AMS Centennial | Earth Networks Recap

  • Jan 10, 2020

Happy Birthday, AMS!

AMS 100th birthday cupcakes

Early in January 2020, we traveled to Boston for the American Meteorological Society Annual Conference to celebrate 100 years!

Earth Networks at AMS 100

The Earth Networks Team gathered at AMS 100 to present posters, sit on panels, and network with up-and-coming meteorologists at WeatherFest and the Career Fair.

Interested in what we we up to? You can read our recap below.

Earth Networks Sessions

The Outcomes of the 2019 WMO Congress: What is the Path Forward for International Cooperation and Coordination across the Weather Enterprise?

Earth Networks SVP, Global Sales, Jim Anderson, will sat on the panel for this session. The panel included other leaders from the Public and Private Sections who discussed challenges, benefits, and next steps across WMO’s regions.

Policy & Execution in Support of the Weather Research and Forecast Innovation Act

Earth Networks VP of Federal Programs, Bill Callahan, led this session exploring the amount and usability of weather information in public-private partnerships. This session also focused on how industry and academia should work together within the weather enterprise.

A great discussion followed, diving deeper into ideas, plans, and ongoing activities that will lead to deeper and broader collaboration across the American and International Weather and Climate Enterprises.

Earth Networks Posters

Using Total Lightning Data to Optimize Airport Shutdown Costs (Poster #42)

A picture of an airplane at a terminal at an airport with a pretty multicolored dusk sky

Airport closures due to lightning threats total over $30 billion each year. These costs are mostly burdened on the major air carriers. Determining the optimal threat parameters to trigger closures could save the airlines billions of dollars each year.

Using the Earth Networks Total Lightning network data for 2014 – 2019, statistical analysis was done on lightning occurrences at 10 major U.S. airports with varying geography.

Theoretical closures were tabulated for varying trigger radii and lightning free “all-clear” windows. These data were then used to calculate the frequency of closures, the amount of closure time and probabilities for subsequent “hits” in the airport “strike zone.”

The data shows there are optimal radii and all-clear intervals that will limit lost time due to evacuations and re-deployments and subsequent re-evacuations due to short all-clear windows.

Analyzing Thunderstorms for Improved Lightning Safety (Poster #155)

A lit up city view taken from a high elevation with a thunderstorm overhead, complete with bright lightning strikes

Lightning poses a significant safety hazard for all sectors of the nation (public, private, and government). The ‘best practices’ for lightning safety used in those sectors vary significantly but are all based on passed statistics of lightning in thunderstorms.

As the climate changes and thunderstorms generally become more powerful, these statistics need to be re-evaluated so as to provide the most accurate data.

In this study, we develop a lightning clustering algorithm that takes individual lightning strokes and creates thunderstorms based on their spatiotemporal proximity. We use lightning data from the Earth Networks Total Lightning Network. Once these thunderstorms are obtained, we can test the efficacy of various safety protocols and practices such as the 30-30 rule.

Preliminary results consisted of 482 storms from a single day and found that 16 of those storms (4%) produced a final flash more than 30 minutes after the previous. Furthermore, 8 of those storms (1.7%) produced a cloud-to-ground that is greater than 25 km from the nearest stroke in the storm with the largest outlier being 43 km, which was probably limited by our clustering algorithm.

These results will help in deciding whether the current safety protocols need revision to provide the greatest safety for the public.

Thunder-Day Climatology Using Modern Lightning Location Data (Poster #156)

A green field with dark grey storm clouds overhead and three cloud-to-ground lightning strikes

A thunder-day is defined as a 24-hour period during which thunder is heard. Records of thunder-days for specific locations go back more than 100 years, making this one of the few measurements suitable for long-term lightning climatology studies. More recently, lightning climatology studies have used lightning flash density instead of thunder-days (e.g. Cecil et al. 2014, Hodanish et al. 1997).

Since lightning location systems have been operationally available for only a handful of decades, the history available with the measurements is much shorter and prone to biases due to ever improving detection technology. Further, for inter-seasonal analysis measurements of flash density are arguably worse than thunder days because the number of flashes observed in any particular location in a season is frequently dominated by a small number of thunderstorms. This requires flash density observations to be averaged over longer periods of time to smooth out the variability.

Luckily, it is possible to generate a synthetic thunder-day measurement using the stroke locations of a lightning location system. As an added benefit, these synthetic measurements heavily reduce the effects of temporally and spatially varying detection efficiency inherent in ground based lightning location systems.

Thunder-day observations are also more intuitive and meaningful to a general audience, and still make stunning visualizations. In this study, we will develop a synthetic thunder-day product, apply the technique to stroke data from multiple lightning location systems, and compare the results to direct thunder observations available in NOAA’s Global Historical Climatology Network, as well as other thunder-day sources.

We will then show maps of multiple years of thunder activity for a number of locations on the globe, and demonstrate the utility of these maps for inter-seasonal trend analysis.

AI-Powered Chatbot for Effective Weather Communication (Poster #1357)

Weather forecasting is a complex and often challenging skill that involves observing and processing vast amounts of data. Even though there are Weather systems built to generate automated Weather warnings, they still suffer from many User communication problems. Effective communication of weather information is important both for experts who disseminate crucial, life-saving information about threats and for the public, who need to receive and interpret in ways that allow them to act.

Among the communication problems are the difficulties to answer weather domain related queries to the public, delivering abstruse weather forecast, report delayed weather news, unavailability of 24×7 customer service during severe weather situations, and not effectively communicating the operational and technical information of the weather system used by the public.

To resolve these problems, we aim to design a chatbot that is specifically tailored for handling Weather queries from the public. The Chatbot that we proposed, is designed to 1. Provide effective Weather warnings 2. Educate and provide information on the current Weather, 3. Serve as a 24×7 customer care 4. Provide personalized weather forecasting to help User better plan their day.

The methodology that we followed to build the Chatbot makes use of Natural Language Processing (NLP) to extract useful information from the public queries and provide a relevant response using Machine Learning-based approach. The Machine learning-based approach allows the chatbot to use a repository of predefined responses and heuristics to pick an appropriate response to the public. The heuristic could vary as simple as a rule-based expression match, or as complex as a Machine Learning classification.

For building the Chatbot’s Weather domain knowledge, we would be using techniques for extracting Structured knowledge from Email messages of a Weather organization. The knowledge extraction will be an on-going process and the extracted weather knowledge will be useful for the public for better understanding the current weather.

The test results of the built chatbot show that it serves to be extremely useful for tool for communicating weather information to the public. Moreover, the Chatbot developed can be used as a guideline for future improvement and study.

What is the AMS?

The American Meteorological Society is the nation’s premier scientific and professional organization. It promotes and disseminates information about the atmospheric, oceanic, and hydrological sciences.

AMS currently has over 13,000 members including researchers, educators, students, enthusiasts, broadcasters, and other professionals in weather, water, and climate.

You can learn more about AMS, their history, and their future by watching the short 100 Year Anniversary video below. We hope to see you in Boston!