=Paper=
{{Paper
|id=Vol-1952/Disaster
|storemode=property
|title=Open Data And Disaster Management
|pdfUrl=https://ceur-ws.org/Vol-1952/Disaster.pdf
|volume=Vol-1952
}}
==Open Data And Disaster Management==
https://ceur-ws.org/Vol-1952/Disaster.pdf
Open Data and Disaster Management
Ditsuhi Iskandaryan
Universitat Jaume I
Spain
ditsuhiiskandaryan@yahoo.com
Abstract
This paper is describing the role of open data during disaster management.
The frequency and intensity of disaster occurrence force to focus on
management and planning. Day by day more data is becoming open, which
add transparency and support decision makers to react faster in an extreme
situation. The method was based on research and comparison between
countries, tools, services and apps. The result shows with new technologies
continuous improvements are doing for covering existing gaps, such as
handling with big data, analyzing faster.
1 Introduction
Disasters, such as earthquake, flood, hurricane and so on, are events which occurred suddenly and cause human, economic
and environmental damage. Population growth, spread of disease, climate change affect on frequency and intensity of disaster
occurrence. During Haiti earthquake disaster took 220,000-336,000 lives[7], during Hurricane Katrina it took 1245-1836
lives[8]. For decreasing these numbers, for sustainable development it is important to concentrate on disaster management.
Organizations and agencies working on disaster management need to collaborate with partners, find more data, find a way
for mitigating losses and risk. Open data is one of the valuable sources, which plays crucial role in disaster management.
Open data is defined as data that can be freely used, re-used and shared by anyone. Data to be open must be open
legally and technically. First one refers open data license which allows anyone freely to access, reuse and redistribute [9,10].
There are many types of license like Creative Commons License, Open Data Commons, the Open Government License and
so on[3]. Technically open means that data must be machine-readable and in bulk form[9,10].
2 Use cases
In 2010 during Haiti earthquake because of the lack of data many issues were appeared. It was difficult to find people who
needed help. After two days Google, GeoEye and DigitalGlope together got high resolution image which was widely used.
But still there was a lack of data and volunteers realized that they can help from distance. They started to do online mapping.
After a few weeks in Open Street Map were done near 10,000 edits to the Port-au-Prince region. Another service, that was
widely used, is Ushahidi. It was created in 2008 in Kenya for reporting violence during election, however it became more
popular and during Haiti earthquake many volunteers reported SMS, MMS geotagging in interactive map. The fact that
people could report SMS was one of the main advantages of Ushahidi, because only 11% has access to internet and more than
30% has mobile phone[1,6].
Another case is about Colombia. In the beginning of April in this year huge water flood and landslide almost
destroyed a town Mocoa (Putumayo) in south of Colombia. More than 200 people died, 500 families were affected by this
disaster. There were issue related to available open data and it was difficult to generate data in a short time. Some authorities
Copyright © by the paper's authors. Copying permitted for private and academic purposes.
do not have the capacity to generate data in the speed that the disaster needs. The Open Street Map and the Mapping Humanity
community worked together to map the affected zone. They used social networks, some data which are available in the official
portals. After several days they finished mapping task of the area and the output was used even by the official authorities for
saving lives, for distributing the resources[11,12,13].
3 Challenges and barriers
Despite the fact that the role of Open data is crucial, there are many challenges which need to solve in future. Challenges are
related to access, usage, dissemination, data collection. For decision making it is important to have up-to-date information
about emergency situations, to collect data regularly and filter these huge data. Related to the latter one is remarkable to
mention about Karnataka state natural disaster monitoring center(KSNDMC) in India. India is considering to be as vulnerable
country to disasters. Only this year more than ten disasters happened, including earthquakes and landslides. The main function
of KSNDMC is to collect data, analyze these data and to visualize output on map. But there were problems regarded handling
large data, real time analysis and conveying result to end user. However, later using ESRI technologies, such as Arc SDE
Spatial Database, ArcGIS Desktop, ArcGIS Server and Web GIS Technology, they overcame these issues[14].
It is necessary to mention also about accuracy, completeness and dependency of volunteer geographic information.
Volunteers using their mobile devices can provide near-real-time information about some events (including electronic reports,
pictures, videos, location), they are more active and information provided by them is increasing hugely. However, they are
with different background, they can report some information about certain geographic problem without deep knowledge. So,
for decision makers it is very important to check accuracy and one of the way is to compare these data with reference dataset
(for example compare with dataset provided by government) and after analyze the differences[2].
During Haiti earthquake there was duplication of effort, there was a problem of combination data created by different
software. Researcher from the University of Virginia worked on one project, which aim is to find way for increasing
trustworthiness of crowdsourcing data. In their web site first strategy is classify group membership (e.g. police is more trustful,
than unknown group). The second strategy is about ranking of the posts. Each post is tagged with five-stars rating system and
user can evaluate a post after reading They also suggested to comment report and to rate viewer. Actually, these suggestions
are more technical and mainly depend on viewer activity[5].
The team from University of Munster did survey to understand which type of crowdsourcing is more usable by
asking questions to experts from different organizations related to disaster management. The result showed that most of
disaster managers used Twitter, but mostly for broadcasting not for collecting data. Half of the participants was aware about
Ushahidi, but only 21% was using this. Experts brought several reasons, such as uncertainty, trust and semantic problems.
According to these experts they need some improvements in training of volunteers, filtering and rating of information and
etc[4].
While disasters occur disaster management will stay actual discipline for study and for improvement. One big
achievement is the creation of following free software such as InaSAFE[15] and WebSAFE[17]. The aim of these software
is, using data from scientists, local governments and communities, to generate consequences of future disasters.
4 Conclusion
As a conclusion we can say Open Data is valuable tool for disaster management. It can help to make quick and effective
decision. When data is not fully available volunteer geographic information with existing open data can save lives, resources
and mitigate losses. Also, we have to take account that early-warning systems can save and protect huge numbers of lives.
During Cyclone Phailin in 2013, when new early-warning system was applied, forty five people died, In 1999 during the
same size storm were died 10,000 people[16]. So, we can see the benefit of early-warning system.
References
[1] J. Heinzelman and C. Waters, “Crowdsourcing Crisis Information in Disaster,” Phys. Rev. Lett., vol. 96, no. 25, p.
258102, 2010.
[2] S. Jackson, W. Mullen, P. Agouris, A. Crooks, A. Croitoru, and A. Stefanidis, “Assessing Completeness and
Spatial Error of Features in Volunteered Geographic Information,” ISPRS Int. J. Geo-Information, vol. 2, no. 2, pp.
507–530, 2013.
[3] N. Korn and C. Oppenheim, “Licensing Open Data : A Practical Guide,” no. June, pp. 1–8, 2011.
[4] J. Ortmann, M. Limbu, D. Wang, and T. Kauppinen, “Crowdsourcing Linked Open Data for Disaster
Management,” Proc. Terra Cogn. 10th Int. Semant. Web Conf., no. January, pp. 11–22, 2011.
[5] A. C. Weaver, J. P. Boyle, and L. I. Besaleva, “Applications and trust issues when crowdsourcing a crisis,” 2012
21st Int. Conf. Comput. Commun. Networks, ICCCN 2012 - Proc., 2012.
[6] M. Zook, M. Graham, T. Shelton, and S. Gorman, “Volunteered Geographic Information and Crowdsourcing
Disaster Relief: A Case Study of the Haitian Earthquake,” World Med. Heal. Policy, vol. 2, no. 2, pp. 6–32, 2010.
[7] https://en.wikipedia.org/wiki/2010_Haiti_earthquake
[8] https://en.wikipedia.org/wiki/Hurricane_Katrina
[9] http://opendatahandbook.org/guide/en/what-is-open-data/
[10] http://opendatahandbook.org/glossary/en/terms/open-data/
[11] http://blog.openstreetmap.co/2017/04/02/llamado_mocoa/
[12] https://www.humanitarianresponse.info/node/77/search?search=mocoa
[13] http://pierzen.dev.openstreetmap.org/hot/leaflet/OSM-Compare-before-after.html#14/1.1455/-76.6512
[14] http://geoithub.com/esri-indias-arcgis-disaster-management-karnataka/
[15] http://inasafe.org/
[16] https://www.bloomberg.com/view/articles/2014-02-18/seven-steps-to-surviving-a-disaster
[17] http://www.rappler.com/science-nature/environment/62263-indonesia-philippines- disaster-risk-reduction-gis