This will be the last thing that I write for Forbes in 2019 and this decade. Atmospheric sciences, the collective of meteorology and climate, have come along way over the past decade with new weather radars, modeling, satellites, and observation capabilities. Irrespective of misguided myths and perceptions, weather forecasts are the best they have ever been, and we have a solid grasp on how climate is changing. As we move into the next decade, I wanted to pull out a “crystal ball” and see what’s next on the horizon for the weather and climate community. I reached out to a cohort of 20 weather and climate experts for their projections.
Before surveying the projections of my expert colleagues, I thought that it would be useful to offer my own thoughts first:
Messaging and its consumption will focus on impacts rather than category or rating levels. Communication, psychology, and sociology will be fully immersed in warning processes. Phased-array weather radar systems will also gain traction. Satellite systems will continue to evolve beyond “seeing” weather with new data available for model assimilation. On the forecasting front, we will see the fruit of the Earth Prediction Innovation Center (EPIC) and its community collaboration approach. Attribution of extreme weather events to climate change will become more reliable. I envision a percentage contribution (%) or likelihood metric attached to future forecasts or post-analyses. I also worry about continued trends in Arctic Sea Ice loss, sea level rise, and Greenland dynamics that, in some cases, are ahead of projections from decades ago.
The experts that I queried span the public, private and academic sectors of the weather-climate enterprise. Key themes that emerged from their projections center around: Artificial Intelligence (AI) and Machine Learning, Advances in Prediction, Communication, Societal Risk, Health, and Evolving Technology.
Artificial Intelligence, Machine Learning and Forecasting
Victor Gensini, an atmospheric sciences professor at Northern Illinois University, shared his vision of the next decade, and it he has some interesting perspective on weather modeling and machine learning:
I see the next decade featuring a significant application of machine learning methods to numerical weather prediction output. Systems like MOS will likely be replaced by such algorithms making for even better local point forecasting. Think of it as using statistical downscaling on output with horizontal grid spacings likely 1km (by the end of the decade) for hyperlocal forecasts. I do wonder if we have the proper calibrated observations to train such models, but perhaps we will see improvements in this area over the next ten years too. We may also see the first “operational” convection permitting global climate model by the end of the decade. That would be a significant leap for the climate research community.
University of Illinois professor Steve Nesbitt echoes some of Gensini’s thoughts but extends them to further elaborate on the emerging role of AI:
My thought is “artificial intelligence will revolutionize how we observe, simulate, and forecast weather and climate. It will improve the quality of observations for forecasters and models, improving how information and data is incorporated into model forecasts and decision support systems. It will also improve the mechanics of how computer models are run, reducing complexity and computational cost, enabling better accuracy and spatial resolution”
Kevin Petty is head of science and forecasting at The Weather Company, an IBM business. He is also bullish on AI and machine learning. Like Gensini and Nesbitt, he believes we have only scratched the surface. He adds:
We must stop treating weather equally across the globe. It’s not. Some populations/businesses are more vulnerable than others. The democratization of data and analytics is something else that will give rise to more participation from individuals and groups not directly immersed in fundamental aspects of weather, water, and climate, but have the knowhow and interest to take advantage of weather-related data. In other words, data and analytics will become more accessible to the broader population, leading to new, innovative products and solutions. Image processing.
Other experts that I queried like Dr. Jason Furtado at the University of Oklahoma (OU), CIRES postdoctoral Sam Lillo, and Brian Etherton at Maxar Technologies strongly affirmed the role of AI in weather – climate science in the next decade.
Advances in Prediction
Texas TV meteorologist Richard “Heatwave” Berler evokes images of something common to all of us, the smartphone camera:
Like a camera with more megapixels, weather observations/forecast models will be taken/run at a higher resolution, be it from more satellite platforms, even from crowd sourcing citizen weather stations and vehicles at the surface. This, run on ever faster computers will lead to greater resolution of small detail in the first 12 hours of a forecast, and give us a longer horizon on the large scale weather features. I hope to see some skill (now seen 7 days out) extend to 10-12 days out by decade’s end.
Subseasonal-to-seasonal forecasting is increasingly critical for planning and operations related to industry, public safety, and agriculture. Barb Mayes Bousted is a meteorological trainer at the National Weather Service and the developer of the Winter Misery Index:
One thing that jumps to mind would be leaps in the subseasonal-to-seasonal forecasting of high-impact weather events, crossing those blurry lines between weather and climate. I could foresee advances in longer-range predictions of severe thunderstorms and tornadoes, hurricanes, floods, heat waves, cold snaps, untimely freezes, flash droughts – the range of extremes that can impact life, property, and ecosystems. We are on the path already with at least a few of those hazards, but we need – and, I believe, will achieve – greater understanding and communication of the links between global climate patterns and potential for hazardous/high-impact weather.
Ashton Robinson Cook believes important advances in tornado prediction are on the near horizon. Referencing work by Gensini and others, he told me:
As far as tornado prediction goes I think the advancement of the high-resolution ensembles are the big advancement for severe local storms now but maybe only for the 2-4 year timeframe. For predicting tornadoes in the long range, coarser grid climate prediction models also seem to have demonstrated potential but those models will also need improving as well. AI will be the biggest player in improving forecast capabilities.
Communication and Societal Risk
Legendary TV meteorologist Gene Norman is now a meteorological consultant and freelance broadcaster with CNN. He had an array of thoughts on the next decade:
Weather advances in the next decade: 1) more accurate and targeted weather alerts that incent safety action to cut down on the ” we didn’t know it would be this bad” reaction after a major weather event. 2) improved accuracy of forecasts due to greater computing abilities 3) better and more consistent communication (by the media) of the linkages between extreme weather events and climate change to eliminate skepticism and erroneous reporting. This likely also prompts intentional political action
Susan Jasko was a candidate for President of the National Weather Association (NWA) in 2019 and a leading voice at the intersection of weather and communication. Her thoughts:
We seem to be at the edge of a transition from an idea about humans dominating nature and controlling the natural world to the notion that human life is dependent on a holistic understanding of the natural world. I think the weather and climate enterprise will make breakthroughs in ways to engage communities in using science to inform crucial decision-making and transform us into an environmentally resilient and wise nation. Of course, it will take the expertise of all of us – physical and social scientists, policy and resource experts, economists and artists, teachers and families- to make this transition successful.
Kim Klockow has a similar view. She is a research scientist and Societal Applications Coordinator with the Cooperative Institute for Mesoscale Meteorological Studies (CIMMS) at the University of Oklahoma and the National Severe Storms Laboratory (NSSL):
I think the next big advance will be more locally-focused weather communication & resilience activities. That’s the next big frontier for the social science side of our enterprise – what works in one community may be different from what works in another, and reaching people where they are will provide the best service. I see this as an integral part of the AMS Centennial Initiative focused on local action networks. All disasters are local, and so are the solutions to many of our most pressing weather and climate change-related problems!
Miri Marshall, a meteorologist at WUSA 9 in Washington D.C., is one of the most talented weather communicators that I have seen in recent years. She has a particular gift for making weather information accessible and relatable. Her thoughts come from participating in the cohort of the Yale Climate Change and Health Certificate Program:
I think we’ll see a bigger emphasis on climate change and health impacts. Climate change is not just about chunks of melting ice, but about the potential for an increase in cases of heat stress, vibrio infections, and mosquito borne illnesses just to name a few. Climate and health are a package deal. It’s all connected.
Kevin Kloesel is also thinking about health but from a weather perspective. He thinks about everything from lightning safety to campus preparedness during severe weather at the University of Oklahoma. He has some forward-thinking ideas about the deadliest weather per annum in the U.S., heat.
With heat waves forecasted to become longer and more intense, heat illness will likely continue to be the leading cause of death from weather. Therefore, I can envision a day when biometric information from our smart watches are integrated with hyperlocal environmental information and personal health inputs such as nutrition, hydration, prescription medication consumption, metabolic rate, etc. to create a heat stress index that is customized for every person. Personalized early-warning heat advisories could then be issued via a person’s smart device whether sedentary or in the midst of an athletic endeavor.
Gavin Schmidt is a climate scientist at NASA Goddard Institute for Space Studies. He is a leading voice in sounding the alarm with science (not opinions and innuendo) on climate change. He recently tweeted that 2019 will likely be the second warmest year on record. He made the following projection about forthcoming space missions:
When Plankton, Aerosol, Cloud, ocean Ecosystems (PACE) satellite mission launches we have the possibility of global scale aerosol composition, which will greatly constrain the total aerosol forcing. The James Webb Space Telescope might give a glimpse into exoplanet climates that might solidify (or demolish) the idea that we understand climate on a broader scale.
Recent episodes of the Weather Geeks podcast revealed several technologies to keep an eye on in the next decade. Phil Chillson, a professor at the University of Oklahoma, discussed unmanned aerial vehicles and drones as new sources of lower atmosphere data for weather networks and models. The Weather Channel’s Mike Chesterfield discussed the evolution of Immersive Mixed Reality (IMR) to better convey weather messaging and warning. Mark Powell introduced RMS HWind technology for unparalleled visualization of a hurricane wind field. Catherine Edwards, a researcher at the University of Georgia, has been using autonomous drones to sample water temperatures and other parameters in hurricane environments. Companies like ClimaCell are utilizing unconventional sources of information like cameras, cellular networks, and more of information to deliver pinpoint weather information.
For the last word, Deanna Hence, an assistant professor at the University of Illinois, sums things up nicely:
I believe the next great advances will be increasing sophistication in the merging of weather and climate science with the understanding of human systems. In all of the ways we are increasingly incorporating the connection between human decision making and weather and climate, breakthroughs in this area will improve not only any effort connected to weather and climate prediction, but any other science, engineering, technology, or industry that the atmosphere touches.
There are far more advances not covered herein. Climate models will continue to evolve, and climate change will finally take a meaningful place in the 2020 (and beyond) election cycles.. NASA will continue to develop new missions in response to community guidance in the Decadal Survey that advance Earth System science. NOAA will forge into its Weather Ready Nation initiative. I also expect Smallsat and Cubesat systems to further find a footing in the weather-climate portfolio as well.
It is an exciting time to be a part of a rapidly maturing scientific field.