How to mitigate flooding in Bandung City

Almost one year ago, Bandung, Indonesia, experienced a devastating flood. The West Java Regional Disaster Mitigation Agency (BPBD) reported an estimated 16 billion rupiah (1 million USD) in damages and losses from this flood alone.

Despite floods becoming more regular weather events, flood preparedness and response remains a formidable task for local authorities. As one of the leading smart cities in Indonesia though, Bandung City government has a strong commitment to nurturing new technologies.

The city has already installed more than 350 CCTV cameras, and specifically for monitoring weather conditions, 27 new sensors in sub-district offices are still to come.

These advances motivated the question: How can the City Government harness data from new technologies to better mitigate repeat flooding?

To answer this, the Pulse Lab Jakarta team embarked on a six-day exploration in Bandung City.

We met with a handful of individuals from roughly a dozen organisations, many of whom were involved in waste management and water management, and a few were everyday citizens who are on the frontline. Here are our key findings:

1. The Players — Flood mitigation is a joint-effort; it is one that requires cross-institution and cross-administration coordination, covering the scope of preparedness, response as well as recovery efforts. Because of this broad scope, we decided to group the participating organisations according to their involvements and functions.

Ultimately, we ended up with five unique groups: the thinkers, the coordinator, the data collector, the monitor and the field actors, which as a whole encompasses a complete set of skills and expertise needed to effectively design information-driven flood mitigation strategies.

2. The Data — While existing throughout different government units, without a sense of holistic coordination, the available data related to floods cannot be properly utilised.

For instance, BMKG (Meteorological, Climatological, and Geophysical Agency) provides information on weather and rainfall forecasts from weather stations and sensors, while Bandung Public Works Agency monitors floods from CCTV cameras installed in seven flood-prone areas.

In an additional scope, certain sub-districts also monitor their own regions during similar events using CCTV cameras. From these examples, it is evident that Bandung City has reasonable capacity to gather data and information for flood mitigation. Nonetheless, a fully integrated system is lacking.

3. Data Utilisation — Despite the data sources dedicated to flood mitigation, the data gathered has not been fully leveraged by the community.

When BMKG officials provided information on rainfall forecast in the past, for example, some citizens complained that the information was overly technical.

As a result, the tendency is to ignore these official forecasts. Some village officials also mentioned that they prefer to simply look at the clouds and then make a judgment based on instincts.

So here, our team noticed the challenge of trying to socialise reliance and utilisation of the data available, as well as how to simplify it in a way that it can be understood by citizens.

4. Point of Contact — The free online messaging platform, WhatsApp, is the preferred medium for disseminating information. Favoured by the key stakeholders, the application is said to be effective and reliable, especially when coordinating large group activities.

Typically, each key player will be an active member of more than one WhatsApp group (on average they are in 6–8 different groups related to their responsibilities) in order to update information daily and even more frequently during a crisis.

It allows for the communication of instant updates about floods via images, texts and voice recordings, the broadcast of emergency alerts, as well as the coordination with group members, all at once.

These findings helped us put together the following recommendations:

1. Understanding the existing data journey is important

The data journey is the mapping of data from collection to usage. In Bandung City, we outlined the journey based on existing practice. What is often overlooked is that the data journey may not always reflect the regulations or standard operation procedures.

By reconstructing the actual flow of data, we were able to observe the different role of each actor and identify the recurring pain points.

Understanding data governance within the current decentralised context was also important. More effective flood mitigation strategies will rely on inter-government and inter-agency cooperation, which means data sharing is a necessary component to consider.

At the moment, the provincial government is well-positioned to facilitate this aspect due to its existing cross interaction with the various administrations.

2. NGOs should work with the subdistrict heads

As we have established, many government units have their own data collection methods. The downside is that they are not always integrated with other units.

And while tech-focused alternatives to the traditional quantitative and qualitative data collection methods are now being explored, the next step is to think through how to employ these methods not just for planning but as well for timely response during crisis.

The subdistrict head is normally well-connected to different government units and local communities, thus this individual already embodies the role of field coordinator.

These sub-district heads as government officials are especially vital for data verification and utilisation in the event of a flood. Still, their roles will need to evolve and adapt as new tools and new data sources become available.

Our hope is that from these findings as well as from others that may emerge, we can foster collaboration with other stakeholders in order to develop useful prototypes that can help mitigate repeat flooding in Bandung City, including designing both short-term and long-term measures.

This blog is published as part of our #DataCollab Series. #DataCollab is a data innovation collaboration for Bandung City, initiated by HIVOS, Data Science Indonesia, Pulse Lab Jakarta and Code4Bandung. Contact us if you are interested in joining our collaboration.