From the Himalayas to the Andes: Crowdsourced disaster risk mapping

By Wei Liu, IIASA Risk and Resilience Program

What do Rajapur, Nepal; Chosica, Peru; and Tabasco, Mexico all have in common? Flooding:  these areas are all threatened by floods, and they also face similar knowledge gaps, especially in terms of local level spatial information on risk, and the resources and the capacities of communities to manage risk.

To address these gaps, I and my colleagues at IIASA, in collaboration with Kathmandu Living Labs (KLL) and Practical Action (PA) Nepal are building on our experiences in Nepal’s Lower Karnali River basin to support flood risk mapping in flood-prone areas in Peru and Mexico.

Recent developments in data collection and communication via personal devices and social media have greatly enhanced citizens’ abilities to contribute spatial data, called Crowdsourced Geographic Information (CGI) in the mapping community. OpenStreetMap is the most widely used platform for sharing this free geographic data globally, and the fast growing Humanitarian OpenStreetMap Team has developed CGI in some of the world’s most disaster-prone and data-scarce regions. For example, after the 2015 Nepal Earthquake, thousands of global volunteers mapped infrastructure across Nepal, greatly supporting earthquake rescue, recovery, and reconstruction efforts.

Today there is excellent potential to engage citizen mappers in all stages of the disaster risk management cycle, including risk prevention and reduction, preparedness and reconstruction. In this project, we have successfully launched a series of such mapping activities for the Lower Karnali River basin in Nepal starting in early 2016. In an effort to share the experience and lessons of this work with other Zurich Global Flood Resilience Alliance field sites, in March 2017 we initiated two new mapathons  in Kathmandu, with support from Soluciones Prácticas (PA Peru) and the Mexican Red Cross, to remotely map basic infrastructure such as buildings and roads, as well as visible water surface, around flood-prone communities in Chosica, Peru and Tobasco, Mexico.

@ Wei Liu | IIASA

March 17th, 2017, staff and volunteers conducting remote mapping at Kathmandu Living Labs @ Wei Liu | IIASA

Prior to our efforts very few buildings in these areas were identified on online map portals, including Google Maps, Bing Maps, and OSM. Through our mapathons, dozens of Nepalese volunteers mapped over 15,000 buildings and 100 km of roads. The top scorer, Bishal Bhandari, mapped over 1,700 buildings and 6 km of roads for Chosica alone.

Having the basic infrastructure mapped before a flood event can be extremely valuable for increasing flood preparedness of communities and for local authorities and NGOs.  During the period of the mapathons, the Lima region in Peru, including Chosica, was hit by a severe flood induced by coastal El Niño conditions. Having almost all buildings in Chosica mapped on the OSM platform now makes visible the high flood risk faced by people living in this densely populated area with both formal and informal settlements. These data may support conducting a quick damage assessment, as suggested by Miguel Arestegui, a collaborator from PA Peru during his visit to IIASA in April, 2017.

Recognizing the value of crowdsourced spatial risk information, we are working closely with partners, including OpenStreetMap Peru, to mobilize the creativity, technical know-how, and practical experience from the Nepal study to Latin America countries. Collecting such information using CGI comes with low cost but high potential for modeling and estimating the amount of people and economic assets potentially being affected under different future flood situations, for improving development and land-use plans to support disaster risk reduction, and for increasing preparedness and helping with allocating humanitarian support in a timely manner after disaster events.

This slideshow requires JavaScript.

Having the basic infrastructure mapped before a flood event can be extremely valuable for increasing flood preparedness of communities and for local authorities and NGOs.  During the period of the mapathons, the Lima region in Peru, including Chosica, was hit by a severe flood induced by coastal El Niño conditions. Having almost all buildings in Chosica mapped on the OSM platform now makes visible the high flood risk faced by people living in this densely populated area with both formal and informal settlements. These data may support conducting a quick damage assessment, as suggested by Miguel Arestegui, a collaborator from PA Peru during his visit to IIASA in April, 2017.

Recognizing the value of crowdsourced spatial risk information, we are working closely with partners, including OpenStreetMap Peru, to mobilize the creativity, technical know-how, and practical experience from the Nepal study to Latin America countries. Collecting such information using CGI comes with low cost but high potential for modeling and estimating the amount of people and economic assets potentially being affected under different future flood situations, for improving development and land-use plans to support disaster risk reduction, and for increasing preparedness and helping with allocating humanitarian support in a timely manner after disaster events.

Flood-inundated houses and local railway in Chosica, Peru, 18/03/2017 @ Miluska Ordoñez | Soluciones Prácticas

The United Nation’s Sendai Framework for Disaster Risk Reduction states that knowledge in “all dimensions of vulnerability, capacity, exposure of persons and assets, hazard characteristics and the environment” needs to be leveraged to inform policies and practices across all stages of the disaster risk management cycle. CGI has a great potential to involve citizens from around the world to help fill this critical knowledge gap. These pilot mapathons conducted between Nepal and Latin America are promising examples of supporting community flood resilience through the mobilization of CGI via international partnerships within the Global South.

This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.

Learn from the past, prepare for the future

By Roman Hoffmann, Wittgenstein Centre for Demography and Global Human Capital (IIASA, VID/ÖAW and WU), Vienna Institute of Demography, Austrian Academy of Sciences

Flooded street in Meycauayan, Bulacan, Philippines (credit: Kasagana-Ka Development Center Inc., 2016 )

Floods, droughts, and tropical storms have significantly increased, both in frequency and intensity in recent years. The burden of these events—both human and economic—falls in large part on low and middle-income countries with high exposure, such as coastal and island nations. In a recent study, with IIASA researcher Raya Muttarak, we found that education significantly contributes to increasing disaster resilience among poor households in the Philippines and Thailand, two countries which are frequently affected by natural calamities.

In these countries, public disaster risk reduction is important, yet public measures, such as investments in structural mitigation for large buildings or infrastructure, implementation of early warning systems, or planned evacuation routes and shelters, may not be enough to sufficiently protect communities from the devastating impacts of natural calamities. In addition, the undertaking of individual preparedness measures by households, such as stockpiling of food and water, strengthening of house structures, and having a family emergency plan, is crucial. Yet, even in areas which are heavily exposed to disasters, people often do not take any precautionary measures against environmental threats.

How people can be motivated to take precautionary action has been a fundamental question in the field of risk analysis. In the new study, which was based on face-to-face interviews in both Thailand and the Philippines, we found that prior disaster experience, which is influenced by geographical location of the home, is one of the key predictors of disaster preparedness. For those who were affected by a disaster in the recent past, education does not seem to play a significant role—they have already learned by experience.  However, among those who had not previously been affected, educational attainment becomes a key determinant. Even without having experienced a disaster, the educated are more likely to make preparations. In fact, educated people who haven’t experienced a disaster have preparedness levels that are as high as those of households who were only recently affected. Since education improves abstract reasoning and abstraction skills, highly educated individuals may not need to experience a disaster to understand that they can be devastating. This suggests that education, as a channel through which individuals can learn about disaster risks and preventive strategies, may effectively serve as a substitute for (often harmful) disaster experiences as a main trigger of preparedness actions.

In additional analyses, we investigated through which channels education promotes disaster preparedness by looking at the relationship between education and different mediating factors such as income, social capital and risk perception, which are likely to influence preparedness actions. We found that how education promotes disaster preparedness is highly context-specific. In Thailand, we found that the highly educated have higher perceptions of disaster risks that can occur in a community as well as higher social capital (measured by engagement in community activities) which in turn increase disaster resilience. In the Philippines, on the other hand, it appears that none of the studied mediating factors explain the effect of education on preparedness behavior.

Emergency shelter, San Mateo, Rizal, Philippines (credit: Kasagana-Ka Development Center Inc., 2013 )

Certainly, it remains important for national governments to invest in disaster risk reduction measures such as early warning systems or evacuation centers. However, our study suggests that public funding in universal education will also benefit precautionary behavior at the personal and household level. In line with recent efforts of the UN to promote education for sustainable development, our study provides solid empirical evidence confirming the important role of education in building disaster resilience in low and middle-income countries.

Reference
Hoffmann, R. & Muttarak, R (2017). Learn from the past, prepare for the future: Impacts of education and experience on disaster preparedness in the Philippines and Thailand. World Development  [doi:10.1016/j.worlddev.2017.02.016]

This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.

How the Tohoku disaster is changing Japan

By Junko Mochizuki Research Scholar, Risk and Resilience Program

After a 9.0 magnitude earthquake and subsequent tsunamis struck the northeast of Japan on March 11, 2011, large-scale destruction of the coastal communities, including nuclear accidents, turned into a political maelstrom. Debates over the country’s alternative energy futures became intense; worries over ailing energy infrastructure, public safety, and the lack of transparency and accountability led to the most extensive restructuring of its power sector in the country’s recent history.

Against this backdrop, renewable energy was heralded as one of the important solutions: A new nation-wide Feed-in-Tariff (FIT) was introduced in July 2012, replacing the Renewable Portfolio Standard (RPS), which many had perceived, until then, as largely inadequate.

Nearly six years have passed since. Japan’s reconstruction, originally envisioned to last for 10 years, is now in its latter phase. The coastal communities are slowing recovering, with many focused on the idea of ”building back better.” We now hear less about the country’s energy future in the national and international media. But less documented is how well these communities are performing in terms of the ambitious reconstruction plans that they had proposed.

The 2011 earthquake, tsunami, and nuclear disaster led to major destruction in Northeast Japan. But did it also bring an opportunity to “build back better?” ©mTaira | Shutterstock

This was the context in which my colleague Stephanie E. Chang and I began our research titled Disaster as Opportunity for Change, recently published in the International Journal of Disaster Risk Reduction. We evaluated renewable energy transition trends in the 30 coastal communities in Tohoku, Japan from 2012-2015. We focused on energy transition as one measurable dimension of ”building back better (BBB),“ because this is a popular concept that is often talked about, but rarely analyzed through empirical modelling.

In this study, we sought to answer three simple questions. First, are the disaster-affected regions really “building back better?” Have they introduced more renewable energy than the rest of Japan?. Second, why did some communities achieved greater renewable energy transition than others during recovery? Third, what was the role of government policy? We were interested in answering these questions through quantitative analysis, instead of an in-depth case study, since such in-depth analyses are rare in the field of disaster recovery.

In a reconstruction study, we typically need about 10 or more years to make major conclusions. Since we did our study in year six, this study doesn’t provide the final answer, but rather whether the disaster led to opportunity to build back better.

Our research indicated some clues in answering the above three questions, but many puzzles remain. First, it was clear that the disaster-affected regions achieved a greater transition to renewable energy, particularly in both small and mega-solar adoption. Other renewables including wind and geothermal are lagging due to many factors such as more complex approval processes. We focused our analysis on energy transition, measured in terms of per capita approved renewable capacity, as opposed to indicators such as installed capacity or power generated, since the latter depend on many factors such as the readiness of grid systems in accommodating intermittent renewables.

We also found that the relationship between a transition to renewable energy and the extent of disaster damage, and other post-disaster changes such as number of houses being relocated, appears to be non-linear. This means that the destruction caused by disasters, and subsequent decisions to relocate population, provided at least some momentum for wider societal change. Clearly, when communities experience very large destruction or extensive change such as land-use adjustment, this can overwhelm the local capacity to implement broader changes such as major investments in renewable energy. Balancing competing reconstruction demands is, therefore, an important policy question that must be dealt with, most ideally, prior to any large-scale disasters.

Japan is building mega solar installations like this one in the region affected by the tsunami and earthquake ©SE_WO | Shutterstock

 

Third, our results remain somewhat inconclusive as to the contribution of government policy. Counter-intuitively, communities having renewable energy plans prior to 2011 adopted significantly less solar energy after the Tohoku disaster. Statistical modeling such as ours tells little about how different aspects of national and prefectural policies have fostered or hindered local energy transitions and these are better answered through other means such as in-depth interviews.

Overall, we find potentially complex drivers of “building back better” and we hope that this study motivates further systematic studies of societal change in the context of post-disaster reconstruction. Of course, a better articulation of what policies work in promoting change and why will also help foster the sustainability transition even without the impetus of a disaster.

Reference
Mochizuki J & Chang S (2017). Disasters as Opportunity for Change: Tsunami Recovery and Energy Transition in Japan. International Journal of Disaster Risk Reduction DOI:10.1016/j.ijdrr.2017.01.009. (In Press)

This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.

Mapping flood resilience in rural Nepal

By Wei Liu, IIASA Risk and Resilience and Ecosystems Services and Management programs

Disasters caused by extreme weather events are on the rise. Floods in particular are increasing in frequency and severity, with reoccurring events trapping people in a vicious cycle of poverty. Information is key for communities to prepare for and respond to floods – to inform risk reduction strategies, improve land use planning, and prepare for when disaster strikes.

But, across much of the developing world, data is sparse at best for understanding the dynamics of flood risk. When and if disaster strikes, massive efforts are required in the response phase to develop or update information. After that, communities have an even greater need for data to help with recovery and reconstruction and further enhance communities’ resilience to future floods. This is particularly important for the Global South, such as the Karnali Basin in Nepal, where little information is available regarding community’s exposure and vulnerability to floods.

Karnali Basin in Nepal © Wei Liu | IIASA

Karnali Basin in Nepal © Wei Liu | IIASA

That’s why we are working with Practical Action in the Zurich Flood Resilience Alliance to try to remedy this situation. Participatory Vulnerability and Capacity Assessment is a widely used tool to collect community level disaster risk and resilience information and to inform disaster risk reduction strategies. One of our first projects was to digitize a set of existing maps on disaster risk and community resources where the locations of, for example, rivers, houses, infrastructure and emergency shelters are usually hand-drawn by selected community members. Such maps provide critical information used by local stakeholders in designing and prioritizing among possible flood risk management options.

From hand-drawn to internet mapping
While hand-drawn maps are ideal for working in remote rural communities, they risk being damaged, lost, or simply unused. They are also more difficult to share with other stakeholders such as emergency services or merge with additional mapped information such as flood hazard. With the recent increase in internet mapping, platforms such as OpenStreetMap have made it possible for us to transfer existing maps or capture new information on a common platform in such a way that anyone with an internet connection can add, edit, and share maps. As this information is digital, it makes it easier to perform additional tasks, such as identifying households in areas of high risk or measuring the distance to the nearest emergency shelter, to support effective risk-reduction and resilience-building.

Practical Action Nepal, the Center for Social Development and Research and community members discuss the transfer of community maps to online maps © Wei Liu | IIASA

Practical Action Nepal, the Center for Social Development and Research, and community members discuss the transfer of community maps to online maps © Wei Liu | IIASA

From theory to practice
In March 2016, the Project team travelled to two Nepal communities in the Rajapur and Tikapur districts, to pilot the idea of working with a local NGO (the Center for Social Development and Research) and community members, to transfer their maps into a digital environment. The latter can easily be further edited, improved and shared within a broad range of stakeholders and potential users. Local residents in both communities were excited seeing their households and other features for the first time overlaid on a map with satellite imagery. The Center for Social Development and Research was also very enthusiastic about integrating their future community mapping activities with digital mapping, without losing the spirit of participation.

Hand drawn maps produced from community mapping exercises in Chakkhapur, Nepal © Practical Action

Hand drawn maps produced from community mapping exercises in Chakkhapur, Nepal © Practical Action

 

The resulting online maps in OpenStreetMap of Chakkhapur, Nepal, showing the location of drinking water, an emergency shelter and medical clinic. ©OpenStreetMap

The resulting online maps in OpenStreetMap of Chakkhapur, Nepal, showing the location of drinking water, an emergency shelter and medical clinic. ©OpenStreetMap

Increasing resilience through improved information management
The first stage pilot study in the Karnali river basin confirmed the great potential of new digital technologies in providing accurate and locally relevant maps to improve flood risk assessment to support resilience building at the community level. The next step is to further engage local stakeholders.  A wider partnership has been established between Practical Action, the Center for Social Development and Research, the International Institute for Applied Systems Analysis and Kathmandu Living Labs to further build local stakeholders’ capacity in mapping with digital technologies, including a training workshop for NGO staff members in September, 2016.  The plan is to have more communities’ flood risk information mapped for designing more effective action plans and strategies for coping with future flood events across the Karnali river basin. A greater potential can be realized when this effort is further scaled up across the region and the results are placed into shared open online databases such as OpenStreetMap.

Further information

  • Flood Resilience Portal
  • Geo-Wiki Risk 
  • McCallum, I., Liu, W., See, L., Mechler, R., Keating, A., Hochrainer-Stigler, S., Mochizuki, J., Fritz, S., Dugar, S., Arestegui, M., Szoenyi, M., Laso Bayas, J.C., Burek, P., French, A. and Moorthy, I. (2016) Technologies to Support Community Flood Disaster Risk Reduction. International Journal of Disaster Risk Science, 7 (2). pp. 198-204. http://pure.iiasa.ac.at/13299/

Note: This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.

When global lessons are not so easily learned

By Junko Mochizuki, Research Scholar, IIASA Risk and Resilience Program

Experts in the field of emergency management like to emphasize that there are important “lessons learned” in the aftermath of disaster situations. After large disaster events such as the 2015 earthquake in Nepal, and 2013 super typhoon Yolanda in Philippines, forensic investigations are often conducted to reveal ”what went wrong”  in the chains of command, identifying what we can do differently when the next big one strikes. Such forensic investigations are not only relevant for the field of emergency management, but also for the field of disaster and climate risk management, which seeks to identify the underlying causes of what went wrong in the long chains of developmental policy intervention.

Survivors of Super Typhoon Yolanda in Tacloban City, Philippines, 2013. (cc) UN Photo/Evan Schneider

Survivors of Super Typhoon Yolanda in Tacloban City, Philippines, 2013. (cc) UN Photo/Evan Schneider

Over the years, researchers have identified a number of root causes that increase disaster risk—such as weak building codes and land use policy enforcement and overemphasis on ex-post emergency response as opposed to proactive management of disaster risk. Also, decades of economic studies looking at the costs and benefits of risk reduction investment show that such investment often pays off in the longer run. Yet, as the recent global trends of rising disaster risk unfortunately testify—we are far from learning these lessons effectively, or at least fast enough to beat the rising risk posed by future climate change: Global annual average disaster loss is estimated to have risen to approximately $300 billion in 2015 according to the UN Office for Disaster Risk Reduction (UNISDR).

As the special representative of the secretary general for disaster risk reduction, Robert Glasser wrote in the Guardian last week, “Every time there’s a mega disaster, there are lessons learned…  The key question is always, how do you keep up the awareness after a couple of years?”

That is why the IIASA Risk and Resilience program’s research is increasingly focused on cognitive, behavioral, and governance aspects of societal learning on disaster risk reduction. We are currently working with public, private, and civil society stakeholders, asking the questions of why we, as a collective society, continue to fail to act on these lessons learned in disaster risk management and what we can do to change it. By combining both quantitative and qualitative systems analysis approaches, we are untangling why we make decisions the way we do, and what processes and institutional mechanisms directly and indirectly affect disaster risk and developmental outcome over the long term.

Given that catastrophic disasters are by definition rare events (hence opportunities for learning is naturally limited), we are doing this using novel methods such as participatory gaming or policy exercises in which we create virtual opportunities for stakeholders to experience complex decision-making in a safe learning environment. By creating stylized context for common decision-making (such as rural farmers making longer-term decisions on livelihood diversification, or urban planners addressing rising disaster risk due to rapid population growth), these gaming spaces serve as mechanisms through which stakeholders can not only learn about their cognitive and behavioral assumptions, but also through which learning can be accelerated, repeated, and shared among different communities facing similar development and disaster risk reduction  challenges.  We are running such policy exercises in the context of our flood resilience project and internal gaming project .

Decades of research have shown that there are common global lessons on development and disaster risk reduction but they are not so easily learned in practice. It is too often that that the windows of opportunities for policy learning are limited and we continue with business-as-usual of “lessons unlearned.” Creating an enabling environment for iterative learning is no easy task under these pragmatic constraints, but we hope that a bit of creativity and lots of hard work will eventually pay off in the long run.

Note: This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.