.

The weather stations record data to a micro-SD card every fice minutes, and send a text message to the mobile phones of the reservoir managers daily or whenever a rainfall threshold of 10 mm per hour is exceeded.

Developing open-source, real-time, probabilistic drought-risk visualisation National Institute of Water and Atmospheric Research

Predicting the onset of drought in Small Island States in the Pacific region — with an adequate understanding of antecedent conditions, current observations, and probabilities of near-term forecasts — is a significant challenge for climate experts where resources are limited. This results in a lack of actionable, early-warning information that would allow for proper drought preparations to be activated. For example, there have been several recent cases in the Pacific region where communities have run out of potable drinking water during drought events, endangering lives and hampering recovery efforts. This has also meant that ad hoc emergency supplies have had to be urgently organised, resulting in higher expenses and increased risk for citizens.

A team at New Zealand's National Institute of Water and Atmospheric Research (NIWA) focused on working with national meteorological services in three Pacific countries — Fiji, the Solomon Islands and Samoa — to provide open-source drought-risk monitoring and visualisation tools to help overcome this challenge.

Based on consultations with climate services staff at national meteorological services and disaster management offices, software for new and improved drought information was written using open-source tools and scripting.

Enabling institutions to manage storm-surge risks New York University

Often, vital information never reaches those that need it most, despite recent advances in methods for collecting and analysing hazard data. In addition to accurately assessing risk, governments and other stakeholders are increasingly focused on better communicating risk to communities and development practitioners to increase disaster preparedness and resilience.

In November 2013, Tacloban City was struck by a record magnitude storm surge produced by by Typhoon Haiyan. Despite the weather agency's accurate storm surge prediction, thousands lost their lives.

In response, a team from New York University set out to improve communication practices linking national agencies to local governments and the public. The goal of the project was to identify improved strategies for crafting and communicating risk messages, to develop this into new practices, and to ensure that communities are able to respond quickly to vital risk communication to better protect lives and livelihoods.

The project team combined insights learned from evaluating the experience of Typhoon Haiyan, lessons from the literature on risk and hazard communication, and field surveys to construct guidelines for constructing hazard warning messages.

Multi-hazard school safety in Indonesia through science-based information UNESCO

Natural hazards often damage or destroy school infrastructure, threatening educational opportunities and risking the lives of schoolchildren. In collaboration with the University of Udine, a team from the United Nations Educational, Scientific, and Cultural Organization (UNESCO) set out to create a multi-hazard school safety assessment methodology called VISUS (Visual Inspection for defining the Safety Upgrading Strategies) to help policymakers decide where to focus risk reduction efforts based on available resources and scientific evidence.

The VISUS methodology helps assess schools using a holistic, multi-hazard approach that considers five aspects:

  • site conditions
  • structural performance
  • local structural criticalities
  • non-structural components
  • functional aspects

The VISUS methodology was based on a technical engineering approach that can be used by engineering students and building construction students to assist with local and regional government agencies, to better prioritise safer school programs.

VISUS was used as an effective decision making tool for planning risk mitigation actions at the regional scale, yet remained grounded in local contexts and needs. VISUS aligns with the 'Comprehensive of School Safety Framework', which aims to make all school buildings fully resilient to natural hazards by 2030.

Phase 2 (nine projects)

Title Team Description
Multilingual films for resilience to risks from volcanic hazards University of Bristol

In Phase 1, the team produced hazard and impact films on lahars and pyroclastic flows. The films are available in English, French and Spanish.

For Phase 2, the team are developing six additional hazard and impacts films on lava flow, ash fall and gas. Additional language options will be introduced as appropriate to evaluation needs. The team will also develop one to two experiential films.

Participatory terrain data and modelling Deltares In Phase 1, the team developed a method to derive high-resolution terrain data by fusing data from OpenStreetMap (OSM) and freely available Shuttle Radar Topography Mission data. The combination leads to a very high-resolution terrain datasets in well-mapped urban areas, allowing urban risk managers to identify critical hazardous flood areas. The project was piloted in Tanzania. In Phase 2, the team refined the methodology and the tool was adapted to be more user friendly. In addition, the team worked closely with partners on the ground by conducting training on flood inundation modelling and critical infrastructure impact assessment.
Web map services to improve real-time flood data in Africa DFO

In Phase 1, the firm developed a web-map service and geospatial portal, which visualises high-resolution flood maps for the Africa region with data derived from NASA satellites.

Through Phase 2 funding, DFO will extend the water extent coverage to all continents in the world. The team will move the portal from the proprietary ArcGIS server to the open-source Tethys platform, and enhance the functionality.

Real-time urban flood risk data via cellphone network analysis IRD

In Phase 1, the project has provided a proof of concept that in African countries where the cellular phone network is developing rapidly, rainfall measurement based on cellular networks can provide a cost-effective, reliable and sustainable solution to improve weather/climate monitoring for risk assessment. The project was first piloted in Ouagadougou, Burkina Faso.

In Phase 2, the team will carry out a vulnerability assessment in Ouagadougou to understand how the public can better act on the early-warning information derived from the real-time urban flood risk data via cellphone network analysis tool. The methodology will also be replicated in Niamey, Niger.

Open-source mobile weather stations for flood resilience IWMI In Phase 1, the team developed the first of its kind open-source mobile weather station, and the stations were installed in the upper catchment of the Nachchaduwa Reservoir in Sri Lanka. In Phase 2, the team will be further improving the weather station for better accuracy and reliability, and establishing a community of practice for users of open-source weather station data to share experiences on costs, operations and maintenance with a local online knowledge-sharing platform. The team will also support national stakeholders to repair and maintain devices, and to pursue policy for use at both national and international scales.
Developing an open-source, real-time, probabilistic drought risk visualisation NIWA

In Phase 1, the team developed a drought risk visualisation toolkit (DRVT) for three Pacific Island countries: Samoa, the Solomon Islands and Fiji. The objective of the tool is to assist national meteorological services provide better and more timely advice and early warnings to drought-sensitive communities, government agencies, and other key clients in the Pacific.

In Phase 2, NIWA are developing and extending the DRVT to Kiribati, Vanuatu and Papua New Guinea, with an additional focus on uptake and end-user pre drought contingency and action planning

Enabling institutions to manage storm-surge risks NYU

In Phase 1, the team developed a storm-surge risk communication tool. The tool is a template for storm message design, including guidelines backed by extensive research.

In Phase 2, the team are refining the communication toolkit and workshop design, piloting the communication toolkit in the Philippines and Bangladesh, developing a film capturing narratives from storm-surge survivors, and building an online tutorial and learning platform.

Using Twitter data to map flood risk FloodTags Through further funding from the Belgian Building resilience through innovation and open data in sub-Saharan Africa project, the team are scaling the Using Twitter to map flood risk project in Tanzania. The team are also developing a flood impact warning tool to better communicate the hydro-meteorological warning provided by the Tanzania Met to the early responders and the public.
Multi-hazard school safety in Indonesia through science-based information UNESCO

Through further funding from the Belgian Building resilience through innovation and open data in sub-Saharan Africa project, the team are scaling the VISUS methodology in Mozambique.

The project's main objective is to strengthen the safety of students and teaching staff by providing decision makers at the Ministry of Education and Human Development and the National Institute of Disaster Management, as well as the different regional and local authorities, with tools and information to make science-based decisions on where and how to invest their available resources. Ine hundred schools in Inhambane, Gaza and Maputo provinces will be surveyed.