Contd from previous issue
The preparation of National Disaster Management Plan in 2016, in line with theSendai Framework for Disaster Risk Reduction 2015–2030, affirmed the country’s strong determination to effectively counter the growing frequency and intensity of disasters.
Today, the country has a wide range of institutions related to disaster monitoring and forecasting, which have enhanced disaster management activities. However, one of the biggest drawbacks in managing the disasters in India, as identified by the Ministry of Home Affairs, is the lack of data availability, the study notes.
Potential focus areas for app-based crowdsourcing in various stages of disaster management cycle. Credit: Rajib Shaw/Mongabay
The lack of data in the context of disaster resilience, which can be plugged through app-based crowdsourcing, is in sharp contrast to the smartphone and internet usage boom in the country, researchers said.
India has the second-largest internet user base after China, with over 432 million subscribers and with more than 300 million smartphone users, notes the 2018-’19 annual report of the Ministry of Electronics and Information Technology. The number of internet users in India is expected to reach 829 million by 2021.
The key shortcomings that are limiting the large-scale use of these disaster-related apps are bad application interface; lack of inclusivity in terms of language, age and gender; lack of communication infrastructure; and absence of routine communication channels.
“These challenges can only be overcome when the user sees some value in using the apps against their regular communication scheme. For example, visualisation of disasters in maps, such as water inundation during flooding or building collapse during an earthquake, or a one-stop-shop format where the application shares multiple information such as awareness, what to do, early warning and citizen science,” explained Shaw.
Additionally, there is a need to work towards making the current disaster-related apps more user-friendly with simple login features and easy-to-understand processing. According to Saptarshi Ghosh of IIT Kharagpur’s Department of Computer Science and Engineering, well-designed disaster management apps can be very helpful when disaster strikes, but the main problem is to have trained volunteers to use those apps.
“It is not possible to develop this know-how during an actual disaster. A serious effort should be made to train people to use these apps during normal times so that they can be easily used at the time of disasters,” Ghosh, who was not associated with the study, said.
Role of crowdsourcing
It is not as if app-based crowdsourcing of data during disasters has not seen some measure of success in India, despite persistent issues like the cost of building the app and handling crowdsourced data.
For instance, the scope of citizen science in aiding government authorities to assess the scale of damage and mobilising required personnel came to the fore during cyclone Hudhud in 2015, when the government of Andhra Pradesh, with the help of National Remote Sensing Centre, launched an Android app for a crowdsourcing project where people could upload photos from their smartphones.
“Over 3,000 downloads of the app were recorded. The uploaded pictures were automatically integrated with NRSA data and helped the government in assessing the damages and mobilising required men and material to reach the last point,” said a report.
“GIS, GPS and remote sensing technologies were used to spot the damages. The government engaged four satellites to monitor cyclone – RI Sat1, RISat2, Radar Sat and Resource Sat2,” according to the report. Usage of apps such as PetaBencana.id and Ushahidi are two examples of crowdsourcing during disasters despite their limitations.
PetaBencana.id is a crowdsourcing tool that combines citizen reports over social media, mainly Twitter, and other apps, with hydraulic sensor data to produce real-time flood maps in Jakarta. It is now being scaled up to other cities.
Ushahidi is an open-source crisis-mapping software that was first developed in Kenya to chart the 2007 election violence. The application of this platform post-2010 earthquake in Haiti was one of the first major examples of mobile phone-based crowdsourcing being used in disaster response.
“Japan has mobile apps that work [alert you] even when your phone is off the ringer mode. One important app is for lightning alerts, which gives location-specific alerts a few minutes ahead of a lightning strike,” said Shaw.
In India, Karnataka State Natural Disaster Monitoring Committee, has, for example, launched an app that provides early warning for lightning strikes through colour-coded messages.
Shaw said that trials are underway for a smartphone crowdsourcing application, which requires students and residents of Varanasi to snap photos during urban flooding incidents and upload them onto the app. “Rainfall pattern is changing and the northern part of the city receives sudden, heavy rainfall which inundates pockets. We are engaging the local to take photos of the inundation and to upload on the app. We are able to map flooded areas. This system helps local traffic officials to divert the traffic in another direction,” said Shaw.
Further, integrating disaster-related operations with dedicated open platforms – like in the case of Ushahidi – can boost relief efforts humanitarian organisations. By sifting through crowdsourced data such as photos and posts, organisations can gain an idea of the situation on the ground and reach out to affected communities faster and in a targetted manner.
India’s draft report to UN notes that India has witnessed many tropical cyclones in recent years. Credit: Indian Navy/Wikimedia Commons
Filtering relevant data
While apps based on crowdsourced data are yet to take off on a larger scale in India, the Nepal earthquake in 2015, the floods in Chennai in the same year, and the 2018 Kerala floods have brought to the fore the increasing effectiveness of online social media – Twitter, Facebook, Instagram and others – and messaging services in disaster resilience in South Asia.
“SMS [and] social media are definitely useful. It is not [an] either/or choice. But apps can enhance citizen participation with specific target groups and users for a specific location, which is found to be more effective in disasters since most of the events are local in nature. So an app [as an internal link], in conjunction with SMS, social media [as an external link] will be useful,” explained Shaw.
“An app can help you locate nearby shelter[s], emergency centre[s]...which Whatsapp or SMS services can not do,” Shaw added.
IIT-Kharagpur’s Saptarshi Ghosh said while posts on social media and internal messaging services could be extremely helpful for carrying out relief operations – or planning in advance for disaster management – only about two percent of the information tweeted in a disaster scenario turns out to be useful for relief operations, which is humanly not possible to identify.
Ghosh and colleagues have developed algorithms that can extract critical information in real-time and in summarised form, which could come in handy in disaster management. The critical information is immersed in large amounts of conversational content where people mostly express sympathy for the victims of the disaster.
The team is developing web-based systems and mobile applications for aiding post-disaster relief operations. The systems will utilise the algorithms and perform tasks such as identifying and extracting actionable information and summarising the actionable information.
“The technology has reached a level where it can be deployed for use by relief operators. For instance, a person sitting in the control room can get live updates about what resources are needed in which location [and] what resources are available, and coordinate the relief operations accordingly. Or a relief worker can get updates on his or her smartphone about people being trapped in the vicinity, so that they can be rescued,” explained Ghosh.
This article first appeared on Mongabay.