Reliable communication is key during natural disasters
On Sunday 16 September, Hong Kong was battered by Typhoon Mangkhut, the strongest tropical storm to hit the city in recent decades, with strong winds up to 103 mph (166 kph) that smashed windows and teared off parts of buildings and roofs, leaving more than 100 people injured but thankfully no fatalities.
OpenSignal measured the experience of Hong Kong smartphone users both before and during the storm. We were able to see the degree to which the typhoon affected the mobile experience on an hourly basis, for example: whether connectivity suffered, and the extent to which Hong Kong residents chose to shelter from the storm.
The city had been preparing for the worst, with the local authorities issuing the highest typhoon warning — a signal 10, warning residents to take refuge at home, and deciding to temporarily shut down air travel and the city’s bus and railway services.
When a natural disaster occurs, reliable communication plays a vital role in ensuring efficient first response and providing critical information to people in the impacted areas.
During the storm, many people stayed at home connected to Wifi
The number of devices in the Hong Kong area contributing measurements to our analysis did not vary significantly in the wake of the storm. This suggests that the majority of Hong Kong residents decided to face the typhoon by taking refuge at home instead of evacuating to mainland China or elsewhere.
Mobile users often connect to Wifi networks when they are in familiar locations where they spend a considerable amount of time, such as at home or at work, and these locations are likely to be indoors.
Analyzing OpenSignal’s time on Wifi metric, which represents the percentage of time mobile devices are connected to Wifi networks, a clearly defined daily pattern appears for Hong Kong. Normally, time spent on Wifi peaks during late evenings and is at a low in the middle of the day, suggesting the metric broadly reflects the daily routine of going out to work or play, and then returning home in the evening where people are most likely to stay connected using Wifi.
As we expected, weekdays and weekends emerge from the analysis with two distinct trends. During the central hours of the day, weekdays present three negative peaks at 8:00, 13:00 and 18:00, likely representing commuting and lunch times, while weekends show a smoother curve, reaching a minimum around 14:00, suggesting the citizens are still most likely to go out during the warm hours.
Focusing now on the time Hong Kong residents spent on Wifi during the typhoon, it is evident that Sunday, September 16 was a day out of the ordinary. OpenSignal data shows how smartphones in the city were on average connected to Wifi more than 65% of the time throughout the day of the typhoon. On a normal weekend, time on Wifi would drop to below 40% in the middle of the day. This is clearly a different pattern compared to other weekends, suggesting the vast majority of the residents heeded the authorities’ warnings and took shelter at home.
Mobile networks remained available despite the strength of the storm
OpenSignal analyzed more than 47 million measurements collected from over 5,600 mobile devices during the thirty days ending on Sunday, September 16, when the typhoon hit Hong Kong. 4G availability suffered some small bumps during the central hours on Sunday, but the time smartphone users spent connected to 4G did not fall below 90%, which is a remarkably good result given that the average 4G availability during the previous month was 91.6%. This mobile user experience demonstrates the resilience of Hong Kong’s mobile infrastructure.
Such consistent availability could also be partly explained by the strong time on Wifi shown during the storm. In fact, we assume that much of smartphone users’ data traffic was absorbed by the fixed (Wifi) infrastructure, easing the strain on mobile networks and facilitating a good mobile experience for those smartphone users who did venture outside of Wifi range.
Overall, OpenSignal's data indicates that Hong Kong’s smartphone users did not face a significant loss of mobile network experience during the great storm. Mobile proved to be a durable and reliable means of communication, even during extreme weather conditions such as Typhoon Mangkhut.
When a natural disaster occurs many people either don’t have the time or resources to evacuate the danger zone, and a durable mobile network is critical to ensure communication is available during the emergency. On September 16, Hong Kong’s mobile operators proved they can deliver a good smartphone user experience even when facing the most challenging weather conditions.
OpenSignal measures the real-world experience of consumers on mobile networks as they go about their daily lives. We collect 3 billion individual measurements every day from tens of millions of smartphones worldwide. Our measurements are collected at all hours of the day, every day of the year, under conditions of normal usage, including inside buildings and outdoors, in cities and the countryside, and everywhere in between. By analyzing on-device measurements recorded in the places where subscribers actually live, work and travel, we report on mobile network service the way users truly experience it.
For this particular analysis, 47,841,439 measurements were collected from 3,492 devices during the period: Aug 17 - Sep 16, 2018.
We aggregated the data on an hourly basis and compared mobile experience during the day Typhoon Mangkhut hit Hong Kong, versus the hourly average experienced during the previous thirty days.
We have calculated statistical confidence intervals for every hour and plotted them on all of the graphs. When confidence intervals overlap for a certain period, our measured results are too close to declare a statistical difference.
Opensignal, Inc retains ownership of blog articles including all intellectual property rights, data, content, graphs & analysis. Blog articles produced by Opensignal, Inc may not be quoted, reproduced, distributed, published for any commercial purpose (including use in advertisements or other promotional content) without prior written consent.