Blog

NSF website features AER research on worst-case “exigent” weather forecasting

Daniel Gombos
January 8, 2014

Research performed by AER scientists Dr. Daniel Gombos and Dr. Ross Hoffman is featured on the NSF Science, Engineering, and Education (SEE) innovation website.

The NSF website reports on a new technique developed by AER scientists called exigent analysis (EA) that produces a forecast map of worst-case or “exigent” damages for a particular weather event based on an ensemble of forecasts.  For any chosen forecast confidence level, EA supplies an upper bound of the number of people and value of property that the storm would affect at each location.  The resulting exigent damage map gives the worst-case damage scenario for a given probability, or equivalently, the most likely pattern of damage for a given value of area-wide damage.

This research has direct application for emergency situations affecting life and property.  In advance of a disaster, emergency responders could use exigent damage maps as extreme but plausible scenarios, and agencies such as the Federal Emergency Management Agency and the Red Cross could use exigent forecasts to preposition resources and personnel.

Drs. Gombos and Hoffman applied EA to several recent major weather events including Superstorm Sandy (2012).  In the figure that appears at the SEE website and reproduced here, the left panel shows the ensemble mean forecast wind damage to households for Superstorm Sandy based on the ECMWF ensemble forecasts of 10-meter wind from 00 UTC 26 October 2012, approximately 96 hours before New Jersey landfall.  In the most likely scenario (left panel), total residential property damage is forecast to be $5.8B, with damage concentrated along the mid-Atlantic and New England coasts and Chesapeake and Delaware Bays.  The exigent damage at the 90% risk level (right panel) is greater than the most likely scenario at these locations and more inland penetration of damage in all areas north of the landfall location and especially near New York Harbor.  The total forecast damage for the exigent scenario is $16.5B.
 

AER Recognized by Northrop Grumman for Supplier Excellence

Brenda Kelly
December 19, 2013

AER has received a highly coveted Supplier Excellence Award from Northrop Grumman Corporation’s Information Systems (NGIS) sector for technical innovation and quality products and services.

AER was recognized for its work on the System Engineering Maintenance and Sustainment (SEMS) II program at the U.S. Air Force Weather Agency, the largest supplier of terrestrial and space weather data for military users worldwide.

Under the SEMS II program, AER provides subject matter expertise and development support to Northrop Grumman’s team, including those members assigned to the Cloud Depiction and Forecast System (CDFS) II. CDFS II, based on AER-developed science algorithms, is the world’s only operational, hourly, global cloud analysis and forecast system.

“This award from Northrop Grumman underscores AER’s ability to innovate and solve operational problems that affect the Air Force’s ability to provide warfighters with value-added environmental data for efficient decision making. The close collaboration between our companies not only translates into good science and engineering but also provides a substantial increase in Air Force Weather capabilities, with across-the-board benefits to defense and national programs,” said Ron Isaacs, president of AER.

 

 

AER SEMS II Team:Standing: Marc Hidalgo, Mark Conner, Eric Hunt.Seated: Becky Selin, Suseela Sarasamma.

 

Trent Dalton from Northrop Grumman (far right) presented the award to AER team members Suseela Sarasamma, Becky Selin, and Marc Hidalgo.

AER SEMS team members Bob d'Entremont, Gary Gustafson, and Hilary (Ned) Snell.

 

33 AER scientists contribute 28 research studies at AGU Fall Meeting 2013

Eli Mlawer
December 12, 2013

AER is proud that 10 of our scientists were the lead authors presenting their research on a wide range of topics at the American Geophysical Union (AGU) annual Fall meeting. In all, 33 AER scientists contributed to 28 research studies in 7 science categories ranging from greenhouse gases to land surface processes to space weather.
 

Science category AER studies contributed AER is lead author and presenter
AER science research presented at AGU
Greenhouse gases 10 2
Climate analysis 3 1
Numerical Weather Prediction 3 1
Land Surface Processes 5 3
Air Quality 4 1
Space Weather 2 2
Gravity and geodesy 1 0
Total 28 10

Click here to view the authors and their research topics in each science category.

AER scientists at the cutting edge of GHG research

Janusz Eluszkiewicz
November 26, 2013

AER scientists involved in greenhouse gas (GHG) research, Janusz Eluszkiewicz and Thomas Nehrkorn, are co-authors of a Harvard-led study "Anthropogenic emissions of methane in the United States" just published in the Proceedings of the National Academy of Sciences (PNAS) and receiving considerable attention worldwide for its conclusion that methane emissions in the U.S. are significantly greater than previous estimates.

AER's contribution to this research was the development of the atmospheric transport simulations that underlie the study. These simulations utilize the Stochastic Time Inverted Lagrangian Transport (STILT) model driven by customized meteorological fields from the Weather Research and Forecasting (WRF) mesoscale model. The coupled WRF-STILT model is the linchpin of multiple GHG-related efforts worldwide and has been developed at AER over the past decade with support from NSF, NASA, NOAA, the intelligence community, and private industry. Among on-going applications of WRF-STILT are urban methane leaks, monitoring, reporting, and verification (MRV) of CO2 emissions in the Boston-DC corridor, and GHG sources in the Arctic.

Besides Janusz and Thomas, scientists contributing to GHG research at AER include John Henderson, Marikate Ellis Mountain, Jennifer Hegarty and Scott Zaccheo.

Additional information about this study can be viewed in many news sources including:

15 Research Studies by AER Scientists Cited in IPCC WG1 AR5 Report on Climate Change

Ron Isaacs
October 27, 2013

When the U.N.’s IPCC released the WG-1 AR5 report, we were proud to see the breadth of research by AER scientists cited in the report. It’s an indication that AER scientists have helped advance the scientific discourse about the Earth’s climate by providing research covering a broad range of topics from the ocean depths to the top of the atmosphere.

15 peer-reviewed publications by AER scientists were cited in 7 of the 14 chapters in the WG-1 AR5 report and cover a broad range of topics from the ocean depths to the top of the atmosphere.

 

For example:

  • Climate modeling and simulations
    • Evaluations of climate model performance with respect to climate patterns such as El Niño and the Pacific Decadal Oscillation as well as water vapor in the upper tropospheric
    • Developing and validating aerosol and cloud models
  • Ocean
    • Developing state-of-the-art datasets of ocean surface winds
    • Measuring and modeling regional variability in sea level
    • Evaluating the accuracy of ocean mass trends
  • Atmosphere
    • Assessing sudden warming trends in the stratosphere
    • Retrieving information about size distributions of ice clouds from satellite measurements

Many of the global models that provided simulations of future climate for the WG-1 AR5 report use an AER-developed computer code to calculate accurately the radiative impact of greenhouse gases and other atmospheric constituents.

“For more than two decades, the atmospheric community has relied on AER scientists for expertise in the field of radiative transfer including the impact of increased greenhouse gases,” said Robert Morris, Ph.D., AER vice president of Research and Development.

The cited research studies were sponsored by a broad cross section of federal agencies, including the National Science Foundation, NASA, and the Office of Naval Research. The audience for AER's scientific environmental expertise includes academic and government communities who look to AER for authoritative analysis of environmental trends past, present and future.

In addition to research, contributions by AER staff included peer review of portions of the IPCC WG1 AR5 report.

GRL features AER research on Ocean Bottom Pressure

Christopher Piecuch
August 13, 2013

Knowledge of changes in ocean bottom pressure, and their relation to fluctuations in sea surface height, is important for understanding aspects of ocean circulation and climate variability; for example, facilitating estimation of ocean heat content variations and elucidating causes for regional sea level variability. However, historical measurements of bottom pressure have been extremely sparse, limiting observational understanding of the nature of ocean bottom pressure behavior on scales relevant for climate studies.

Taking advantage of nearly a decade of measurements of ocean mass variability from the Gravity Recovery and Climate Experiment (GRACE) — a collaboration between the space agencies of the United States and Germany — along with observations of sea level from satellite radar altimeter missions, AER authors Christopher Piecuch, Katherine Quinn, and Rui Ponte recently published the first observation-based investigation of the relationship between ocean bottom pressure and sea level over the global ocean on interannual time scales.

The authors found a tight relationship between sea level and bottom pressure variations over a number of oceanic regions; for example, abyssal plains in the Southern Ocean and shallow seas around the Indonesian archipelago. Their results demonstrate the high quality of mass measurements derived from GRACE, which is a “first-of-its-kind” mission, and also highlight important exceptions to the generally held paradigm that year-to-year changes in regional sea levels mostly represent redistribution of heat within the ocean.

The findings were published in Geophysical Research Letters and appear as a featured article on the journal's homepage.

See related research articles in the AER Research Library.

AER_OBP_for_GRL

Shading represents standard deviations of large-scale, low-frequency ocean bottom pressure signals from the Gravity Recovery and Climate Experiment (GRACE) satellite mission in units of equivalent cm of water.

Blackout Risk Forecasting: HSB and AER will develop tech tool to forecast blackouts

Robin Luo, Vice President of HSB
June 12, 2013

Electrical blackouts are on the rise in the US, costing businesses and consumers more than $100 billion annually and inconveniencing millions of people. But so far insurers, companies, municipalities, government agencies and other organizations lack a widely available and dependable tool to calculate the financial consequences.

Understand and assess blackout risks

In response, Hartford Steam Boiler and Atmospheric and Environmental Research (AER) have announced a strategic alliance to develop technology and services to comprehensively understand and assess the blackout risk. The technology will be used by insurers and reinsurers, commercial and industrial firms, and other businesses.

Plan for widespread electrical blackouts

“HSB and AER are commercializing the blackout risk data and technologies that our firms built, calibrated and applied within our businesses over the past decade,” said Greg Barats, HSB president and CEO. “Our joint teams will now make it possible for stakeholders to plan for widespread blackouts caused by a broad range of perils to help clients minimize and manage the impact across all lines of business.”

“Infrastructure reliability has become a critical subject for insurers, supply chains and governments at all levels,” said Ron Isaacs, AER president and CEO. “The strategic alliance brings unique expertise and experience and enables a significant advance in the blackout assessment field. New technology and tools are essential as weather events become more extreme and frequent, the power grid ages, and other risks threaten our power supply.”

Technical knowledge, environmental risk management

The technology will be built upon the 147-year track record of HSB’s technical knowledge and leadership in equipment breakdown insurance and loss prevention and AER’s expertise in climate and environmental risk management solutions. Read more about how the strategic alliance combines the technical resources of two industry leaders.

Important benefits for insurers and reinsurers

Through the alliance, HSB and AER are developing technology that will enable insurers to:

  • Predict the severity of business impacts from future power outage events
  • Determine portfolio risk accumulations and the probability of insured loss with confidence
  • Calculate the loss potential to current and future coverages which will inform risk-based pricing
  • Assess how blackout risk varies by geographic and grid area

Important benefits for businesses and broker clients

The technology developed by HSB and AER will enable businesses to:

  • Assess the likelihood of a power outage
  • Understand backup power resource needs to sustain continuity
  • Consider where to locate backup facilities
  • Assess coverage needs related to power outages

Assess and optimize blackout risk exposure

The technology will empower risk managers in insurance and reinsurance, corporations and government agencies to take a strategic perspective on blackout risk, assess and optimize their risk exposure from potential events and calculate losses when an event is imminent.

HSB and AER are leaders in risk analysis services

Both HSB and AER have a long history in risk analysis services. HSB, part of Munich Re, is a leading engineering and technical risk insurer in North America and the largest insurer of electrical equipment. AER, a unit of Verisk Analytics, is a leading provider of climate and environmental risk management solutions.

Learn More

  • Download the brochure “Power outage risks are on the rise: are you prepared to manage the risks of blackouts to your portfolio?”
  • View more about the AER and HSB strategic alliance to develop technology and services to calculate the consequences of power outages [This links to the HSB-and-AER landing page which will be updated in summer and autumn with more links as we introduce more content.]

Read the press release AER and HSB to Develop Tech Tool to Forecast Blackouts.

Solar Storm Risk to the North Am. Electric Grid

Nicole Homeier and Lisa Wei
May 22, 2013

Lloyds and AER published a ground-breaking study of solar storm risk to the electric grid in North America.

The Space Weather Study by Lloyds and AER analyzes the likelihood of extreme geomagnetic storms, the specific vulnerabilities of the North American power grid, the regions at highest risk and the implications for the insurance industry and society generally.

“These innovations to address the risk posed by geomagnetic disturbances comprise an interdisciplinary study tying together historical records, space physics, geophysics, engineering, and economics.” – Lisa Wei, PhD, AER

 

 

AER scientists who co-authored the report and provided research and analysis:

  • Nicole Homeier, PhD, Director of Product Innovation and Senior Staff Scientist at AER
  • Lisa Wei, PhD, Staff Scientist at AER

Proactively Monitoring Wildfires in the West

Dr. Arindam Samanta, FireLine Operations Manager/Scientist
May 6, 2013

The 2013 wildfire season in the western United States is looking to be a difficult one, fueled by a combination of factors, including a drier-than-normal preceding winter, generally drier vegetative fuels, and warmer current conditions. In addition, projections of higher temperatures and lower-than-normal precipitation this summer raise the prospects of what could become an intense wildfire season.

A number of fierce wildfires are raging, threatening hundreds of thousands of properties, in both Northern and Southern California. They include the Panther Fire (Tehama County), the Summit Fire (Riverside County), and the Springs Fire (Ventura County). Those fires have already damaged homes and are threatening hundreds of thousands of additional properties. Since it started last Thursday, the Springs Fire has spread along the coast north of Los Angeles, carried by winds in excess of 43 miles per hour. The fire has also partially shut down the Pacific Coast Highway.

To date, the fire has reportedly damaged 15 homes, destroyed 26 outbuildings, and laid waste to more than 27,500 acres of brush.

We at Verisk and AER are monitoring these wildfire developments through cutting-edge satellite imaging technology, some of which we also use in FireLine™, our property insurance wildfire risk assessment solution.

AER provides near-real-time data on current wildfires through AER Respond™ for Wildfire Service. The service applies proprietary algorithms to the most up-to-date satellite imagery — both moderate (>100 meters) and high (<10 meters) spatial resolution — to provide the following analytics:

  • accumulated burn area — reanalyzed when new imagery becomes available

  • new burn area — assessed relative to burn area from prior analysis

  • hot areas — (if any) provided at burn area analysis times

AER has scheduled the analytics for daily releases plus intermediate releases when new data becomes available in a variety of geospatial formats (for example, shapefiles) that seamlessly integrate into existing GIS systems.

AER Respond for Wildfire Service can help with your insurance operations in a number of ways:

  • timely assessment of potential claims activity

  • support for underwriting moratoriums

  • timely allocation of resources for claims adjusting

  • post event analysis for claims adjusting

  • overall exposure management

AER Respond for Wildfire Service is tracking several California wildifires, including the Panther Fire (Tehama County), the Summit Fire (Riverside County), and the Springs Fire (Ventura County). The insets above show polygons of burn area, changing over time. Also shown are hot areas/hotspots (red dots) in the Springs Fire area.

AER’s “amazingly accurate” winter outlook

Dr. Judah Cohen
April 29, 2013

The AER winter forecast has now been correct for four years in a row (Read Press Release). Even before that we had published a paper documenting that the AER model was the most skillful, e.g., accurate, winter seasonal forecast model that can be verified. Yet before every new forecast, winter or summer, there is great pressure to be correct and the need to validate the accuracy and usefulness of our techniques. And to make things more difficult this past winter, December 2012 got off to a blazing start. The consensus quickly became this winter would simply be a repeat of last year’s record mild winter. But our predictors continued to signal bullish for cold from January onward and we stayed the course.

It is very satisfying knowing now that the winter is over, that the warm start in December was an aberration and the remainder of the winter was indeed cold and in many cities very snowy, including in our hometown of Boston (though pity the people of Duluth that had their snowiest month ever this April!). Perseverance has paid off and it is nice to be recognized for another correct forecast.

Throughout my career I have heard that we are not good just lucky. But getting the forecast correct even once is extremely difficult and we have produced, the best forecast four years in a row, including this winter; a result that is highly improbable simply due to chance. The correct forecast is not just for the US but also for the entire Northern Hemisphere.

AER Forecast Proves Exceptionally Accurate for 2013 U.S. Heating Demand
Forecast from autumn 2012 outperformed all public forecast benchmarks.. Achieved average error of 1°F (0.52°C) for the U.S., and 1.5°F (0.87°C) for the Northern Hemisphere.

Forecast Skill

One metric of the forecast skill or accuracy is the pattern correlation between the predicted and observed temperatures. A perfect forecast would produce a value of 1 and any value above 0 is considered skillful. The pattern correlation for both December through February (Dec., Jan., Feb, or DJF) and January through March were comparable at 0.6 and 0.65 respectively. The skill for Eurasia was also comparable, for DJF it was 0.67 and for JFM 0.64. But the best overall score was for North America for the January-March period with a pattern correlation of 0.79.

Another metric of skill is the root mean square error (RMSE), which tallies the average error at every point. For the US the area average RMSE was less than 1ºF.

"Job Well Done!"

Both of these metrics are not only great scores for a three-month forecast; they are great scores for a three day forecast!

The accurate forecast has not gone unnoticed and wishes of congratulations and job well done have come from clients, fellow scientists and even the media including Harvey Leonard the chief meteorologist at WCVB. But the most effusive praise so far has been from the Washington Post’s Capital Weather Gang. Whose blog title says it all: “AER’s Judah Cohen produces amazingly accurate winter outlook.” It seems that jaw dropping was the most common reaction to the verification of the forecast, which was Jason Samenow’s reaction as written in the article and from a client.

Here at AER we believe that we produce the best seasonal forecast even with less manpower and resources at our disposal than the government forecast centers. A big reason is because we made the decision from the beginning not to follow the herd and obsess with ENSO as the only seasonal forecast predictor. Instead our focus has been on the Arctic Oscillation and of course Siberian snow cover; focusing on these predictors has paid off for our clients. And as Jason concludes in his blog: “Perhaps it’s time … government centers begin to incorporate Cohen’s methodology into the preparation of their outlooks.”

It is very gratifying for others to acknowledge our leadership role in seasonal forecasting. But there is no time to rest on our laurels and Jason Furtado, Justin Jones and I are already sweating the summer forecast.

Download the Seasonal Forecast brochure

Read the Press Release

Read more about the Winter Seasonal Forecast