Weather community at a crossroads: 5 recommendations for improving predictions and their value

It’s a somewhat stormy time for the nation’s weather enterprise. The sequester promises painful cuts at the National Weather Service which has already prompted a hiring freeze and restricted travel.  The U.S. main weather forecasting model trails the Europeans in accuracy and computing power.  And a gap in weather satellite coverage may compromise weather forecasting in future years unless it’s addressed.

As weather enterprise leaders – from both the public and private sector – convene in Washington, D.C. next week for the American Meteorological Society’s Washington Forum (its theme is “The Value and Sustainability of the Weather, Water and Climate Enterprise”.), I propose 5 recommendations for the community to consider.

1) Commit to developing the world’s best weather model

The fact that the European Centre for Medium-range Forecasting (ECMWF) runs a superior computer model to the National Weather Service (NWS) is well-known.  In a guest post on this blog, University of Washington atmospheric science professor Cliff Mass described U.S. modeling as  “second rate” and argued we can and should do better:

As a numerical modeler myself I firmly believe that our global model could far exceed the performance of the rather conservative EC[MWF] effort.  I believe that … we could produce vastly improved forecasts, far superior to the EC[MWF], with huge economic and safety benefits to the nation.

NWS director Louis Uccellini told Climate Central that he views improving our modeling a “high priority” and said some computer upgrades are planned by the end of this summer.

But I’d suggest going further than this.

Uccellini and his colleagues should develop a detailed strategy for developing the world’s best weather forecasting model that includes an aggressive timetable, estimate of costs and an analysis of benefits. It should then present this strategy to both lawmakers and the media to demonstrate its commitment to the cause.

Related: To be the best in weather forecasting: Why Europe is beating the U.S.

2) Find cost effective solutions to looming satellite coverage gap

A possible gap in weather satellite coverage could begin as soon as 2014 and last 17 to 53 months. The satellites feed important data into models for weather forecasting and a gap could degrade forecasts.  The situation is serious enough that the Government Accountability Office listed the gap among the biggest threats facing the U.S. government.

The looming gap has arisen from mismanagement, massive cost overruns, and technical development challenges in NOAA’s satellite programs.

Conventional weather satellites are expensive.  USA Today’s Doyle Rice reported Monday that the latest polar-orbiting satellite launched cost $1.5 billion. But a new breed of smaller satellites, which can collect much of the same data, may present an attractive low-cost alternative.

Rice wrote:

“…a private company — Bethesda, Md.-based PlanetIQ — is proposing to bridge that [satellite] gap: PlanetIQ’s solution includes launching a constellation of 12 small satellites in low-Earth orbit to collect weather data.”

The cost to the government for these satellites would be around $70 million according to the company’s CEO, Anne Hale Miglarese.

But Cliff Mass penned a blog yesterday noting a similar fleet of satellites, known as COSMIC – which has demonstrated success – is already in orbit through a partnership between the government of Taiwan and the U.S.   While COSMIC is aging, Mass stressed that NOAA would do well to invest in a COSMIC-2 initiative.

“COSMIC-2 could greatly mitigate the upcoming weather satellite gap caused by [NOAA's] previous errors and mismanagement,” Mass concluded.

Irrespective of the merits of COSMIC, a proposed satellite fleet from PlanetIQ or some other solutions, NOAA should work with private sector and international partners to identify the most cost effective cure to the satellite gap crisis.

3) Invest in effective delivery of weather forecasts to inform decisions

A good forecast is only a good forecast if it leads to a good decision. In my view, the National Weather Service and the weather community more broadly have not put enough energy into making their forecast information as salient as possible.

For his part, NWS director Louis Uccellini seems to get it.  In an interview with the University of Wisconsin, he spoke extensively of the communication challenges facing forecasters:

How we communicate the forecast and the warnings is still a very critical problem. We have as a strategic goal for the entire country of making the United States a Weather Ready Nation so that everyone can take the proper action with respect to an impending storm and not wait too late, or ignore it and then get themselves in trouble in a life threatening situation. This means that the way we communicate a forecast, the watches and warnings, really need to be looked at. We still have situations in which the message we give is not equal to the message received. There’s a lot of work that needs to go on if we want to take this to the point where people will actually take the action you expect.

Yet, at the same time, NWS forecasts provide no information about the level of confidence. And its effectiveness in communicating key messages during storms is uneven from office to office.

The NWS, academic community and the private sector all need to place more emphasis on making forecast information as useful as possible for a range of audiences and should invest resources in social science and communication research to help accomplish this.

4) Find ways to cut costs, eliminate redundancy in National Weather Service

We’ve written a lot about NWS budget, travel (and training) and hiring cuts. Travel and training are critical for the professional development of NWS employees and important for scientific advances (to improve weather models, for example).  And the NWS forecast offices need to be fully staffed to issue timely, accurate forecasts and warnings.

But in this constrained budget environment, the reality is that NWS management must find new ways to be lean and mean and will have to prioritize, make unpopular decisions, and, likely, sacrifice.  This is particularly true if it wants to build a world-class computer model, have a robust network of observations to feed models from both Earth and space, and develop novel ways to convey forecast information effectively.

Within a large organization like the NWS, there are surely redundancies that can be eliminated.  And to be a bit provocative, should the leadership (if it’s not already) consider consolidation of offices as a solution and/or contracting out certain functions? That may be necessary.

5) Gain a better understanding of the economic value of weather services

The costs of extreme weather events can be astronomical. Economic damages in 2012 alone were estimated to exceed $60 billion dollars. Could better weather services reduce these costs? And – among all the possible improvements – which would have the greatest benefits and soonest?  The AMS Washington Forum has set up a session to discuss this, and teed it up with the following summary (excerpt):

…little is actually known about how to quantify the value of the efforts of the Weather and Climate Enterprise to reduce these impacts…it is difficult for policy and decision makers to assess whether additional investment by the public or private sectors would return benefits by reducing adverse impacts of severe weather events. Moreover, there is little guidance for the optimum distribution of such investments between improved observations, enhanced computer hardware and software capabilities, new research directions, university education of future professionals, or advanced training for present professionals in the management of weather and climate risk.

As a recommendation, perhaps the weather community should ask the National Academy of Sciences to conduct a study to gain a better understanding on the economic value of weather services and what improvements would bring the biggest bang for the buck, building off its 2012 report “Weather Services for the Nation: Becoming Second to None.”  While there is really no time to waste in addressing the challenges laid out in my first four recommendations, such a report could help inform the level of effort dedicated to them.

Also on Capital Weather Gang

D.C. area forecast: A windy mix of sun and clouds, Spring returns Saturday