This week’s House S&T Committee EPIC hearing.

Wednesday afternoon I had occasion to participate in an iconic 21st-century American pastime – decamping from the office in order to wait at home for the cable guy, who then never shows up until well after the agreed-upon time window (essentially valuing the residents’ time at zero). The silver lining? An opportunity to watch the live-streamed  U. S. House of Representatives Committee on Science, Space, and Technology hearing on EPIC.

Folks living outside the Beltway might be forgiven for thinking that nothing is going on in Congress outside the much-ballyhooed impeachment hearings. But the reality is quite different. For the most part, the nation’s business continues. There’s a focus on meaningful priorities, and there’s comity and bipartisanship, all resulting in worthwhile discussion. That was the case in Wednesday’s hearing.

What made the hearing EPIC was not this conjunction of rational elements, but rather the subject matter: the notional Earth Prediction Innovation Center. (Taking a few liberties), here’s a thumbnail version of the background: In 2017 Congress passed the (remarkably nonpartisan) Weather Research and Forecasting Innovation Act, which among other things called for a national weather forecasting capability that will surpass that of the Europeans. The hearing was focused on what’s been done to date, future plans, and the resources that will be needed. The committee heard from a distinguished panel:

  • Dr. Neil Jacobs, Assistant Secretary of Commerce for Environmental Observation and Prediction, performing the duties of Under Secretary of Commerce for Oceans and Atmosphere, NOAA
  • Dr. Clifford Mass, Professor of Atmospheric Sciences, University of Washington 
  • Dr. Peter P. Neilley, IBM Distinguished Engineer and Director of Weather Forecasting Sciences and Technologies, The Weather Company, An IBM Business 
  • Dr. Thomas Auligné, Director of the Joint Center for Satellite Data Assimilation, University Corporation for Atmospheric Research (UCAR)

The full discussion is worth your viewing, but here are a few takeaways/reflections:

Goal. To people of a certain age, this goal of surpassing the European weather forecast capability calls to mind the 1960’s challenge of reaching the moon by the end of that decade. At the time, President Kennedy referred to this as a race that the United States would win, but he also referred to a loftier goal of building on innovation more generally to “explore the stars.” EPIC’s framing has some elements of a race as well. Some comments at this week’s hearing seemed to suggest that U.S. research expertise in weather prediction is unsurpassed, so that achieving superior weather forecasts for societal benefit was largely a matter of (1) NOAA’s opening its doors and providing greater academic-research-community and private-enterprise access to operational models and model development (unifying not just the codes but also the supporting community computing infrastructure), and (2) substantially increasing the computing power available to government, academia, and the private-sector.

These steps can indeed accelerate U.S. progress. But there’s probably little joy in seeing this as a race, for several reasons. First, unlike going to the moon, where there’s little ambiguity about “who gets there first,” superior European weather-forecasting skill is something that can only be discerned from statistical analysis. Second, the European capability is not fixed, but a moving target. Equaling their current level of skill will be easier than catching up to where they’ll by that time. Third, and most fundamentally, European Center and U.S. centers have historically cooperated, and the Europeans are not an enemy. By contrast, during the Cold War, the connection of the space race to nuclear missile capability was obvious, and the Soviet threat to U.S. and indeed world interests was tangible and real.

It’s therefore important to keep in mind, and perhaps even elevate, the broader idea that America hosts the world’s widest variety of weather challenges: great cycles of drought and flood; winter storms matching those of other high latitude countries; hurricanes matching those of lower-latitude nations; severe warm-season convection; and a virtual lock on the world’s tornadoes. Better weather forecasts are therefore fundamental to every American aspiration: renewable energy, food production, and water management; life and safety in the face of extremes; and protection of the environment and ecosystems. The goal is a better life for every American, and indeed every person, plant, and animal on Earth. (The idea is not to choose between either of two quite different framings, but rather balance them and realize the benefits of both).

Time frame. The prepared statements, the questions, and the answers to those questions could have left viewers with unfounded expectation of a “quick win.” The idea that American university research expertise is a vast underutilized resource only needing to be tapped is a bit simplistic. Incentives for advancement within the academic research community are not exactly aligned with the personal investments needed to contribute to the practitioners’ world. The cultures are wildly different – and for good reasons. The difficulties of “unification” of codes, hardware, etc., were largely unaddressed or glossed over. Big data, artificial intelligence, and other nations are competing for the same talent pools. The governance issues are themselves non-trivial, as is the need for…

Resources. There was universal agreement across the panelists and committee members that more resources will be needed. Discussion focused on money and computing power. The current budget discussion in Congress shows perhaps $7.5-15M available the first year, depending on the Senate or the House mark. All parties agreed that either mark would be wholly inadequate, but there were no suggestions about how the needed higher levels of funding might be achieved.

Governance. The funding discussion played into considerations of governance. Everyone agreed that the Center should be hosted/housed outside of NOAA (but with NOAA (and other federal agencies strong participants). This point was emphasized beginning with the opening statements and reaffirmed throughout, as if to assuage fears that NOAA would somehow capture the entity, and the country would wind up no better off than before. The NOAA-veteran-of-32-years part of me was a bit put off by this tone, maybe even defensive, but it is right decision. One exchange suggested that this extramural location would ensure sustainability following the end of the current administration. However, follow through over the long term will stand or fall based more fundamentally on whether at its core EPIC is a good idea.

Other comments under this heading included whether EPIC should be a completely new entity, whether it should eventually morph from a virtual organization (demanded by the small size of the budget increment) or something more concrete featuring bricks and mortar. A cautionary note: Formation of a new entity risks creating yet another marginally-supported institution likely to be drawn into turf wars with existing (also marginally funded) players. Even a proposal-driven process that results in an existing institution taking on this additional role will be similarly vulnerable. This suggests that all stakeholders, including the Congress, devote equal attention to community-building across the so-called Weather, Water, and Climate Enterprise.

Near the hearing’s close, two questions were raised that ought to be of special interest to all of us.

Return on investment. The first was: What will be the return on investment here? It’s likely that none of the panelists is satisfied with the answers they were able to mount. The interlocutor seemed satisfied after hearing that several trillion dollars a year of U.S. economic activity is weather-dependent. This answer is a bit glib. The foot-shuffling highlights the importance of ongoing efforts across the Enterprise to develop better analysis and characterization of the value of environmental intelligence.

What will change? The second was: Let’s go out five years. EPIC has been a success. What will be/feel different to the average American? The answers were variations on a theme that Americans can have more confidence in their forecasts.

True enough. But’s that not going to change the experience for those in harm’s way. That’s because forecast improvements are barely keeping pace with increasing demands on forecast quality imposed by population increase, urbanization, the rise of social media, and other social change and technological advance. So long as emergency response requires mobilizing larger numbers of Americans ever-sooner in the face of approaching threats, under major uncertainty about their options and associated risks, American lives will continue to disrupted, even forever changed by weather, water, and climate extremes. Americans can feel safer and enjoy more control  in the face of natural hazards only when land use, building codes, and more robust critical infrastructure systemically reduce hazard risk; when public education and public policies restore equity and agency.

EPIC alone, as currently viewed, will not by itself change that. Still a lot of work ahead.

This entry was posted in Uncategorized. Bookmark the permalink.

4 Responses to This week’s House S&T Committee EPIC hearing.

  1. Jimmy Correia says:

    Regarding the final question. Turn it around. Its 5 years later, why did EPIC fail?
    Funding. turf wars. death by committee. All rational and reasonable given we have those now.

    I would offer that EPIC failed because so many offered solutions that no one entity could decide what to prioritize (just look at the number of physics options we have in WRF, an already community built modeling system).

    So we have one UFS with 50 different equally viable “improvements”. As it is we are already on course to have 5 applications of one UFS. 5*50 is 250 versions of a model. And we havent even mentioned Data Assimilation.

  2. William Hooke says:

    Thanks, Jimmy, for the clever flip of the question and for a thoughtful perspective. More than welcome.

  3. Neil Stuart says:

    Bill,

    Your post is timely for me as I attended the inaugural RISE conference here in Albany NY, (Resilience In Sustainable rEconstruction (RISE)) which was hosted at UAlbany on 18-20 November. It is a global initiative that the president of UAlbany helped organize. I was fortunate to attend and be a panel expert for one panel discussion but if you access the RISE web site, you will see the theme was “Transforming University Engagement In Pre- and Post-Disaster Environments: Lessons from Puerto Rico”.

    https://www.albany.edu/rise2019/home.html

    Without getting into too much detail, there were many though-provoking, heart-felt and creative discussions on how all components of the pre-disaster, during disaster and recovery process can best work together to minimize the costs, impacts and human suffering associated with the increased frequency of disasters, using the experience of Hurricane Maria (2017) in Puerto Rico as a case study. There was considerable time devoted to discussions on what niche RISE could fill since there are already so many entities and organizations that work together to research and develop solutions to all the problems associated with disasters.

    So, as inspiring as the RISE conference and movement is, there is the same risk that EPIC faces, that RISE may just be the formation of a new entity that risks creating yet another marginally-supported institution likely to be drawn into turf wars with existing (also marginally funded) players. I encourage you to glance at the RISE web site and look at the agenda and the huge spectrum of participants to see all the important topics that were covered in the 3 day conference. As fractured as our weather, water and climate enterprise and efforts may be at times, we still have good reason to be optimistic that we will continue to make progress. Thanks for your time.

    Neil

Leave a Reply

Your email address will not be published.