Lost in Translation

By: Melissa Huffman

Social science has really risen to prominence recently in operational meteorology, given the frequency of devastating natural disasters the United States experiences and the increasing number of people being affected by them. This is pretty cool when you think about it… two seemingly unrelated fields working together to help keep people safe.

One of my favorite time-wasting activities is typing random words into Google Translate to see what they look like in different languages. Admittedly, the translations aren’t always perfect but it gives me a good idea of a few phrases I can use if the need arises. (I live in a pretty diverse city… just go with the metaphor.)



While more sophisticated, operational meteorologists are also translators. We help people figure out what weather to expect so that they can make decisions based on that information. Do I need to wear a jacket? Will my kid’s soccer game be rained out? The sky looks really dark over there; I wonder if that means bad weather is coming.

Translating weather information is not an intuitive process, however. Operational meteorologists are very good at recognizing what lots of complicated atmospheric features represent. However, a description of thermodynamic and kinematic information obtained from radiosonde observations in advance of a major tornado outbreak doesn’t really tell a community that today is a day they need to be prepared to take shelter. We need help in finding the best way to translate all the data we analyze into something people can use.

Enter social science.

Social science has really risen to prominence recently in operational meteorology, given the frequency of devastating natural disasters the United States experiences and the increasing number of people being affected by them. This is pretty cool when you think about it… two seemingly unrelated fields working together to help keep people safe. There’s a big problem though with this team that a lot of people don’t seem to want to talk about: these two fields, which work together to translate weather information to the public, are speaking two different languages to each other.

We both think we know what aspect of the human condition or hazard characteristic the other field is describing, but sometimes we just don’t. And we’re not asking each other to clarify. One of the best examples of this centers around the concept of using impacts to convey weather information. The importance of this concept stems from research conducted by the social science side of the team, which has evaluated qualities of effective warning messages and found that “more specific messages produce higher levels of warning belief and perceived risk” (Perry, Lindell, and Greene 1982[1]). Using impacts to describe hazards can provide contextual information for individuals who have never experienced a hazard before and also help elevate the urgency associated with warning messages, should they be required. This helps increase the likelihood that an individual will respond to a warning message. This is the goal of our translator team: keeping people safe. This importance has been emphasized for operational meteorologists in recent Service Assessments from the National Weather Service. For example, Recommendation 5 of the Hurricane/Post-Tropical Cyclone Sandy Service Assessment[2] states:

The NWS should develop more effective and consistent products to communicate severe weather impacts, specifically: 

Concise summaries of weather and its impacts using non-technical text and graphical material provided in a short and easy-to-read format 

Confidence or uncertainty and worst-case scenario information

These products should be pretested using evidence-based social science. NWS should also provide effective training on the use of these products to ensure WFO personnel know how to best communicate with decision makers and the public during extreme weather events.

This isn’t the only Service Assessment in which a similar recommendation has been presented. In fact, the 2011 Joplin, MO Tornado, the April 2011 Historic Tornado Outbreak, Hurricane Irene, the June 2012 Historic Derecho, the May 2013 Oklahoma Tornadoes and Flash Flooding, and the 2015 Historic South Carolina Floods all contained findings and/or recommendations relating to the communications of impacts. That’s just a sample of recent Service Assessments too.

So here is our problem… what is an impact?

Ask a social scientist and you may get a complex description of different factors that could enhance or lessen the effects of an external event on a community. Maybe a summary of physical impacts from analog events, such as fatalities, injuries, property damage, or even an analysis of more social impacts like psychosocial, sociodemographic, socioeconomic, and sociopolitical issues that continue long after an event has ended.

Ask an operational meteorologist and you may get something closer to a more quantitative description of environmental characteristics surrounding an anticipated event. Maybe phenomena that could be observed within a specific time window, like hail, damaging winds, tornadoes, or winter weather:


Example: Impact Communication for Winter Weather


Example: Impact Communication for Tropical Weather

Both of these groups have ideas of what impacts are that are close enough, but not quite the same. Operational meteorologists are being told to communicate impacts, but there are instances in which what we perceive the impacts to be are different than how social scientists would have us communicate them. What does 2 inches of snow on a major interstate represent for transportation groups? What could baseball size hail mean for agricultural interests?

Social scientists do a fantastic job of conducting studies about how meteorologists communicate impacts. In April 2016, the NOAA Social Science committee even released a report[1] to help summarize some of these studies as well as effective risk communication practices through communicating impacts. However, there’s an inherent assumption by social scientists that operational meteorologists know what impacts will occur in a community… or even how to determine what those impacts are.

Operational meteorologists can use analog events and previous experience to anticipate some of the impacts, no doubt, but communicating impacts wouldn’t be a recurring issue if operational meteorologists and social scientists were on the same page. That’s not to say that efforts are not being made by both groups to communicate impacts; the National Weather Service has moved to providing Impact Based Warnings[2] to provide a better explanation of damage potential associated with tornadoes and severe thunderstorms. Social scientists and meteorologists have also worked together to produce Potential Storm Surge Inundation Maps[3]; these represent the storm surge heights that a person should prepare for before a storm, given the uncertainties in the meteorological forecast.


Example: Potential Storm Surge Inundation Map

More can be done though. We both have a novice understanding of each other’s language, but we’re not fluent yet.

Fluency is not a one way street. More dialogue between social scientists and operational meteorologists has to occur. We have to go beyond vaguely understood phrases and conceptual models for public response and work to make sure we can communicate with each other. We have to recognize the constraints that each other operates within and the skills that each other brings to weather communication in order to be able to serve our communities. When you learn a language, fluency is not reciting vocabulary words out of a textbook or using an online translator; it’s being able to fully converse with another person. Think of how powerful our communication of weather information will be when we’re both speaking the same language.


[1] Perry, Ronald, Michael Lindell, and Marjorie Greene. 1982. “Threat Perception And Public Response To Volcano Hazard”. The Journal Of Social Psychology: Vol 116, No 2.

[2] http://www.nws.noaa.gov/os/assessments/pdfs/Sandy13.pdf

[1] http://www.ppi.noaa.gov/wp-content/uploads/Risk-Communication-and-Behavior-Best-Practices-and-Research-Findings-160425.pdf

[2] http://www.weather.gov/impacts/

[3] http://www.nhc.noaa.gov/surge/inundation/


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.