Day: December 6, 2019

There are a number of changes to IPAWS that have been in the works for years now and are finally close to being released by FEMA. These include: 

• Increasing the maximum character count from 90 to 360.
• Adding support for Spanish language WEA.
• Adding two new alert categories:

1. Public Safety Message – for less severe situations.
2. WEA Test Message – supports state and local WEA testing.

• Geo-targeting improvements to reach 100% of an area with no more than 1/10th of a mile overshoot.
• Messages stay on the receiving device for 24 hours unless deleted by the user. 

These changes were supposed to go into effect at the end of November, but have been delayed by FEMA. While we don’t yet know the “go live” date, we’re hoping it’s a matter of weeks, rather than months.

Here are a few points to keep in mind about these changes: 

• Every aspect of IPAWS WEA messaging is potentially affected by each of the components of the system. These include: 

1. The IPAWS origination software. If you’re using Hyper-Reach, you can be assured that we’re on top of IPAWS requirements and are either supporting all new changes or will in very short order. After all, we’re one of the leading proponents of IPAWS among ENS providers: we were one of the first to implement IPAWS and have the highest rate of IPAWS adoption among our customers compared with other major ENS providers.
2. The IPAWS network, as provided by FEMA. This is where the changes we’re expecting are in process.
3. The mobile carriers, such as AT&T, Verizon, T-Mobile, etc.
4. Individual mobile devices. 

• The different parts of the system affect these changes in different ways.  For example: 

1. The geo-targeting improvements depend heavily on the carrier’s ability to support them AND the capabilities of individual mobile devices. As a result, the “1/10th of a mile overshoot” change will only be effective on mobile devices that can filter out messages based on their location information. Devices that don’t have the ability to filter or which don’t have location information will receive the message if they’re in the broadcast area of a cell tower in the polygon selected for message delivery.
2. The character count expansion to 360 depends on the origination software, the carrier’s ability to deliver 360 characters and the device’s ability to display that character length. We’ve seen reports that some carriers won’t be able to support 360 characters right away. 
3. The testing components mean that FEMA may start requiring monthly testing once those components are available.  Currently, FCC rules require getting a limited waiver for end-to-end testing and that requirement will go away once the full testing components are available. 

Although these changes have taken a long time to come – the FCC issued rules for some of them more than three years ago – they will provide significant improvements in the usefulness and effectiveness of IPAWS. We’re looking forward to them and are committed to continuing to improve your experience in using this valuable service.

If you’re already a Hyper-Reach customer, you’re probably using our template feature. Templates make it easy to standardize messages by letting you create stock messages where you just fill in the blanks. 

There are many great reasons to use templates. They improve the quality and consistency of messages and they can help you get a message written quickly. And you can use templates to ensure that you follow best practices, such as including all critical information.

You can also use templates to make sure that your messages are a reasonable length. While Hyper-Reach can deliver longer messages than you would normally use, in most circumstances, shorter is better, all other things being equal.

To make it easy for you to design templates that work for you, we’ve put together a little Google sheet that you can find here. You can use it to design your template and also test it with real-world examples of the kinds of messages you would want to send. 

We hope this spreadsheet is helpful. If you want us to make any changes, just drop us a line at jveilleux@ashergroup.com. We’ll be glad to modify the sheet to help you.

Recently, we came across this article about detecting landslides in India, which kill more than 900 people annually. It got us thinking that something similar might work in the US. The article describes a way to modify a smartphone using its accelerometer to monitor for movements in the soil – for less than $300. Early indications are that it works: in one case, by warning officials to close a road that was later washed out.

Landslides are less common and less deadly in the US, killing about 25 people per year on average.

But the danger from US landslides is growing. Because of development and climate change, the conditions that cause landslides are getting worse, including rerouting of surface water runoff, severe rainfalls, and wildfires that destroy tree cover. As a result, experts are predicting that that landslides will become more common and more damaging. Here in the US, landslides are more dispersed – occurring in all 50 states – which makes it more difficult to focus resources on the areas of greatest risk.  

Fortunately, artificial intelligence and other techniques are being developed to help predict the areas most likely to develop landslides. The US Geological Survey has released a new database which identifies many areas with higher projected landslide risk. And there are new tools in development. One research project in Central America has developed a model to identify areas most at risk, with promising initial results. Closer to home, researchers at Carnegie Mellon University are working on machine learning algorithms with the express purpose of identifying high-risk areas in Pennsylvania for infrastructure investment. 

It’s not difficult to imagine a system of early warning sensors like the ones in India, dispersed in high risk areas around the US, which could then send automated alerts in a method similar to the ones sent by the National Weather Service. For example, there’s a new system developed for earthquake alerts in California. But earthquakes kill less than a third as many Americans as landslides do – less than 8 people per year over the last century. So when you hear about a new system for sending landslide danger alerts, remember that you first heard about it here.