During and in the aftermath of a natural disaster, reliable communication plays a vital role to ensure efficient first response and provide critical information to people in the impacted areas.
OpenSignal measured the experience of smartphone users before and in the aftermath of hurricane Florence, which hit North Carolina on Friday, September 14. Florence, which was downgraded to Category 1 just before marching ashore near Wilmington, battered the area for days producing widespread flooding, damaging hundreds of buildings and affecting thousands of residents.
We focused our analysis on the cities of Wilmington, Jacksonville, Myrtle Beach and Fayetteville, both before and during the storm. We are able to see the degree to which the hurricane affected the mobile experience on a daily basis, for example: whether connectivity suffered and the extent to which residents chose to shelter from the storm.
LTE availability for smartphone users was negatively impacted for five days
Our analysis shows that the percentage of time smartphone devices connected to LTE networks – LTE availability – decreased from the previous thirty days’ average of 95% to the minimum value of 86.8% on Sunday, September 16. Users’ LTE experience was adversely affected for five full days, and only returned to normal levels on Wednesday, September 19.
Although the widespread flooding likely increased the challenges in restoring the downed cell sites, thus extending the time mobile operators needed to fully restore the network, our analysis shows users could still connect to LTE services more than four fifths of the time.
Overall mobile connectivity remained strong during the hurricane
OpenSignal analyzed more than 10 million measurements during the thirty days before the storm hit North Carolina and the week after. As we expected, the number of active smartphones in the area dropped, suggesting that many people heeded the authorities’ warnings and evacuated the area. Nevertheless, two thirds of the smartphones remained in the area, showing that a sizeable part of population decided to stay at home.
Hurricane Florence’s impact on the LTE network experience of smartphone users lasted from Friday, September 14 – the day the Hurricane landed ashore – to Wednesday, September 19. However, the sum of 2G, 3G and 4G availability remained close to the previous thirty days’ average of 99.3%, hitting its lowest on Sunday, September 16 at 97.7%. With most of the smartphones usually connecting to 4G, when the hurricane downed a number of LTE cell sites, an increasing number of mobile users connected to the available 3G and 2G networks, showing that although users could not experience the best speed, they still had access to mobile technology.
When a natural disaster occurs many people either don’t have the time or resources to evacuate the danger zone. As a result, a durable mobile network experience is critical to ensure smartphone users continue to enjoy communication services and mobile data during the emergency. Even though extreme weather conditions can negatively impact the smartphone experience, U.S. mobile operators proved that they could ensure mobile connectivity during hurricane Florence.
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, 10,073,034 measurements were collected from 1,499 devices in the cities of Wilmington, Jacksonville, Myrtle Beach and Fayetteville during the period: Aug 14 - Sep 20, 2018.
We aggregated the data on a daily basis and compared mobile experience during the days following hurricane Florence hitting North Carolina, versus the average experience of the previous thirty days.
We have calculated statistical confidence intervals and plotted them on the graph. When confidence intervals overlap for a certain period, our measured results are too close to declare a statistical difference.