OpenDRI Field Guide

1. Preface

In the year 869, the residents of Miyato-jima (Japan) felt a strong earthquake. Knowing that a tsunami might follow, many fled to the top of the nearest hill. Those who had fled in another direction watched in horror as the first wave from the sea combined with a second wave that came over the rice paddies, washing over the hill and pulling the helpless villagers out to sea. Those that survived that day constructed a monument on the hill, telling the story of how two tsunami waves fused together. For 1142 years, the shrine served as a warning future generations not to flee to the top of that particular hill. When the 11 March 2011 earthquake struck off the coast of Sendai, residents of Miyato-jima remembered the story and fled further inland, where they watched two tsunami waves crash over the shrine. (ref LATimes, Japan’s 1,000-year-old warning).

This story is an outlier—an example of collective memory reaching across 50 generations. In contrast, humans tend to remember how to plan for events that happen frequently with relatively low intensity, like seasonal flooding. Beyond isolated examples like the Miyato-jima shrine, there exists no catalogue of ancient disasters that provides sufficient data to model the periodicity or impact of infrequent events.

There are many cycles in deep history which are invisible to science, because many powerful cycles of nature are not measured in years but millennia. Volcano may erupt in 100, 1000, 10,000, 100,000 or 1,000,000 year cycles. Earthquakes may have similar periodicities. The historical record dates (at most) a few thousand years. Accurate scientific measurements of such low-frequency, high-intensity events are sparse and spotty until only the past few decades.

At the same time, the underlying causes of risk are changing. While the world is becoming more interconnected and interdependent, new challenges are making it less likely that wisdom received from previous generations will prudently inform decisions about the present and future (Suarez, 2013). From urbanization and climate change to population growth and technology, emerging risks are altering traditional approaches to disaster risk management. Among the large cities exposed to cyclones and earthquakes, the population will double from 2000 to 2050 (0.68 billion to 1.5 billion) (NHUD, 2012). At the same time, these cities will create more concrete and asphalt–making them less permeable and more liable to flood. Models of climate change estimate that damage from cyclones will increase 50-125% over the 21st century. Climate change is unfolding at an accelerating pace–a rate of change that will make prediction always behind. Such patterns of natural hazards require rethinking current understandings of assessing and managing risk.

Introduction to Understanding Risk and Risk Information

Understanding dynamic risks requires better information. Currently, many governments have very limited information about their exposure to natural hazards, and therefore cannot effectively manage their vulnerabilities. In many countries, base maps have not been updated in 50 or more years, the census may be more than 10 years old, and there may be little data to distinguish those buildings that have been built to withstand natural hazards and those that have not.

Without data, vulnerable societies cannot make different choices. They cannot prepare for expected failures or make reasonable (often low-cost) investments in mitigating risks. They cannot enforce building codes and zoning restrictions that create resilience in the face of expected adversity. They cannot compare their historical understandings of a risk (inherited with the wisdom of their society) with new understandings of a dynamic planet, where climate risks are changing historical understandings and opening them to hazards that they have not previously faced.

Understanding New Risks Requires New Approaches

Within this changing context, GFDRR is taking a new approach: It is rare for the leaders in these communities to have the data they need to see the invisible risks, hidden in history or behind concrete.

Data is not enough

That said, having data is not enough to create changes to behavior. To alter mental model of the risks across a population, data needs to be available to everyone, and knowledge of how to analyze and apply those data need to be widespread. Data need to be open–legally open, in terms of intellectual property licenses that permit them to be reused, repurposed, and redistributed without cost, as well as technically open, so that any software can open them, manipulate them, and save new analyses in open formats. More importantly, the data needs to be collected, analyzed, and curated by the people in the place facing the risks. Only through this process of having the data be available to all and curated by those who are potentially affected can behavior fully change.

Open Data for Resilience: Building Open Data Ecosystems around Risk Management

In 2011, the Global Facility for Disaster Reduction and Recovery created the Open Data for Resilience Initiative (OpenDRI) to help people in vulnerable regions better understand the historical and changing risks they face from natural hazards. OpenDRI is a partnership of governments and international institutions that are building a deeper understanding of risk by sharing information about their hazards, exposure, vulnerability, and risks.

This guide outlines practices that a network of peers is developing. All of the partners to this guide are working to build better data around the Disaster Risk Management (DRM) cycle. Science agencies are developing better models of earthquakes, cyclones, floods, droughts, tsunamis, volcanic eruptions, and other hazards. Governments are cataloguing the exposure of their built environments to those hazards, at the same time that they are exploring the vulnerability of their populations to the direct and indirect effects of disasters. In parallel, donors are looking at how best to help their clients target investments.

Who is this Guide for?

This guide is for those who wish to apply open data approaches to develop a deeper understanding of how to prepare, reduce, and transfer the potential risks from natural hazards. It is aimed as much to government officials and community leaders as it is to to international organizations, non-governmental organizations, and donors.

Goals of the Guide

This guide intends to:

1. Share disaster risk management activities that have harnessed or promoted open data
2. Explore shared challenges in using open data to increase resilience of societies who are facing risks from natural hazards
3. Start to grow consensus around how to implement open data initiatives inside of client governments
4. Articulate the workflows, partnerships, tools, and practices around OpenDRI
5. Collate shared documents that can be used as templates for future implementations of OpenDRI practices.

A Note on Style of this Guide

The guide is written as an assembly of concise descriptions of practices around OpenDRI. This approach allows for several dynamics:

1. Easy Reading
2. Simplified Translation
3. Multiple Layouts: A3, A4, A5, and web.
4. Easy reference in the form of a growing body of knowledge
5. Simpler version control around a “small pieces, loosely joined” model.

The guide also focuses on cataloguing the set of practices that have emerged over the management of open data, the curation of community maps, and the rise of practices around risk communication that illustrate the potential impact of natural hazards on a given place. Each of these practices is set forth as closed wiki: a module in a larger whole that authorized individuals can edit as they learn and improve upon the original idea of OpenDRI. Overall editorial decision making will remain with GFDRR until a governance structure can be established to perform this function.

Collective Intelligence around Open Data for Resilience

The intention of this guide is to make the work itself a collective effort. It is seeded from the experience of a handful of partners, and shared with communities that work in open data or hope to emulate this approach. Curators will continue to steward the content, ensuring that added knowledge reflects the community’s shared experiences.

---

1. Introduction: The Need for Disaster Risk Management Data

Understanding Risk

Disasters reveal chains of decisions about risk (Natural Hazards, UnNatural Disasters (NHUD), 2011). When infrastructure fails under strain of an earthquake, citizens may point to the failure of the construction firm to adhere to building standards. Or to the failure of a government to set and enforce code around retrofitting a school to seismic risks. Or to the owner of a factory to have inquired into the exposure of the structure to hazards and developed a strategy to cope with these potential vulnerabilities.

In each case, critical information is missing. Information that might have driven a different choice about architectural designs, building materials, or the site for the building (siting). Information that might have driven a community to question choices. Information that might have driven a legislature to pass laws or officials to allocate staff time to enforcing them.

Credible information about risk is an essential element of Disaster Risk Management (DRM). Across the disaster risk management cycle, institutions are now engaged in a process to build this stock of information. The aim is to improve the chain of decision across entire system, from the donor who funds retrofitting of schools to the individual business person who need to mitigate potential losses from natural hazards while caring for his or her household.

---

Risk Assessment and Communication

Building risk assessments follows an internationally agreed approach that combines three elements: hazard, exposure and vulnerability.

• Hazard: The likelihood (probability/chance) and intensity of a potentially destructive natural phenomenon. For example, ground shaking in an earthquake; severe wind/storm surge in a cyclone; inundation level and velocity in a flood event.
• Exposure: The location, attributes and value of assets that are important to communities (people, property, infrastructure, agriculture/industry etc).
• Vulnerability: The likelihood that assets will be damaged or destroyed when exposed to a hazard event. For example, the relationship between the intensity of earthquake shaking correspond and the level of damage for different buildings (Figure 2).

The iterative process of understanding the threat from a given hazard unfolds through several levels of complexity. At the most basic, an analyst can model the impact of hazard estimates what might happen to people, buildings, crops etc from a single event, such as 1/100 year flood or an 8.1M earthquake. This is Impact Modeling.

Impact modeling is not the same as risk assessment. Risk is the composite of the impacts of all potential events (e.g., using looking at the impact from 10,000 cyclone events); this allows an agency to determine the annual average loss and probable maximum loss from individual or multiple hazards. Risk models can be very useful for financial planning and protection, prioritizing DRM investment within a country, and cost-benefit analysis of different risk reduction options. They are the basis for projects that build preparedness, focus risk reduction investment and action, and implement policies that slow the creation of new risks.

Managing Risks Requires Managing Risk Data

Effective disaster risk management requires a commitment to collect, curate, and analyze data over the long term. However, many governments have lacked the resources to be stewards of the data that are essential for risk assessment. Many more lack the capacity to turn the data into models that show the potential impact of a given hazard upon the important elements of their society: people, properties, economies, and the natural environment.

The partners to this Field Guide are working to reduce the costs of collecting and curating data and transforming it into useful models that can guide policies and investments. Several have built non-traditional methods of creating exposure data, which tends to be the most costly to build and maintain.

---

Challenges: Constraints on the application of information to risk management

Most countries lack the resources, training, and software to place hazard and exposure data under a management process that allows for the assessment and mediation of risk. In many nations, the information necessary to catalyze this type of risk management thinking is blocked by a range of problems:

Fragmentation of Specialists

Risk assessment is a multidisciplinary process, but these experts rarely sit in one organization. Specialization has driven the design of modern bureaucracy towards hierarchies. While this structure is efficient for for transactions and coordination of workflows, the flow of information across (and between) organizations can be a challenge. Gatekeepers can prevent the timely flow of information, or may limit it in ways that hinder its use and reuse by others outside the original organization.

Data Fragmentation

Multidisciplinary analysis requires data from across specializations, yet these data are often segmented into silos. They may be in proprietary formats or locked under intellectual property licenses that require expensive payments. Some ministries may charge other parts of their own governments for use of the data, or might even have installed platforms that allow others to access the data but not download it.

Data Duplication

While donors may not set out to fund two or more collections of the same data, the result of having closed data is often just that. One ministry may not know what that other ministries possess or are currently collecting. This problem becomes more acute when NGOs are involved, as communication across partners is often not as good as it might be. Fusing separate datasets may not be possible, or may be very costly, if the groups are using different standards, software, and practices around its collection and quality assurance.

Data Access/Availability

While policies may allow data to be made available, the data curators might limit access or use of the data to specific parties. In this sense, access is discriminatory: it is only for certain approved entities.

Data Staleness and Incompleteness

Data may reflect best knowledge from an investment made more than decade before. In some countries, the last census or high-quality map may be decades old. Data about exposure may have never been collected at the level of resolution necessary to build risk models. Data may also be outdated and/or incomplete.

Exposure Data can be Expensive to Collect and Maintain

Data about buildings and built environment is resource-intensive to collect and maintain (time, cost, personnel, etc). The stock of buildings and infrastructure changes at the rate of construction minus the rate of destruction, mediated by a range of other factors, including the the rapid increase in population and the rate of urbanization. In mathematical terms, the stock of buildings is technically the partial integral of the rate of construction minus the rate of construction for a given time period between t(0) and t(n): ∫(construction_rate - destruction_rate).

---

Policy Challenge: underlying risks are accelerating

The processes of urbanization, population growth, accelerating rates of poor construction and urban planning, and increase in subsidence are changing the nature and magnitude of risks many developing countries. This is particularly pronounced in those countries most at risk of increased cyclones, floods, and droughts in cities with swelling peri-urban slums sited in the most vulnerable areas. For these policy makers, it is become ever more difficult to get a handle on dynamic risks.

That said, even dynamic risk can be mediated and managed.

The levers to manage risk are not always obvious. They may be at level of policy, such as building codes. They may require direct investments in retrofitting infrastructure to higher seismic standards. Or they may require soft infrastructure in a better prepared populace who stockpile supplies because they expect to be cut off from outside aid for a period of several weeks after the next major disaster.

Managing dynamic risks requires higher resolution data

Most governments function with data collected at best annually. In many places, data on the built environment have not been updated in decades, or are collected across a small sample of the country. Land cover estimates may be at greater than 1km or even 5km resolution (average building abstracted from sample across grid square.) There may be policy barriers to collecting the data, particularly where surveys of informal settlements by government officials might create political pressures to turn peri-urban slums into recognized municipalities. Yet, to assess the risk in places facing 5% annual population growth and increased probabilities of droughts, fires, floods, and landslides, governments need higher resolution data, both temporally (collection intervals) and spatially (grid square averages).

But the collection of required components for risk evaluation are no longer activities that most governments can afford to do alone; there is an increasing need for collective action.

To collect higher resolution data in times of economic uncertainty and tight budgets is a difficult choice. Professional surveys of urban areas can be very expensive. Analysis of the data can also be costly. But there is another way: collective action.

In Indonesia, Nepal, and a growing number of countries, governments have been mobilizing their ministries and citizens to collect and curate the data necessary to make everyone safer. Because much of the labor is done by community organization, the resulting maps of built environment are being created at ultra low cost. Working together, members of the private sector, public sector, and community organizations are building a shared understanding of their probable futures. With the aid of risk managers, the government is guiding a conversation about how to invest in resilient communities.

This need for collective action is emerging at time when communications tools and practices are introducing disruptive changes in the methods of coordinating collective action.

When communications are expensive, coordination generally occurs in centralized hierarchies, where decision makers at each level have specific authorities. Since the widespread use of the Internet, the costs of communications have been falling dramatically. In parallel, ever more computing power have been concentrated in devices whose production costs are falling according to Moore’s Law. (sidebar). As a result, billions of citizens of the world have access to low cost communications via handheld devices that allow for data collection, geolocation, and photography. They can organize their operations as swarms, collective intelligences that follow resilient networks of the Internet shaped into human form.

---

Open Data for Resilience: Harnessing Collective Action

Integrating these opportunities and new technologies has become the mission for a two-year old initiative called Open Data for Resilience. This initiative has drawn together international institutions, client governments, and community organizations harnesses collective efforts to create, collate, and analyze data about natural hazards and the risks that they poses to a nation.

These partners include a mix of donors, science agencies, and development institutions. Each has take a slightly different role in the construction of open data for resilience:

(note to each partner: this is a space for you to write a paragraph on your open data work)

Geoscience Australia

Global Facility for Disaster Reduction and Recovery

UNDP

UNICEF

UNISDR

United States Agency for International Development

---

2. Open Data for Resilience

Open Data for Resilience (OpenDRI) works with governments to harness the value of Open Data practices in service of more effective disaster risk management and climate change adaptation. An OpenDRI project offers a menu of DRM tactics for building high-resolution exposure data with collective action, including:

• Collation of data and their publication in an open geodata catalogue. Data about the exposure of a nation to natural hazards are often fragmented across multiple institutions, often under policies that hinder the aggregation of those data into more comprehensive models. GIS and Data Management System (DMS) platforms that enable this kind of aggregation are also rare and (until recently) were very expensive.

• Open Data Working Groups. The development of a community between DRM practitioners is critical for fostering information sharing, providing training, and creating the network of analysts who become the primary users (and sustainers) of risk data.

• Collection of Exposure Data with Participatory Mapping. In many places, there is no geospatial database of the built infrastructure that aggregates key attributes about its vintage, construction materials, elevation, or number of stories. OpenDRI works with communities to build this asset database from the bottom-up using low-cost participatory mapping techniques, where collection and curation of the data is done by the communities that those data describe.

• Software Development around Open Data. Once risk data exists as a public good, it creates a strong incentive to build software applications and services around the application and analysis of those data. Some products are directly related to DRM. Some harness the information–maps, imagery, and data about the built environment–for uses beyond DRM, including navigation, logistics, and business analysis.

• Living Labs. Building a community of technologists and organizers around risk data can be accelerated by using a living lab or innovation lab model, where a place serves as an incubator for new projects.

• Risk Communication Tools. To raise awareness of the potential impact of disasters that have not occurred in living memory, it is important that DRM experts create mechanisms to communicate threats to decision makers and the general public. Risk communication tools provide a platform for this work.

OpenDRI also relies on parallel but separate efforts around improving the modeling of natural hazards, including weather forecasting and mid-term meteorological/climate forecasting. It also relies on cross-support from peers in risk assessment and modeling, who can take data curation that is at the core of OpenDRI and turn that information resource into actionable recommendations around risk management. Note: Risk Assessment is not part of OpenDRI.

History

OpenDRI began from catastrophe. The 2010 earthquake in Haiti killed many of the staff of the national mapping agency (CNIGS) and destroyed a growing knowledge of the geography of Haiti. The building collapse also buried servers with the sole copies of geographic data that would have aided the response and recovery to one of the biggest humanitarian disasters of the last century. Then something unexpected happened.

Several satellite companies collected fresh high-resolution imagery of the damage and made the data available for free. The World Bank collected imagery via aircraft at even higher resolution. More than 600 volunteers of the OpenStreetMap community started tracing the imagery, creating a highly detailed map of Haiti. Volunteers made about 1.2 million edits to the map, performing an estimate year of cartographic work in about 20 days–all for no cost. This effort catalyzed rethinking of community mapping and open data within the World Bank.

The practical reality of what had happened in Haiti fused with a growing movement around open data and open government. Within the GFDRR team, a question emerged: if community mapping could map most of a country in a crisis, what could be done before a disaster? Could GFDRR invest in collecting better data about the exposure of the built environment to natural hazards as a form of technical assistance? Could the communities then curate this data, creating the opportunity for better spatial and temporal resolution of the exposure of a country to threats?

Before the team could set about researching this question, another major disaster emerged: the 2010 famine in the Horn of Africa. GFDRR convened a meeting where partners to the response agreed to share their operational data using a shared data catalogue. This effort—Open Data for the Horn—created a shared catalogue of the various data being collected around the famine, from the Famine Early Warning System to regional maps, geospatial data, and satellite imagery. It has become one of the key points for coordinating activities among OCHA, the World Bank, RCMRD, and WFP. The Sahel Response data catalogue followed soon thereafter.

But the activity at GFDRR was only a small part of a larger movement in open data. In September 2011, Open Government Partnership announced that 8 governments had become the founding signatories on an agreement to make government data far more open to citizens than in the past. Subsequently, 47 more governments signed the declaration. One part of this declaration reads that each government commits to:

We commit to pro-actively provide high-value information, including raw data, in a timely manner, in formats that the public can easily locate, understand and use, and in formats that facilitate reuse... We recognize the importance of open standards to promote civil society access to public data, as well as to facilitate the interoperability of government information systems.

The World Bank itself announced an open data policy in April 2012. All its data and publications would be made available under the Creative Commons 3.0 Attribution License (CC-BY), which permits free reuse and redistribution so long as the data or publication is attributed.

With communities, governments, and international institutions all pursuing open data, the natural next step was to explore packaging open data into a set of approaches around risk.

Packaging Open Data

In 2012, GFDRR began to package these open data efforts under one label: the Open Data for Resilience Initiative (OpenDRI). Teams from GFDRR began to offer World Bank regions and client governments technical assistance around how to use open data to catalyze better information on risks.

These projects centered on applying the principles of open data, open source software, and open standards to the disaster risk management cycle. The objective was to open several types of data for analysis to a wide range of stakeholders:

• Hazard
• Exposure
• Vulnerability
• Risk Information

Where appropriate data did not exist, OpenDRI would catalyze its collection and curation, where possible as open data. Where data was already part of an archive, OpenDRI staff would work to negotiate its release as open data, or establish the appropriate controls on the data with host nation officials. The resulting ecosystem would have far more data on which to base decisions about investing in DRM.

---

Building Partnerships

OpenDRI starts from communities in which it gets implemented. It has connected a wide range of partners:

1. Government Clients
2. Science Agencies
3. Reinsurers
4. Development Partners
5. Local NGOs
6. Voluntary Organizations
7. Incubators/Social Entrepreneurs

---

Work Process Overview

Like open data initiatives, OpenDRI starts small and scales virally. It deploys in one site, and then another, expanding in utility as the amount of data increases. It might start in a smaller city, then migrate to other areas as understanding and trust in the process builds. In this way, OpenDRI is an iterative process.

In general, OpenDRI unfolds in five stages:

1. Scoping: a dialogue to determine the risks, readiness, relationships, and use cases that would form the core of an OpenDRI pilot.
2. Designing: a collaborative, multi-institutional process to customize the OpenDRI package to the unique context of a client government.
3. Piloting: creating an initial presence for OpenDRI, seeding the initiative and building a sustainable community around DRM data.
4. Scaling: expanding the open data ecosystem around the DRM cycle when an OpenDRI implementation sticks.
5. Sustaining: creating the conditions to hand an OpenDRI initiative to the communities that built it.

---

Tools

OpenDRI uses a growing menu of tools to develop the open data ecosystem:

Building a DRM Data Catalogue

Building trust to release data into public or semi-public data catalogues enable the strategic release of certain government datasets to the commons, where they can be curated, emended, amended, and (most importantly) reused in ways that governments alone cannot do. Data catalogues do not mean that a government needs to release of all data to the public.

The collation of links to critical data sets enables first recommendation from NHUD:

First, governments can and should make information more easily accessible. People are often guided in their prevention decisions by information on hazards, yet the seemingly simple act of collecting and providing information is sometimes a struggle. While some countries attempt to collect and archive their hazard data, efforts are generally inconsistent or insufficient. Specifically, there are no universal standards for archiving environmental parameters for defining hazards and related data. Data exchange, hazard analysis, and hazard mapping thus become difficult.

Open data empower decision makers at all levels of government, as well as in the private sector. Open data creates a common space where community can gather around shared problems and co-develop solutions with a wide range of partners.

Principles

For data to serve decision makers across a society, it needs to be fully open. This means:

1. Technically Open: Many government datasets are locked in data formats that can only be read by proprietary software (and sometimes hardware, like obsolete magnetic tape backup drives). The data must be released in ways that allow any device or software can read it.
2. Legally Open: the license under which the data is released must permit redistribution and reuse.
3. Accessible: the data must be available at a public Internet address (URI)
4. Interoperable: the data must follow open standards.
5. Reusable: can be redistributed and reused in ways that were not necessarily anticipated by the curator of the original data.

How OpenDRI works with Governments

OpenDRI advises ministries on the collation, cleansing, and release of data related to risks. These datasets tend to be spread across governments. Sometimes, ministries sell them to each other (though the revenues tend to be low and the administrative/transation costs for managing these sales tend to be high). OpenDRI partners work together to determine which data are appropriate for release. That said, rather than following the traditional method of aggregating data into a central web portal, OpenDRI recognizes that ministries wish to retain stewardship over their own data. So OpenDRI recommends that each ministry release its data using (free and open source) platforms that allow other ministries to subscribe to the data using web services. This model has a number of benefits:

1. Politics: ministries retain control of their own information. Instead of adding a centralized umbrella web portal and the perception of a shift in data ownership, a government adds a free tool into existing workflows.
2. Freshness: the data in the system is always flowing from the source and is as new as the ministry is able to release.

---

Open Data Working Groups

A critical aspect of OpenDRI revolves around the development of an ecosystem of experts who use open data to create analytical products. Hosting open data working groups is a proven tactic to recruit, train, and connect individuals who need better access to DRM information. Working groups provide two critical functions:

• Networking: In many cases, the network of GIS experts, DRM analysts, and disaster managers in a country lack strong relationships. They may also not be aware of the data that other agencies/ministries hold. Open data working groups provide a venue for individuals to become aware of each other’s problems, what data sets they curate and hold, and build sustainable links between institutions. This trust becomes the bedrock on which a sustainable network of DRM professionals curate open data.

• Training: period working group meetings provide a venue where the network of DRM professionals can build capacity around risk assessment. Through partnerships with international institutions, the working groups receive regular contact with DRM experts from around the world, who may present in country or via webinars on topics of interest to the host country. Training sessions also catalyze cross-pollination between countries.

---

Community Mapping of Exposure Data

Building better exposure data is very time intensive, but it need not be costly. It sometimes requires individuals to visit thousands of municipal buildings and locations of critical infrastructure, make a basic assessment about the construction of those sites, take pictures, and ask locals questions about the site. If performed by survey departments of the government or commercial ventures, the costs quickly spiral beyond the means of most governments and donors. In comparison, mapping Kathmandu under OpenDRI cost under $200,000 USD.

The approach taken by OpenDRI is to recruit and train community members to map their own cities. This method creates jobs for youth, trains them in modern geospatial tools, and prepares them for additional work curating the map of their cities. See OpenCities. It also creates a map that is free and open for all to use for any purpose.

OpenStreetMap

OpenStreetMap aims to create a free and open map of the world. Akin to Wikipedia, it allows anyone to draw on “the map” using a wide range of software and devices, including handheld computers and smartphones. To ensure accuracy and data quality, the OpenStreetMap Foundation works with communities in each country to encourage editors and experienced uses to review submissions, and provides software that makes it relatively easy for experienced users to correct the errors of person who has made a mistake or submitted inaccurate data. It is a community managed map.

While some might expect that the accuracy of the map would therefore be far lower than professional cartography, academic studies show that the map is within the margin of error of consumer GPS devices (see Muki Hakley, University College, London in this discussion of the accuracy and reliability of volunteered geographic information)

---

Software Development around Open Data

The release and creation of data under open licenses foments innovation by a host-country’s technology community. In many cases, small companies and non-profit organizations can build revenue models around adding value to open data. A good example is the application of OpenStreetMap data to build routing applications for other business, particularly around the navigation of complex and fast-changing urban environments.

---

Living Labs

To support small businesses and non-profits that emerge around open data, a proven tactic is to use a “living lab” or “innovation lab.” This innovation space provides an incubator where teams from multiple efforts can develop software in a co-working space. Even with basic curation, the environment can foster cross-pollination of ideas and interlinking of both data and business models. The living lab also provides a neutral space where government officials, international staff, local business, and community members can interact as peers and co-develop solutions to shared problems.

---

Risk Communication Tools

To change the mindset of planners at all levels of government, it is not only necessary to give them maps and open up government data; they must also have simple tools that allow them to visualize potential disaster scenarios. Because traditional risk assessment models require a great deal of training and expertise, a range of partners came together to build impact modeling tools that enable a municipal government official to pull hazard data from existing data sources, such as exposure data about a city from OpenStreetMap, and with a few mouse clicks, show the potential impact of a hazard on the schools in the city.

---

3. Scoping

Existing OpenDRI projects have not been standalone efforts, but instead have been components of larger DRM initiatives. In each case, the country context has provide an environment where a network of partners were ready for the challenge of collecting, assessing, and curating data about the built environment.

Assessing the readiness of this network of partners for OpenDRI requires a scoping mission (site visit) that looks at the country content across the four areas that fuse into OpenDRI:

1. Open Government Risk Data. What data does the government collect about natural hazards, the built environment, and other risk factors? How does the government collate, coordinate, and analyze this data? How is this data made available to other parts of the public and private sectors?
2. Open Mapping and Technology. What is the status of the open mapping community in the country? What are the coverage (extent) and accuracy of the mapping data? What kind of open technology community exists to turn mapping data into useful tools?
3. Risk/Impact Modeling and Communication. What kind of risk or impact modeling does the country engage in? Who is responsible for which models? How do risks get communicated to the public and how does the country track how the public uses this information?
4. Hazard Monitoring, Modeling, and Communication. What is the status monitoring of major risks (including weather forecasting)? How does the country communicate those risks?

---

Scoping Mission Objectives

The scoping mission assesses the readiness of government and community to engage in the systematic management of data around the threats that a region faces from natural hazards. In many ways, this process will always need to be customized to the context—it perhaps may always even be bespoke (made to order). For thousands of years, cultures have faced a unique blend of hazards with designs and technologies that reflect local beliefs. With rapid urbanization, globalization, migration, and population growth, many of these traditional ways have come under strain. That said, these approaches are familiar and trusted. OpenDRI missions often begin with problems that reflect the areas where societies are ready to make adaptation, providing a pathway to understanding how new techniques and technologies can be fit to local contexts, beliefs, and practices.

With this need for customization and sensitivity to local context, OpenDRI tends to start from a practical problems that can be solved in the immediate future. The scoping mission looks for the champions||early adopters and works with them to find how possible it is to build the basic feedback loop of OpenDRI using existing data on hazards and exposure, combined with existing models around vulnerability and risk.

In most countries, it will not be possible to aggregate all four types of information from existing datasets. (That said, in countries with sufficient data, the scoping mission team can work with the ministries to develop customized training programs oriented at connecting them with more advanced risk analysis).

GFDRR is developing a diagnostic tool to aid in understanding the readiness of a client country for OpenDRI, focused on assessing each component of the open data process. This tool will complement the existing World Bank Open Data Readiness Assessment Tool, which focused on general aspects of data management versus the more specific DRM data practices.

Staffing the Team

The scoping mission team should be multi-disciplinary, drawn together from the partners that will be sponsoring the project and connected to the contacts/early adopters in the host government. In most situations, the team will include:

• OpenDRI Specialist: surveying the open data ecosystem around the DRM cycle requires experience and expertise. The OpenDRI specialist will build relationships, identify early adopters, and establish the context for an OpenDRI implementation. He or she will lead the design phase and be responsible for building the readiness report.
• Regional/Country DRM Specialist: OpenDRI happens as one component of a larger country and/or regional DRM strategy. The scoping mission should include a DRM specialist who can incorporate elements of the local and regional strategy, and connect local partners to resources from other OpenDRI or DRM/DRR activities.

In an ideal situation, the team would also include:

• Risk Assessment Specialist (optional): some countries may require specialized technical assistance in risk assessment, modeling, and data curation. Such missions should include a risk assessment specialist.

---

What and Why: Defining the Mission and Use Case

The scoping mission team should start by building consensus around the purpose of the trip and determine who will be the primary starting points for seeding the effort. When possible, it is desirable to perform much of this work ahead of travel.

Strategic Intent

The team should explore what outcomes OpenDRI might create in the country context. In some countries, the objective may focus on the aggregation of data that is spread across many organizations. In other countries, the data may not exist at all and may need to be created. Defining the outcomes will focus the interviews and open the opportunity to explore avenues that might otherwise remain unknown or poorly scoped.

Define the Use Case

Mobilizing collective action requires some goal—a problem, use case, or other organizing principle—to focus effort and create a practical, tangible outcome. The objective of the scoping mission is to uncover potential use cases that could mobilize action around open data, both initially as well as in the long term.

---

Who: Create an Initial List of Contacts

OpenDRI generally begins with a request from an official at a government ministry. This champion should be able to guide the Scoping Mission Team on who to talk with. Such entities might include:

Government Ministries

Which government ministries are involved in the DRM cycle? It may be useful to create a matrix of the hazards that the country faces with the datasets that often accompany the study of the

Incubators/Tech Community

Is there a logical place to host OpenDRI? What incubators exist and how well connected are they with the tech community?

Existing OSM Community

What is the state of the existing OSM community and its leaders? What does the map look like? What are their strengths and constraints? How would capacity building change the OSM community?

Universities

Which universities have a geomatics or GIS department? Which have civil/structural engineering?

Civil Society Organizations (CSOs)

What CSOs exist in the areas which need to be mapped? What capacities do they have?

Private Sector

What private sector entities are involved in the collection, curation, and sale of data within the DRM cycle? Some countries contract with outside entities to be stewards over datasets, such as hospitals, schools, and critical infrastructure. What license has the government negotiated for this data?

---

How: Survey the Ecosystem for Opportunities and Constraints

The OpenDRI mission team should set about building connections from the initial list of contacts. The primary purpose is not to sell OpenDRI; it is to listen to problems and think about appropriate solutions, many of which may not be solvable with the OpenDRI approach. OpenDRI fits best in context which are ready for it and where it will address immediate problems.

Identify Champions

The most important factor for success is to find the champions who are already trying to perform this work (sometimes without the authority, resources, or convening power to bring the whole system together on their own) and to LISTEN to their challenges. Often, this work will require toggling between two different modes of thought:

• Inside Government: What factors will drive or constrain the release and integration of existing data that is fragmented across ministries and the organizations that are supporting the ministries or managing their own operations around DRM?
• Development Partners. Within the country’s OpenStreetMap community, local/municipal governments, universities, UN agencies, start-ups, and other development partners, what factors will drive and constrain the collection of new data and the curation of those data by the communities that the data describes? How open is this data?

Determine the Fit

Client ministries and the offices of international organizations/development partners at the regional and country levels set strategic objectives that may include risk assessment, disaster preparedness, mitigation, post-disaster needs assessment (PDNA), and recovery. OpenDRI can fit into each part of the DRM cycle. For the World Bank, regional DRM leads will need to determine if and how an OpenDRI project fit into broader DRM agenda.

The key step in determining if OpenDRI fits into strategic intent is to listen. Listen to the potential partners and the problems they are having managing risks from natural hazards. Where do they want to start? What politics and constraints are they facing and why?

Assess Government Support/Constraints

One of the biggest impediments to opening data is the practice of selling government data. The scoping mission team should ask—are ministries selling data that are core to risk assessment: satellite imagery, maps, demographic data, cadastral data, hydro-met data, etc? What support is there for opening this data? Are there legal constraints or regulatory issues? Privacy issues? If so, are there ways to work through those issues and lawyers/legal advisors who are willing to build solutions instead of putting up roadblocks?

One approach that has worked is collecting and analyzing the revenues against the lost uses of the information for DRM (an opportunity cost). The scoping team may wish to build a spreadsheet, tracking who is selling what data to whom, at what cost, compared to the expenses around the data’s production and the expenses to administer the sales/licensing of the data. While it may seem that data is generating revenues, they are often very small, especially in comparison to the costs of their administration. When the ROI on open data for resilience is higher than the ROI for sales of data, a strong case can be made for ending the sale of certain data sets.

Survey Data Sets

While donors may not fund the collection duplicative datasets (at least not intentionally), entities in country may collect data which already exists. This may happen for a variety of reasons: licensing of existing data may not allow for reuse or derivative works, no one may know of the existence of the data, the data may be of poor quality, etc. The scoping mission should try to find data sets that already exist within government. (note on snowball technique, reference to USAID report on Nepal mission, duplication of open space and building footprint datasets.)

Assess Open Technology and Mapping Community

Another set of tactics around creating open data for resilience centers on harnessing the energy of the open technology and community mapping communities in a client country. By analyzing the efforts of university departments, civil society groups, and the open-source software/community mapping communities, the scoping team can find new avenues for expanding data sets as well as applications that make those data useful to a wide audience.

Community mapping efforts work best when situated in an existing community organization that can act as a sponsor. In places where such organizations exist, the scoping team should explore the readiness and enthusiasm in those organizations for hosting community mapping. In places where no such organizations are ready or willing to help with community mapping, the scoping mission team should look for ways to harness or found an incubator/innovation lab to catalyze the development of both community mapping and open technology communities.

Leadership

The leaders of open technology efforts span a wide variety of leadership styles and capacities. It is critical for the scoping mission team to identify and meet with these leaders to determine if they are aligned with the values of the effort and if they have the leadership and management capacity to administer a complex project. In some cases, leaders may need training in a specific area; the scoping mission team should note this need and plan to provide it as part of the training curriculum in the design phase. (It may be important to think about ensuring that leadership is spread among a number of individuals).

Organizational Capacity

To write a contract to support an effort, it may be legally necessary for the organization to be incorporated or formed into an entity with the authority, governance, and fiduciary structures to commit to performing certain actions and receiving money, and providing a clear account of work performed. The scoping mission team should inquire into the status of each open technology organization. Is it incorporated? Is it capable of receiving funding from the government or international institutions? How capable is it of performing on contracts? What training do members need?

Community Mapping

OpenDRI encourages the collection and curation of data about the built environment to be in open platforms like OpenStreetMap. The scoping team should determine what platform is going to be best in a particularly context. This will entail inquiring into the coverage and accuracy of existing data, while also analyzing the curation capacity.

Assess Risk/Impact Modeling Partners

To determine the data structure for field surveys about the built environment, it is often prudent to build partnerships with structural engineering, architecture, and other risk assessment partners. The scoping mission team should inquire into potential partners and determine the fit with the project. What kind of data do they already have? Will such firms make past, current, and/or future data collections open?

---

Assessing Readiness

When the team arrives back from its mission, it must build a set of recommendations about actions. Is the context ready for OpenDRI? Who will be the primary partners, and what will be the use cases that drive that partnership? Who has invested in DRR/DRM activities and at what funding level?

Convene Open Data Advisory Working Group meetings

It is important to keep the process of engaging and developing relationships among the early adopters between the scoping mission and the design and pilot phases. The scoping mission team should convene meetings of an open data working group. In some cases, this has been done online via voice over IP (VOIP) calls, such as Skype. In many contexts, these meetings are led by a local country representative. These meetings can provide a forum for discussing online issues, exploring ideas, and generating connections between members of the community who might not otherwise know each other.

Outputs

The key outputs from the work after the mission are a mission report and a readiness report.

Mission Report

The Scoping Mission report captures the narrative experience of the team, as well as their analysis of the use case and recommended actions.link to scoping mission report template

Readiness Report

Readiness Report is a diagnostic tool that will be used to assess if a country is ready for OpenDRI. link to readiness report template

---

Malawi

In early 2012, GFDRR received a request to send a scoping mission to Malawi around the Integrated Flood Risk Management Plan for the Shire Basin (IFRMP). The mission had two objectives:

1. Raise awareness about community and data preparedness
2. Solicit input from stakeholders at multiple levels of government around community mapping and a planned data preparedness exercise.

The team traveled to Malawi in April 2012, meeting with the Department of Management Affairs (DODMA), Department of Surveys, Department of Physical Planning, Department of Water Resources, Ministry of Agriculture and Food Security, National Statistical Office, and the Chikhwawa and Nsanje District Civil Protection Committee, as well as UN agencies.

Use Case

Across various interviews with a task force of government officials, the team discovered a need to ensure that data created by a number of past or ongoing projects is maintained in an online platform so that this information remains accessible and useful to the Government of Malawi.

Champions

The GFDRR team found the task force very willing to co-design a data catalogue with the World Bank, and subsequently drafted a Terms of Reference for a firm to develop a GeoNode for Malawi and then train officials in its use through the data preparedness exercise.

Outcome

1. The World Bank gave start-up funding to install a GeoNode as a data catalogue, to be managed by the National Spatial Data Center with the participation of the other members of the Shire River Basin Management Program Technical Taskforce.
2. The task force worked with the firm to collate data from a variety of sources, focusing first on the data from “Economic Vulnerability and Disaster Risk Assessment Study”, “Water Resources Investment Strategy” that was collected for Malawi (http://gfdrr.org/gfdrr/node/148). The result can be viewed at http://www.masdap.mw/.
3. The firm held trainings in partnership with local universities and ministries, including the representatives of departments in the region.

---

4. Design

By expanding the stock of available information about risk, OpenDRI helps its clients target their DRM investments in areas with the highest potential impact. The design of each engagement is highly specific to the client’s context. As a result, the Design phase of OpenDRI starts by listening: reviewing the use cases, concerns, and hopes collected during the Scoping mission. The initial OpenDRI team co-develops a set of practical solutions for collecting and curating the right data for their analytical needs. It focuses on making these solutions immediately useful for the partners and scalable in the local context.

OpenDRI has amassed a set of solutions and is still expanding its repertoire. Other organizations are exploring their own approaches. This Design chapter summarizes an approach for today’s methods, leaving room for expansion and rethinking in the future.

---

Objectives

The core objective of OpenDRI is to build an ecosystem of educated users around a national data management system for disaster risk management.

Unlike many DRM practices, design in OpenDRI unfolds as iterative discovery. It starts with a problem of immediate import to a client’s challenge in disaster risk management. OpenDRI brings those who have direct and specific knowledge of the problem (including various constraints and realities around solutions) into relationship with those who interpret and write policy around data sharing in the country. In the process of exploring the challenge and potential solutions, the teams explore faster, cheaper, and simpler ways to collect data while altering the policy frameworks by which government pursues data sharing. In turn, this process opens new possibilities in the minds of the partners,leading to innovation and catalyzing a new cycle to design a novel approach.

An OpenDRI project will often include both a mix of bottom-up and top-down strategies. The former mobilizes champions and communities around the creation of tangible, practical projects that demonstrate what open data can do. The latter provides the policy guidance and political cover for open data to emerge in a government context. The design of each strategy OpenDRI engagement will generally pull together one or more tactics from a growing toolbox. Each is designed to feed into risk assessment; each requires that its work plan; and each requires the other.

The design for an OpenDRI engagement centers on four activities:

1. Collating. Collating, cleansing, and releasing fragmented datasets via one or more data catalogues, usually hosting on GeoNodes.
2. Collecting data about the exposure of the built environment to natural hazards, usually by developing and sustaining a network of government officials, private sector entities, and community-based organizations that are prepared to curate this data.
3. Creating the policy framework where these data can come together and building the community of practice around the data.
4. Catalyzing efforts around the use of the data for DRM as well as purposes unforeseen by the OpenDRI partners.

This guide is built around these four activities. This first draft will discuss approaches to each activity during each phase of OpenDRI. A second draft will explore structuring the chapters of the guide around the four activities, showing the phases of development inside each activity.

---

1. Collating Existing Data

Once the members of the Open Data Working Group have committed to releasing data, the hard work of obtaining the data begins. In countries like Nepal, data about the built environment is decades old and needs to be collected anew. Any data that does exist need to be put into formats that can be released via an open data platform.

The bottom-up portion of the OpenDRI Design phase mobilizes the champions for open data and provides a framework for them to connect fragmented resources into a more comprehensive, integrated data ecosystem. Two tactics in this approach are data catalogues (usually with GeoNode) and community mapping (usually with OpenStreetMap). The design for each

Designing Data Catalogues

Gathering data from existing archives requires a strategy around aggregating and cataloguing those data. In general, the critical step in this strategy is to implement a data catalogue, which provides a place for all the newly-released data to reside. The design of this data catalogue requires close collaboration between OpenDRI’s implementing partners and the Open Data Working Group. The design of the data catalogue is described in greater detail in the Methods section of the guide: Data Catalogue.

It is worth pulling out some critical points from the Methods section and discussing them here.

1. Approach data collation as a political challenge, not a technology problem. Aggregating data around the DRM cycle is only partly a technical problem. It is mostly a process of building support for the release of existing data and a commitment to collecting new data among a range of actors. The work is mostly about listening to the problems that ministries and partners are having around risk assessment/analysis, then seeking out the underlying problems. The data that ministries need to make decisions may be hidden behind fears about what the data describe (national security), the accuracy and age of the data (data quality), or a lack of knowledge of how to harness the data for decision making (capacity building in risk assessment). Regardless of rationale, partners will often want to know that they are not the first or only organization to release data.
2. Focus on the DRM data, not all data. OpenDRI cannot solve all data sharing problems in a society. It focuses on the issues around the DRM cycle and the datasets that support decision makers around risk assessment, mitigation, preparedness, response, recovery, and reconstruction. The collation of data should center on a) data necessary for solving the use case(s) of early adopters, and b) the data sets that ministries are already prepared to release (low-hanging fruit).
3. Plan on Data Cleansing. Data that arrives from ministries may not always be ready for release or appropriate for OpenDRI. The pilot phase will need to confront data munging/cleansing to make the information immediately useful to others.
4. Keep it simple and practical. The process of cataloguing data often means adding detailed metadata. Developing simple, practical standards that everyone can use consistently will work out better than complex standards that can only be catalogued by experts.

---

2. Collecting New Data

Aggregating data from existing sources rarely creates a usable repository about the current built environment. OpenDRI also develops a network of people who are prepared to collect and (importantly) use data about the exposure of their communities to natural hazards. This approach generally applies techniques from participatory/community mapping The architecture of a community mapping initiative is described in the Methods section of the Field Guide. The design issues here are abstracted from this more detailed method.

Designing Field Data Collection

Collecting data from field environments have special challenges, especially in countries where maps are poor and addresses are often a long narrative than a postal address in the traditional sense. The Design team will need to consider several questions to ensure that the project has a solid field data collection design:

• What data need to be collected to drive which specific analytical problem?
• What other data could be collected alongside these data, without overburdening the system?

The Design team should work with the cross support of risk assessment and DRM experts, comparing data models recommended by this design process with the models from previous OpenDRI engagements.

---

3. Creating the policy environment

Creating the environment where open data can take root is often best done by creating a neutral space where partner to the process can discuss concerns, explore policy options, and review the demonstrations of open data projects from OpenDRI’s bottom-up strategy. This neutral space is usually embedded in an Open Data Working Group. It is also the space from which unexpected synergies emerge.

Open Data Working Group

To address underlying issues that prevent data sharing, it is necessary to confront the technical, policy, and science questions together. GFDRR’s experience points at the importance of establishing a working group around open data. Membership includes champions from various government ministries, international organizations, and community partners.

The goal of the work group including building consensus, addressing underlying issues of sharing, exploring solutions to technical problems preventing data sharing, exchanging stories and best practices, and resolving conflicts. Many problems may not be technical, but may center around licensing, revenues, laws, regulations, and policies. Removing these impediments to data sharing in DRM is only possible by creating consensus among those who need data and control the data.

A key step in the formation of the working group is to get members to commit to opening their data. It is this commitment that provides the necessary policy guidance and political cover for mid-level government officials to perform the practical actions by which data moves from a closed server to an open data catalogue.

It is also important to focus on specific use cases. Working Groups that work on policy questions in the abstract tend to get stuck on edge cases and, as a result, become risk averse. Groups that focus on practical use cases can assess the potential political liabilities around a specific scenario and find solutions that may be imperfect, but get the group started on sharing more data.

---

4. Catalyzing efforts around the use of the data

A critical moment in OpenDRI engagement occurs when new or newly available data gets transformed into an analytical product that tells a risk manager something surprising. It might be a map of which schools are likely going to be destroyed in an earthquake, with a second layer on map showing that the NDMA’s response plan relies heavily on the schools that are most likely to collapse. Catalyzing the ecosystem of thinkers who can probe the data for these linkages and correlations is more difficult that one might think.

Education systems may focus on conveying technical skills and testing competency around those skills. Government ministries may hire competent individuals who have passed these tests, and then place them in a promotion scheme where they are required to perform a task in a bureaucracy to a standard. When education and evaluation align with execution of a task, curiousity can get squeezed out.

OpenDRI works to create an ecosystem where thinkers can probe the data and turn it into products that the design team will be unable to predict. OpenDRI directly invests in the development of software that catalyzes this type of risk thinking (see InaSAFE). OpenDRI also works with university professors, entrepreneurs, and non-profit leaders to foster linkages between curricula and the development of tools that harness high-resolution analysis of the built environment.

---

Who

The design phase is a critical window when the donor and facilitator (e.g., GDDRR) can bring together the network of champions and skeptics into a space where they can explore opportunities and constraints, while also building a different vision for their future of DRM data management. The Design phase is also a process by which various international organizations (donors, development partners, and external experts) can come into alignment on their respective strategies and coordinate their respective projects. In some cases, projects can avoid duplicating data collections and reduce costs for all parties.

Management Team

Management of the Design phase should ensure that key stakeholders are represented in the design of the OpenDRI engagement, including:

• Ministry or Agency responsible for disaster preparation and response
• National Mapping Agency and other agencies with interests in community mapping of the built environment
• Ministry or Agency responsible for natural hazards, meteorology, and land use planning

Management should also ensure that the following areas of expertise are included in creating the architecture for OpenDRI:

• Open Data: expertise in working with governments around open data, particularly around the politics, policies, and laws of the release of data around the DRM cycle.
• OpenDRI Technologies: expertise in the technical mechanisms of collecting, cleansing, and curating open data. This person will specialize in geographic data, including the OpenGeo GeoNode and ESRI GeoPortal.
• Community Mapping: expertise in the implementation of community mapping methods for the collection of exposure data.

Academic Partners

OpenDRI efforts actively engage the partnership of a country’s academic institutions. There are two reasons for this natural fit:

• Students. OpenDRI efforts greatly benefit from students in geomatics/geography/GIS who participate in the design, implementation, and evaluation of community mapping efforts (as well as open government risk data).
• M&E. The evaluation of OpenDRI community mapping data is often best done by professors who can mobilize a research team to ground truth the data collected by volunteers and assess its accuracy in a formal academic paper.

The Design Team should identify academic partners and configure an agreement with them around support to the collection and evaluation of community mapping data. This agreement will usually take the form of a grant to perform research around the OpenDRI effort that culminates in an evaluation of the pilot implementation.

---

Outputs

The Design phase creates an architecture for sharing data. Its goal is to catalyze innovative thinking among the client ministries and development partners, while at the same time, implementing early versions of open-source software to jump=-start open data collection and collation.

For GFDRR, the Design phase creates two types of documents:

1. Concept Note: A consensus concept notes between host government/client, Regional DRM contacts, OpenDRI/GFDRR contacts, and partners to building the open data ecosystem. This note captures the use case and provides a narrative of how the project will collate and collect new data about risks, create the environment for those data to be open, and catalyze the use of that data by those who can change the game around disaster risk management in a country or region.

2. Terms of Reference. The ToRs are the contracts by which the Bank hires expert assistance. The Design team will need to craft ToRs for each individual or firm who will contribute to building the pilot. These ToRs will include:

   • Data Catalogue. GeoNode or other.
   • Community Mapping
   • Incubator/Innovation Lab
   • OpenDRI Consultant Lead
   • OpenDRI Technical Lead

.

---

Sri Lanka

In November 2012, GFDRR sent a consultant to Sri Lanka to scope a potential OpenDRI engagement, based on previous interest in open data expressed earlier in 2012. During a five-week (25 working days) mission to Colombo and Batticloa, the consultant met with ministries and organizations around DRM: Disaster Management Center, Metro Colombo Urban Development Project, Survey Department, Department of Census and Statistics, Nation Building Research Organisation, the Meteorological Office, Information and Communication Technology Agency, UNDP, and several universities and donors.

Use Case

He discovered a situation where use cases centered around data availability for impact assessment and rick communication. In many places, exposure data were not available or very outdated. Because Sri Lanka faces a wide range of natural hazards, he also found that data were fragmented among the ministries that specialize in various aspects of these threats. Ministries and development partners needed a mechanism to share data and to overcome a typical situation: while some officials were willing to share data, others were more reluctant, because their ministries had sold datasets to each other for some time; opening the data would disrupt those revenues.

Recommendations

The consultant worked with GFDRR on a set of recommendations that would enable a parallel bottom-up and top-down approach to data sharing. To build trust around community mapping and data catalogues, the project planned to provide trainings around OpenStreetMap and the GeoNode to build practical, concrete evidence around the effects of collecting and sharing data. To create policy-level support for a open-data strategy, the project would create an Open Disaster Data Advisory Committee. This committee would be charged with establishing guidelines, policies and standards to facilitate data sharing between the government agencies and opening those data to the public. It would be composed of the ministries involved in OpenDRI, plus several community and development partners.

Results

Sri Lanka’s OpenDRI initiative is progressing rapidly, with pilots of the data catalogue and community mapping in both Colombo and Batticloa. The project will also be using InaSAFE as a risk communication tool.

---

{% include JB/setup %}

5. Piloting

The Design team will spend several weeks or even months building consensus among the partners, writing concept notes and terms of reference that require several organizations (and several departments within those organizations) to come to agreement and release funding. Once funding flows, however, work in the field can lead to relatively fast outputs.

The piloting phase collates existing data from government archives, performs the first field data collections, and convenes the stakeholders necessary for developing an environment conducive to open data.

---

Objectives

The primary objective of a pilot is to open the imagination of the partners to what is possible when they share data about risk. The pilot provides a tangible use case to focus the efforts of multiple organizations on building the various parts of an open data ecosystem, some of which may be accessible only inside of government, some of which may be fully public. In some cases, the pilot provides in impetus to get over “last mile” problems, where partners have already made great strides towards building risk data but are prevented from launching their work by a mix of problems that often were discovered only by engaging with the task.

OpenDRI does not require a client country to start a “whole of government” open data program. However, it requires a significant commitment to release data in a narrow band of areas about natural hazards, the built environment, and the threats that disasters pose to a society. The work from the bottom up will build up the proof that this kind of data can be released in open formats on open platforms. There are two parts to this approach:

1. Data Collation and Release. A process with the Open Data Working Group and the GeoNode that centers on obtaining access to existing datasets and then preparing them for release.
2. Data Collection and Release. A process that center on the collection of data about the built environment and its exposure to the natural hazards of the place. Community mapping is one tactic in this area and the main focus of this guide, though other partners will add to these approaches.

Once data is flowing in a pilot, the Open Data Working Group will need to confront policy and management issues around the new repository, including data access rights for certain privileged groups. They will also want to start learning how to apply the data to risk impact models or other analytical products, reinforcing the value of open data and expanding their understanding of why the project is collating and collecting data for DRM.

---

1. Data Collation

The first steps in each OpenDRI engagement has been to create a place where data can be shared and to hire a Data Curator, to manage this collation process. The data catalogue is meant to be a neutral space: a place where the data is legally and technically open for use by all (or at least those who have access to it). In a few cases, this space has solely been OpenStreetMap. In most cases, client ministries wish to have a data catalogue where they can upload datasets that are meant to be opened. The process for building these data catalogues is described in the Methods section.

Data Curator

The Data Curator is generally a technical specialist with GIS training as well as data management background. He or she will be able to perform the following work:

1. Upload spatial data layers into the data catalogue
2. Ensure metadata is entered into the platform in compliance with World Bank GeoSpatial Metadata standards (based on ISO 19115:2003).
3. Style the uploaded datasets using the SLD format to allow for basic visualization on the platform.
4. Create of a suite of basic “mashups” or maps with data made available on the platform in close collaboration with the Open Data Working Group and other stakeholders.

A Terms of Reference is under development for this role.

Collating the First Data Sets

Collating the right data is the key to keeping the project focused on solving problems around the DRM cycle. In general, OpenDRI data will have three attributes:

1. It will be geospatial, with geographic attributes that describe some aspect of the human systems and built environment and the hazards that may affect people and infrastructure.
2. It will be in open data formats and compliant with Open Geospatial Consortium (OGC) standards.
3. It will match the use case of the client.

Collect the data for the use case

Within a single use case, there may be several types of data that need to be aggregated and prepared for release. These include:

1. Imagery

An imagery archive allows for analysis over time. If the archive includes imagery from recent pre- and post-disaster contexts, it may permit analysis of the effectiveness of certain types of interventions. Where possible, imagery should be made available with metadata to describe what the imagery shows.

2. Maps

Mapping data may vary widely in quality and vintage. It is important to get recent data into the GeoNode, less so than an archive.

3. Streets and Critical Infrastructure

Vector data about the streets and major logistics routes are critical during emergencies. Modeling how they would be affected by hazards is a core feature of OpenDRI. These data are needed to make those models.

4. Building Footprint Data

Building footprint data captures the polygons of each structure in the built environment. This data may be protected for various reasons, including potential market value, prior agreements with private firms, and privacy laws. When available, building footprint data may also have issues with resolution, project, accuracy, and vintage. Building attributes (such as number of stories and building materials) are not always present in the data, but a critical for making accurate models.

5. Hazard Maps

To create risk assessment models, analysts require information about local hazards. The OpenDRI team should look for data that describes:

• Floods and Tsunami. For water (hydromet), the resolution of the topolography of the region under analysis becomes critically important. In many parts of the world, the best available digital elevation model (DEM) may be 30 meters in resolution. In flat river valleys where altitude changes very slowly, flood inundation models with 30m DEMs could be wildly off. As a result, it is crucial to assess DEMs available for hydromet analysis.

• Earthquakes. Seismic maps and models that show fault lines, project shakemaps for various epicenters. The GEM may provide the best available data.

• Cyclone Maps. Cyclones are hard to predict, but patterns of cyclonic activity tend to follow pathways over time. (ref for cyclone data?)

• Landslide. Landslide data can be hard to obtain or difficult to get released due to perceptions of commercial value. (research what is needed)

• Droughts. (Will we include droughts?)

Assess Data Quality and Licensing

Data that has been in government archives may derive from a range of sources. It may have been collected by the government, it may have been given to the government by a development partner or UN agency, or it may have been purchased by the government from a private vendor. The Data Curator should ensure that any data referenced in the data catalogue carries the associated sources and licenses associated with the data. GFDRR is build a metadata standards around the GeoNode (see GeoNode guide.).

The Data Curator also will need to assess the quality of the data submitted to the catalogue. In many cases, the vintage of the data may make it more useful for detecting trends than analysis of current risks. The data may have been collected in a geographic projection which makes it difficult to rectify with open standards. Areas where quality needs to be remediated need to be flagged.

Cleanse the Data

All data sets are imperfect; the question comes about balancing the level of effort remediating those inaccuracies with the precision needed for the particular use case. The Data Curator needs to choose which data is highest priority for the time consuming task of data cleansing (or munging, in technical jargon). He or she will likely enlist the assistance of other members of the growing Open Risk Data Working Group, who may have specific domain knowledge or skills around particular types of data cleansing.

Negotiate the release of data

The Data Curator and OpenDRI Lead will spend a great deal more time negotiating the release of data than the System Administrator will spend building the technical infrastructure. Lobbying for the release of information requires having a clear use case that officials can understand and contribute to. It also requires listening to their concerns and addressing those issues with documented cases of how open data has worked in other countries.

Often, fear and insecurity about the data lurks as a hidden issue behind other excuses. Opening data implies loss of control. It also exposes errors that may have been made in the collection and curation of data over time–faults which will only be perpetuated by keeping the data closed. Agencies may be obligated to curate data by law, but may not have the data or may worry that it is so error-filled as to potentially cause legal liability around its release. Licensing and loss of revenues may be an issue, though these tend to be smaller in magnitude than is understood. In each case, the Data Curator and OpenDRI Lead will need to find ways to enable agencies to save face.

Release of the Data Catalogue for Public Use

The audience for the data catalogue generally falls into several classes of individuals:

• Government Data Providers
• Government Data Users
• Private Sector
• Non-governmental Organizations
• General Public

It is important for the OpenDRI pilot team and data catalogue team to develop a strategy around the release of the data catalogue for each population. The release within the smaller DRM ecosystem may require workshops that answer detailed technical questions. The release to the general public may involve more political and public relations concerns about managing perception of the initiative as well as the demand on government officials’ time that open data may create.

Lessons Learned

Opening government data is not always easy, even when the use case is clear and champions inside key ministries want to help. OpenDRI has a growing pool of lessons learned:

Be prepared for less than fully open

Open data may require protections. It may need to have copyrights, acknowledgements, and/or restricted access to make it open. In the early stages of open data, it may be necessary to have some data be available to a limited part of the larger ecosystem.

Invest in release of data rather than perfection of data

A lesson learned from previous engagements is that it is possible to spend inordinate amount of time preparing small percentage of the data, instead of collating data from many sources. In one case, an contractor spent most of his hours fixing the projections of a small number of datasets, instead of building system to aggregate data. Government officials may have a very rational fear of releasing inaccurate data. Their jobs may be on the line. That said, from the perspective the open data effort, it may be better to flag data as needing helping and moving through larger volume than working on specific data and spending consulting hours on big changes to small data sets. OpenDRI Pilot Team staff need to be prepared to balance the need for accuracy with the need for comprehensive releases of data around the DRM cycle. In addition, the team should ensure that any concerns about data quality appear in the metadata.

---

2. Collecting New Data

The first pilot of a field data collection effort initiative generally starts in one city, and potentially, within one neighborhood of a large city. The objective is to train a small pool of energetic mappers into a team that can a) collect the first data and adapt initial approaches to the problems they discover in the field, and b) recruit and train other mappers as the initiative moves into the Scaling phase. OpenDRI has thus far used community mapping as the tactic to implement this objective.

Note: The OpenCities Project has created a toolkit which covers the implementation details of community mapping. This section provides a strategic overview of the OpenCities methodology. A link to this resource will be added when the Toolkit is published.

Team

The team starts with OpenDRI community mapping trainers, who have expertise in the application of community mapping practices to the collection of exposure data, preferably in the local context. The trainers need to ensure that the team builds skills in several areas:

• Mapping and GIS
• Software Development and Graphic Design
• Outreach, training and mobilization

The team consists of several key roles:

Project Lead

The project lead is (ideally) a local national who can work with the trainers to connect the pilot with a network of relationships that they have within government, academia, and the NGO/IO community. The Project Lead manages the operations of the pilot, and therefore must have both management skills as well as a solid understanding of project management, participatory mapping, GIS, communications/social media, and community/grassroots organizing. The Lead will be the coordinator with the incubator, will hire other team members, and coordinate with the OpenDRI trainers on the field data collection methods and tools.

Administrative Assistant

The pilot require a person who maintains project finances, contracts, clerical duties, and emergent needs around the work space. The AA will track and record project activities in a log, which will help sponsors see how mapping is progressing and enable the Communications Lead to message the growth of the project to a larger audience. The AA may be asked to manage equipment sign-ins and sign-outs and to ensure that inventory of mapping (and office) supplies are kept at planned levels.

Champions

Several key skill sets are core to the pilot and generally are found by hiring professional ‘champions’, who can solve technical challenges and train others. These include:

Technology Coordinator

A technologist who can make devices and computers work and can coordinate with the AA over the signout of technology equipment (such as GPS units).

Software Developer

The pilot generally uses open-source software, which often needs to be customized to needs that are only discovered during the field data collection. The software developer should be skilled in programming around the software packages used in the pilot and capable of making customizations to meet team needs. He or she may also be asked to develop applications (‘apps’) that facilitate the growth of the community.

Geospatial Specialist

The pilot will encounter GIS challenges in the data entry. This GIS specialist will assist with building the team’s knowledge of participatory mapping and geomatics. He or she will have experience with land surveying, remote sensing, cartography, and GIS.

Participatory Mapping Specialist

Participatory mapping also has its challenges in training, field data collection, data entry, and quality assurance. The Participatory Mapping Specialist will be a key resource around these issues. He or she will likely have a background in GIS, Geography, Computer Science or a related field and have some experience working on community-based participatory mapping projects. It is preferable to have knowledge of OSM and other open-source GIS tools, such as QuantumGIS.

Field Data Collectors

The core of the project are the teams of field data collectors, who perform the field data collection and data entry into the GIS application. They tend to be students, who may be studying geography/GIS or computer science and are quick to learn.

Occasional Roles

More complex projects may require additional help in operation management, M&E, and communications. The OpenCities Toolkit has a description of additional roles in these areas.

Activities

Setup

The initial stages of pilot center on the logistics of starting-up a new initiative. These activities include:

Contracting Incubation Space

Under its ToR with OpenDRI, the incubator/logistics company will work with the Project Lead to contract a work space. This office may be solely for use by the project, or more commonly, it may work with an existing innovation space or incubator to bring OpenDRI into a larger community of entrepreneurs and technologists. This latter option is preferable because it exposes the team to additional resources and provides a platform for building relationships, which is important for scaling the initiative.

Obtaining Equipment

The logistics company will need to purchase computers, GPS units, PDAs/smart phones, communications tools (radios/phones), and potentially, some office software (though most software needed is free and open source). The OpenDRI pilot management team will need to ensure that this equipment has a plan to ensure its security, and that any devices which go into the field have a basic sign-in and sign-out procedure.

Hiring Staff

The OpenDRI trainers will likely already have the Project Lead under ToR from the Design phase. This management team will reach out to universities and other partners so that they can hire the rest of the staff. It is recommended that the pilot hold an event to raise awareness, recruit surveyors, and tell the story about what OpenDRI is doing.

Developing the questionnaire for field data collection

Based on the data model from the Design phase, the team will work with the Geospatial lead, local universities, local government champion, and other experts to develop the questionnaire that surveyors will use in the field. This document needs to be short, yet cover the necessary data for tens of thousands of buildings. It can be changed, but alterations affect how commensurate data will be between earlier and later versions, and may require resurveying large areas. Careful thought should go into this document. The OpenCities Toolkit has deeper levels of detail on how to build the questionnaire.

Training

Once the space and staff are hired and ready for work, the pilot is ready for training in community mapping of exposure data. In general, this training will consist of the following skills:

• OpenStreetMap: how to enter data into the OpenStreetMap wiki or other geospatial platform.
• Field Data Collection Techniques: how to use a GPS unit, paper maps, and field survey tools to collect data that can be entered into the OpenStreetMap database.
• NOTE CONTROVERSY: Exposure Mapping: how to examine a building and understand how its construction is exposed to natural hazards, including common signs of structural weakness.
• Questionnaire: a training to bring together how the questionnaire developed for the pilot related to OpenStreetMap, Field Survey Techniques, and Exposure Mapping.

The National Society for Earthquake Technology–Nepal has built a training manual to teach how to read a building for structural weaknesses (add link when approved).

Data Collection: Field Surveys and Mapping Parties

When surveyors have been trained, the pilot begins to collect data via mapping parties. These events organizing between 10-40 mappers to collect data in a specific region of a city, with the goal of being as comprehensive as is possible for that region.

Mapping parties generally last several hours, depending on how long the team needs to travel from the work space to the area that needs to be surveyed. If transportation is needed, the incubator/logistics company should arrange for rentals, drivers, or other arrangement suitable to context.

When teams return to the work space, they should enter their data into OpenStreetMap as a social activity, where experts can aid surveyors with questions about how to code specific attributes and answer any unexpected issues that emerge. More information can be found in the OpenCities Toolkit. This routine becomes the daily work for several months, until the goals of the pilot are reached.

Data Quality Assessment: Coverage, Accuracy, reviews with team

As the surveying teams progress, it is important that the geomatics lead work with local universities and other OpenDRI experts to assess the quality of the data that is being entered into OpenStreetMap. OSM novices often make mistakes, and this problem is amplified by the specificity required in the entry of exposure data. When a pattern emerges in the consistent miscoding of certain information, the management team should train volunteers around the issue.

Participatory Mapping: Troubleshooting politics and perceptions

Participatory mapping has been a technique for more the 20 years. It has a wide variance in the quality of work. Several national mapping agencies have reacted to poor quality of participatory mapping data, especially when those activities have not adhered to national or international standards for mapping data. the advent of OpenStreetMap has largely addressed the core concerns in this area. That said, OpenDRI team should work with the national mapping agencies to understand its standards and adhere as closely as possible to them.

---

3. Policy Development

The space for OpenDRI to develop pilots requires coordination and communication between the various ministries and partners. This process is relationship intensive: it requires building trust through consistent contact as well as the delivery of small bits of data across previously unscalable boundaries. The chief ally for this work is the Open Data Working Group and the country offices of the DRM project managers (TTLs) that OpenDRI is supporting.

Open Data Working Group

(this is an area where we need to explore additional cases around how OpenDRI works with ODWG to develop its policies)

Develop a Plan for Data Access

The partners must also decide on who can have access to the data and under what terms. For data to be open, it needs to be licensed so that it can be capable of being redistributed, so that it can be turned into derived works that reuse the original data. Getting to this point is often a process:

1. Manual Interagency Sharing: government agencies transfer data on a case-by-case basis using ad hoc means.
2. Internal Government Data Catalogue Network: government agencies exchange data using a confederation of data catalogues and web services.
3. Open Government Data: one-way sharing of government data to the public, sometimes via specific authorizations for specific individuals.
4. Open Linked Data: the government and public participate in a data commons, where everyone exchanges data.

DRM Project Managers

(this is an are where we need additional data. What are the key lessons learned from developing the pilot with the DRM team in a region or country? What do they need to know on a weekly or monthly basis)?

Reporting Strategy

OpenDRI has developed monthly progress reports to keep partners informed about what data is being collected and released. It provides a summary of the key achievements, a table of the latest data summary figures and participation numbers, as well as an area for lessons learned and future plans. The format may be found here: Progress Report Template.

---

4. Demonstrations of Analytical Products

The goal of OpenDRI is to provide the data that others can transform into tools that explore risks from natural hazards. However, there are few stronger incentives to expand data collation and collection efforts than seeing newly available data turned into an analytical product—especially a visualization that shows something unexpected or confirms a hunch which previously had no evidence to support it. OpenDRI frequently partners with other efforts in DRM to provide this link between DRM data management and DRM analysis.

InaSAFE

Built through a partnership between the Australian Indonesian Fund for Disaster Reduction (AIFDR), Geoscience Australia, and the World Bank, the InaSAFE tool is a free and open-source risk communication tool for non-technical users. It allows anyone to pull information about the built environment together with models of natural hazards and create a scenario that shows the potential impact of the hazard on the built environment.

InaSAFE is built as an easy-to-install plugin for another open-source GIS desktop software QuantumGIS (often called qGIS). It can pull data from the GeoNode or other web services, OpenStreetMap, and models like the Global Earthquake Model or several tsunami and flooding models built by Geoscience Australia and other partners.

CAPRA

(team: we need an explanation and example of the link between an OpenDRI project and CAPRA. Who has documentation: concept notes, ToRs, graphics, etc? Bolivia? LAC?)

---

Who

Management Team

The Pilot Management Team often includes three expert consultants:

• OpenDRI Lead: an expert manager of the integration of various tactics into a comprehensive strategy around DRM data collection and curation.
• Community Mapping Lead: an expert in the implementation of community mapping methods for the collection of exposure data.
• Open Data Catalogue Specialist: an expert in the technical mechanisms of collecting and cleansing data from government ministries and community mapping efforts into a comprehensive data catalogue. This expert will have skills to train local officials in the deployment of an OpenGeo GeoNode or the use of existing ESRI GeoPortal as data catalogue.

---

Outputs and Metrics

The pilot produces data and grow community. As a result, its output metrics center on these two issues.

Data

• Coverage/Extent: what percent of the country’s risk areas are reflected in the data sets?
• Vintage: how new is the data across the coverage of the country?

Community

• Surveyors: how are the number of surveyors scaling over time? What turnover exists?
• Outputs: what is the productivity of the surveyors over time? What is the quality of that effort?
• Social Network: what is the shape and vector of the social network? Who are the supernodes? How many new nodes?

---

Case: Nepal: Pilot of Kathmandu.

In early 2012, the World Bank’s South Asia Region (SAR) engaged in the project with Nepal’s Ministries of Education and Health to retrofit schools and health facilities against seismic risks in the Kathmandu valley. The prioritization of this investment required knowing both where these facilities were located and the relative exposure of these structures to earthquakes. When the teams could not locate a comprehensive list of schools or health facilities to begin their risk assessment, they turned to the GFDRR team that was working on OpenDRI.

Scoping

Through a series of meetings, GFDRR introduced SAR and the GoN to the work that OpenStreetMap had done in Indonesia under AIFDR as well as the OSM efforts in the Kotse Valley with the American Red Cross. Based on the need to collect high-resolution data about more than 10,000 buildings in Kathmandu, SAR and the GoN worked with GDRR to fund an OpenStreetMap effort to map every school and health clinic in Kathmandu. In the process, SAR and GFDRR agreed to create the first site in a new OpenCities Project, where open data and efforts like OpenStreetMap would be used to catalyze better urban planning.

Design

The pilot focused on building a field data collection work flow that would allow for the rapid discovery and mapping of schools and hospitals in the Kathmandu valley. This required establishing a task force of ministries and local partners to advise the effort and provide the authority for the work. It also required building an innovative approach to rapidly mapping an unknown number of facilities.

Community Mapping

The pilot used the innovation lab design, contracting with a local incubator (Biruwa) host and help mobilize field data collectors. The team included an OpenDRI specialist who went on a three-month mission to Kathmandu; a local OSM consultant who had recently completed a PhD around OpenStreetMap; and a part-time knowledge manager, who was responsible for tracking progress and lessons learned. The team trained a staff of 13 interns, who in trained over 350 local mappers. They divided city into 4 quadrants, and used mapmyschool.org to locate schools and health facilities not found on any list. Volunteers were also given a letter from the Ministry of Education to help allay concerns of principals who were reluctant to share information or allow teams onto their property.

Data Quality

In first three months, the mappers collected over 65,000 building footprints using a detailed field data collection form. The project also partnered with the Nepal Society of Earthquake Technology, a local that provides seismic and structural engineering consulting, and two universities, where professors and engineers provided QA on the data.

NSET

NSET reviewed a sampling of small percentage of the buildings surveyed by the volunteers. They flagged issues to adjust training, and then provided that training to volunteers. It helped to taking pictures of every building along the way, enabling engineers to review both photos and forms.

University College, London

Similar to UGM study from Indonesia, the team worked with Muki Hakley, who gave advice to the Geomatics Dept at Kathmandu University on assessment of geometry of data and fitness of use. Professor Hakley also reviewed the accuracy of building footprints, road networks (where roads overlap), and other aspects of the data.

Demonstrating Analytical Products

A team from Stanford University is currently building earthquake risk models from the data collected in Kathmandu, with plans to show how the impact of a seismic event on the municipal facilities that would be critical sites for a response operation.

---

6. Scaling

Pilots demonstrate new possibilities. They allow partners to see firsthand how cost effective, fast, and comprehensive an OpenDRI engagement can be. Pilots can also expose challenges of the local context. Data may be difficult to obtain or data sets may require very time-intensive work to prepare it for release. In either case, there will emerge a point where the partners to OpenDRI will discuss whether to scale the initiative and where the strategic areas of expansion might be.

(note to partners: we are hoping to expand this section with additional practices and ideas)

---

Objectives

The effort to scale OpenDRI pilots is often a matter of replicating the structure of the original pilot in additional locations, rather than increasing the size of a single pilot. This approach is congruent with typical semantic web (Web 3.0) institutions, which tend to follow “small pieces, loosely joined” organizational designs of both information technology industries (and the textile manufacturing industry, where the process of moving a garment from raw material to finished good may have a route through dozens of companies in several countries).

.

---

1. Collating Data

After early adopters have added data to a data catalogue, the slow process of building a network of users begins. The sign for this work to start is if a Data Curator and partners have loaded data and a user base begins to use the data catalogue to solve immediate problems. The usage indicates that the pilot happened in a fertile environment for additional work.

In general, there are three approaches to scaling open government risk data:

• Expanding the number of data catalogues. Data catalogues no longer need to be centralized. Instead, they can form a confederation of catalogues. In this way, each ministry can own and curate its own data, and choose what data to make available to whom. The confederation of catalogues might offer other ministries very different access rights than the rights afforded to the general public.

• Increasing data sets within each data catalogue. Often the initial data sets are limited to information which is considered to be low risk and high reward for its release. Expanding the data available in each catalogue will work towards data which might have lower rewards or higher risks for its release.

• Increasing interconnections. The density of the interconnections between data will greatly affect the perceived value of the network. Each data set can be viewed as a LEGO™ piece. When topography data from one ministry can be combined with meteorological forecasts and river gauge data, flood models can more accurately predict flood damage from an upcoming storm seasons.

In each case, planning how to scale requires considering several factors:

• Meeting the use cases of partners. The availability of data often exposes new use cases, many of which may have additional data needs. The Open Data Working Group and Data Curator can explore which uses meet their priorities and decide which direction to take the additional effort.

• Building relationships with gatekeepers to other data sets. Data catalogues often become gravity wells for information: they pull others with data toward growing supernodes. This credibility enables the OpenDRI to locate islands of data which may be considered core to DRM, but may be held outside of government (often survey or GIS firms). Negotiating the release of these data sets has sometimes proven to be relatively easy when there is a place to host the data without cost to the gatekeepers.

International Interconnection

The partners to this guide are actively seeking to build connections between national data sets around DRM with the datasets used by others in disaster risk reduction, disaster risk management, and disaster response operations. These partners have built (or are building) data catalogues that may augment national efforts. It may also be in a client nation’s interest to ensure that the data used in these international data catalogues is aligned with the data collected via OpenDRI and other approaches to open data run by client nation ministries.

.

---

2. Collecting Data

Research indicates that scaling the efforts to collect data (e.g. via community mapping) is generally a process of building additional pilots then interconnecting them. This strategy roughly parallels grassroots organizing, when organizers carry the effort from one place to the next, building a larger national effort along the way.

Expanding each site

Each new pilot will generally start by cloning the previous cities, with customizations to context. Some areas of customization include architectural traditions of a particular city, natural hazards in the specific place, language, and custom. The organizational design will be similar: incubator plus facilitator and related staff. In general, community mapping efforts have scaled by cloning without problem.

Preparing pilots be sustainable entity

The more difficult part of community mapping with volunteers is converting a pilot into a sustainable entity. The communities that come together to collect exposure data are not meant to be a one-off. They must curate that data over time to create a viable risk management ecosystem. While the costs are low, these organizations need to sustain a small set of paid leaders around a network of volunteer surveyors. The OpenDRI team needs to consider strategies to fund and sustain these efforts. To date, all pilots are still working off pilot funding and none have developed revenue models in either the not-for-profit or for-profit spaces. Some groups might well be able to develop social ventures/social entrepreneurship models. (editor’s note: look to advice of partners here.)

An important element of current work is partnerships with local universities. By integrating risk assessment into the curricula of geospatial and structural engineering courses, the OpenDRI team has been able to expose the next generation of government officials, engineers, and analysts to thinking about risk in terms of probable futures. Students have become an important source of volunteers, especially as they seek to gain experience in a new set of skills.

Data Authority and Quality

Because data collected from community mapping is collected using methods from participatory mapping, traditional GIS professionals need to see proof of the accuracy of the data. They want to know that the data is reliable and to know where it is inaccurate (all data has errors, the trick is controlling for those inaccuracies). OpenDRI team has had success contracting with local academic institutions to perform a QA study on community mapping data. One such report was done by UGM in Indonesia. It catalyzed a change to the conversation about community mapping in the client government, which subsequently expanded its use of OpenStreetMap. University College, London and Kathmandu University performed a similar study for the community mapping work in Nepal.

---

3. Creating Policies to manage linked open data

When ministries begin sharing data that has previously been closed, policy and legal question often arise. Some of these issues center on access, privacy, and standards. The OpenDRI team will need consider several questions:

• Access. Who can view the data? Do some data need to be kept private for security reasons (such as some data about nuclear power plants)?
• Privacy. What the data alone or as a mosaic reveals about others? Do the data reveal information about citizens that needs to be kept private? How can the data be released for DRM purposes in ways that protect citizen privacy?
• Standards. What is the national standard for certain data types? Do ministries use formats that are compatible with each other? What is the cost of translating data from one format to another as it now moves from ministry to ministry and outside partners? If there are problems with standards and data translation, what is the standard that the nation will follow?
• Metadata. How can users find the data they need? Metadata provides a common language to describe the data. In this way, experts in various specialties can define their vocabularies and enable others to find the data that they need.

Building Curation capacity and QA/M&E

As data scales in size and interconnection, the challenges of curating it increase. When a system is flooded with high volumes of poor quality data, it becomes far less useful than it was when it started with a few solid datasets. The Data Curator needs to become the steward of the data. He or she will not only add new data, but also removing data that has become stale, cleanse data that contains errors in accuracy or formatting. The Data Curator will need to establish and apply data typologies and hierarchies. The Data Curator’s role is to leave data better than he or she found it. The Quality Assurance capacity and M&E around OpenDRI need to be tied to the quality, findability, and usage of the data under curation. OpenDRI is developing a guide to this curation process.

Community Mapping Standards and Certification

To be most useful to national mapping agencies, the data from community mapping needs to follow standard methods. One approach to ensuring this standardization is through certification of community mapping organizations. The national mapping agency in Indonesia (BIG), has recommended the creation of a set of certifications around community mapping. This effort should be tracked and evaluated for its effect on the quality of data. Areas of concern for a national mapping agency include:

• Providing Access to Imagery and Base Information. When volunteers start from imagery and existing trusted data, they tend to be more accurate in their mapping than when they start from scratch. National mapping agencies are beginning to find ways to provide imagery and data under license to community mapping groups. See the US State Department Humanitarian Information Unit Imagery to the Crowd.

• License Guidelines. Data received from community mapping needs to be under a license that allows the national mapping agency to redistribute the data and turn data into derivative works.

• Certification Process. Community mapping organizations can participate in trainings that enable them to become certified by the national mapping agency in the standards that geospatial data must follow in the country. The national mapping agency can also observe a community mapping organization’s training program and certify that these methods are teaching approved standards and generating usable data.

• Pilots and Experiments. National mapping agencies often lack a safe place to try new methods, including community mapping. Pilots and experimentation programs with universities offer a way that different approaches, standards, and technologies can be tried in a field environment at low cost.

Training Local Officials in the Use and Extension of OpenDRI

Local officials working in DRM or disaster management are often starved for good data. It is also common to find that their geospatial analysis capability is a single consultant who holds onto the data for job security. Providing these officials with training in how to make use of OpenDRI data—especially with usable tools like InaSAFE—gives them a new set of capabilities. It also opens their mind to thinking beyond traditional approaches to DRM and exposes them to probabilistic thinking, scenario planning, and data sharing with a larger ecosystem of local officials.

---

4. Catalyzing efforts around the use of the data

Growing the network of individuals and analysts who use OpenDRI data is an important approach to increasing the volume and quality of DRM data available in the client nation. In some cases, the data will generate unexpected synergies. Entrepreneurs have used street map data to build smart-phone apps around transportation, logistics, and other location-based services—none of which relate directly to DRM. Universities have found that the data is a ripe target for research not only in GIS, but also in social and physical sciences.

Entrepreneurs

A proven approach to catalyzing entrepreneurial activity around OpenDRI data is to co-locate the OpenDRI in an incubator for local technology community. When community mappers work in the same space, entrepreneurs can discover not only how rich the OpenDRI data is, but also learn how to integrate their software with the open data services offered by OpenDRI.

(Include UNICEF case as well as Nepal)

University Partnerships

Universities train many of the geospatial professionals, government officials, and scientists who will be using OpenDRI data. GFDRR has found great success working with local professors. Training programs for OpenDRI can be given to universities to integrate into their courses, and data access provides students and faculty with a rich source of information to explore through research in graduate programs. In turn, the community of young professionals continues to expand the mental model of thinking in terms of risk management, providing a stable source of critical thinkers in the future.

---

Case: Indonesia: Mapping Jakarta for $20,000 USD

In 2010, GFDRR Labs formed a working partnership with an AusAID and Government of Indonesia initiative known as the Australia-Indonesia Facility for Disaster Reduction (AIFDR) with the goal of applying innovation and technology to the challenges of disaster risk management. AIFDR seconds technical specialists from Geoscience Australia to increase Indonesia’s national disaster management agency’s (BNPB) capacity to use risk information for disaster preparedness, including a free and open source platform called InaSafe. Because Indonesia needed higher resolution exposure data to engage in systematic planning around a wide range of natural hazards, AIDFR contracted with the Humanitarian OpenStreetMap team to engage in a community mapping pilot in Pedang in 2011.

Pedang project with AIFDR

Over the course of 18 months, HOT trained (x) mappers and put 250,000 buildings on the map for $200K, including the development of training materials, their translation into Bahasa, and the development of a several software tools that greatly simplified the OpenStreetMap interface for volunteers. Despite the quality and quantity of data, the GoI remained skeptical about its accuracy. Participatory mapping had a long history in Indonesia; much of the data was collected on paper using techniques that made it understandable from the eyes of a local population, but lacked rigor and standardization necessary to bring those data into a national map. Crisis soon changed this situation.

(note: put in a comment about sustainability of mapping competitions?)

Mapping Jakarta for $20,000

In 2012, the monsoon forecast indicated that the province of Jakarta was at high risk for a greater intensity of flooding than usual. The governor of Jakarta asked his team for a risk assessment and plan of action. After being limited by existing data, the provincial disaster management authorities (under the national disaster management agency, BNBP) asked AIFDR and the World Bank to support a community mapping project, scaling the work from Pedang to all five cities in Jakarta.

The governor enlisted the support of the 5 mayors, who asked more than 200 ward chiefs to work with the community mapping effort. Through several mapping parties, the volunteers trained by the Humanitarian OpenStreetMap Team worked in pairs with the ward chiefs in a process that has since been named co-mapping. This method couples an experienced community mapper with a non-technical representative of local government. Together, they add attributes to geographic features on paper maps, which are then turned into digital geospatial information, printed back out, and authorized (by signature) by the ward chief. The process lends additional authority to the map.

For less than $20K, the effort mapped most of Jakarta. Indonesia now has over 1 million buildings in OpenStreetMap. The National Mapping Agency (BIG) has developed standards for the way that such data must be collected, tagged, and licensed to the government. And BNBP officials worked with AIFDR to develop InaSAFE as a risk communication tool that allows local officials–including those in Jakarta–to perform basic scenario analysis of using the OpenStreetMap data and web services on hazards and vulnerability prepared by AIFDR and Geoscience Australia.

---

7. Sustaining

When OpenDRI has been successful, governments and communities will have new tools and methods to collect and curate data about their exposure to natural hazards. They will have met the goals that the partners outlined in the design phase and adapted along the way.

Sustaining a changed way of working with risk data is the work of the host nation itself. However, sustaining change is difficult. Without continuous funding, communities of practice often fall back to using familiar, older methods; thought leaders move onto the next problem; and experts find opportunities to grow their skills further in new organizations. The risk for OpenDRI is that after the project is over (and the consultants and firms have moved onto other projects), key stakeholders will revert to the approaches that they used for data sharing prior to the OpenDRI engagement.

OpenDRI’s designers take a practical approach to this problem, acknowledging that an occasional influx of energy and resources may be necessary to keep work going. The development partners who funded and catalyzed the change can take steps to ensure the effort remains vibrant and productive.

---

Objectives

The first and last task of the OpenDRI management team is to create the framework into which they can inject these bursts of sustaining energy. The sustaining phase creates an architecture for continued work. This design must include a plan for ongoing training, occasional funding for small projects, and a framework for champions to grow a locally owned, long-term, sustainable open data ecosystems,

A key problem between the scaling and sustaining phases is to determine when OpenDRI has met its intended goals. This task may not be easy. The original objectives set forth in the initial Concept Note may have shifted to more advanced risk thinking. As the pilots evolved into a larger efforts across greater areas, the effort may have grabbed the attention of high-level leaders, who want to continue low-cost, high-output work. This shift is itself a success, but also may require additional capacity building.

What

There is a point in the maturation of risk thinking that OpenDRI must clarify its scope to be around data collection and curation, not risk assessment. When tasks around OpenDRI become more focused on the analysis of data than the management of the data, OpenDRI can cede to other efforts in DRM capacity building and risk assessment.

The goals of the Sustaining phase can include:

• Creating opportunities for training and networking that increase the interconnections between members of the network of data collectors, curators, analysts, and decision makers.
• Linking the projects with other development partners, including handing off the OpenDRI program to other organizations.
• Building trust in the accuracy and authority of the data.
• Encouraging academic institutions to include OpenDRI training materials, software, and data in their curricula.
• Catalyzing the development of organizations around data collection and management, often as social enterprises.
• Funding small software development projects that extend the use and utility of OpenDRI data.
• Funding the start-up costs for Living Labs, where individuals from across the open data ecosystem can gather to work on start-ups, continue projects, and have a neutral space to work on inter-organizational problems.

---

Why and How

The mix of sustaining efforts may change over time, but they will also need to be tailored to the context. The range of tactics include:

Ongoing Training and Networking

The more connections OpenDRI creates between ministries, analysts, data collectors, data curators, and communities, the greater value that network will bring to the individuals who join it. The greater the capacity of each member of this network, the more capacity the network will have to analyze its risks. OpenDRI has been catalyzing these connections and capacity through training programs. Some meetings bring individuals from the region to a site where they can meet each other, discuss shared problems, and explore common solutions. Other meetings occur via webinars, where one or more speakers can broadcast their presentation to dozens or even hundreds of individuals.

Academic Partnerships

Universities are core partners in most OpenDRI engagements. Their academic departments around geography, geomatics, and GIS provide many of the students who volunteer to engage in community mapping. They also train many of the individuals who will curate geospatial data in government ministries and the private sector. When OpenDRI can integrate methods around open data and community mapping into the curricula of university programs, it ensures continued extension of an open approach to data collection and curation.

OpenDRI has also enlisted universities to perform evaluations of the quality of data produced by community mapping. Having a professor from a local university assess the accuracy of data collected by volunteers provides a clear set of recommendations for improving operations at the same time that it provides an honest understanding of the strengths and weaknesses of the data. Knowing the accuracy of the data has enabled national mapping agencies to set policies around the integration of community mapping data into official government data.

Development of Organizations around OpenDRI Data

The more organizations that make use of OpenDRI data, the more likely the data are to be updated. Embedding OpenDRI into entities like a Living Lab, Innovation Lab, or other community-based organization has proven to be an important step in the process of building an ecosystem around open data. It is through these incubators that social entrepreneurs learn about the data, how it is collected and curated, and how they can build applications or even revenue models around services that make of the data. Not all such technology may be related to risk management. Mapping a cities roads and infrastructure is a crucial step to providing information services around mapping, logistics, and transportation. In Port-au-Prince, the Tap-Tap group taxi network has been mapped, as have the matatus in Nairobi—both based on the OpenStreetMap work done in the country.

Software Development around OpenDRI Data

In the process of expanding the use of OpenDRI tools and practices, government ministries may encounter software interoperability problems or discover the need to extend a tool to include features that were not foreseen during the initial OpenDRI implementation. This moment is ideal for a small investment—potentially in the low $10,000s USD—to help add features to the software. Because OpenDRI is an open-source software model, licensing is already established to allow for the feature to be usable by other countries as they encounter similar challenges or desired functionality.

Partnerships with other international development organizations

Several institutions are building open data initiatives, several of which are partners to this field guide. An OpenDRI program can be integrated with this work. At the very least, data services can be connected. More desirable would be full integration and coordination of the efforts, so that open data enables the system of actors to better understand what each is doing, better target their investments and activities, and better listen to the actual needs of the client through better data.

---

Who

Building the plan for sustaining OpenDRI should include those who are the core partners to OpenDRI, including:

Ministry Officials around Open Data

The Open Data Working Group owns the commitment between ministries to share data. As such, they are ideal hosts for the training and networking events. They are also a key source of information about how well software is working for national needs and how it might be adapted for the future.

Academic Community

University professors and their students in geology, geomatics, geography, and GIS are key partners in sustaining OpenDRI.

Social Entrepreneurs

The youth who discovered the Internet and mobile phone are now building software to tackle the challenges with which they grew up. In many cases, their needs are simple: a reliable internet connection and electricity as well as mentors and a space to work. Enabling these social entrepreneurs with guidance around their code and their revenue models is an important aspect of sustaining OpenDRI.

---

Outputs

The most important output from OpenDRI is a changed approach to risk management. When politics, practices, and tools reflect the impact of OpenDRI on a society’s thinking about DRM, the initiative has been a success—even when the specific tactics or projects with which OpenDRI began have ceased.

Other outputs from OpenDRI sustaining phase are more proxy measurements around this change to thinking, and should be used as signs of process that is working rather than direct measure of OpenDRI itself:

Living Lab Sustainment

When the Living Lab becomes a sustainable organization, with its own revenue model, the community gains a stable place for other activities to occur. It also establishes a neutral space between government, private sector entities, and community-based organizations. The financial stability of this living lab is an important aspect of OpenDRI sustainment. The founders of this space should be encouraged to seek grants, contracts, and other financial support to keep the space vibrant and growing.

“Apps” and Software

When entrepreneurs and ministries are producing open-source software based on OpenDRI data or software, it is very likely that these products would not have emerged without the OpenDRI engagement. The success of the sustaining phase can be viewed, in part, by the creation of spaces for this type of software development to occur.

Training Program Impacts

An active training program around the management of risk data and its application to risk assessment is an important activity from the training programs. It may be possible to measure the impact of these trainings on practice, though this particular output is beyond the scope of OpenDRI per se.

---

Case: Caribbean Community of Practice

---