By giving us your feedback, you can help improve your www.NOAA.gov experience. This short, anonymous survey only takes just a few minutes to complete 11 questions. Thank you for your input!Give my feedback
December 7, 2011
Thank you, Mike [McPhaden, AGU President].
It’s a particular pleasure for me to be here among colleagues, partners, and friends.
NOAA’S SCIENTIFIC INTEGRITY POLICY
Before I begin my formal remarks, I want to thank you and the other AGU leaders for your roles in enabling AGU to be such a vital force within the scientific community.
Science underpins all that AGU and NOAA do, and we appreciate the key role you play in championing strong science and its use.
Scientific integrity is at the core of producing and using good science, so I’d like to begin with a brief update on that front.
When I first arrived at NOAA, I made a commitment to protect scientific findings from being suppressed, distorted or altered; to strengthen science; and to encourage a culture of transparency. From my first days on the job, NOAA has been developing its first-ever policy on scientific integrity.
This policy is about fostering an environment where science is encouraged, nurtured, respected, rewarded and protected.
In that spirit, I am pleased to say that today, after my talk, I will be announcing the release of NOAA’s scientific integrity policy.
Developed through a deliberate and inclusive process, this policy amplifies NOAA’s and the Obama Administration’s commitment to scientific integrity.
THE EXTERNAL LANDSCAPE
This is a challenging time for the nation and for science.
The irony is that the demand for services provided by agencies like NOAA is at an all-time high and growing.
One reason for this demand is the increase in number and intensity of extreme events –
NOAA has been busy predicting the weather-related extreme events we’ve seen this year.
However, our capacity to continuing to do so is seriously threatened by downward pressure on our budgets. Budgets and politics threaten observations, research, modeling and delivery of information and other services.
Our ability to do and fund research to better understand the causes of weather extreme events and our ability to improve the effectiveness of response to our warnings are all at great risk.
Today, I will focus on the unusual weather patterns we’re seeing and what I believe we need to do to better predict and manage them. I hope many of you will work with us to innovate new approaches to succeed together in this tough environment.
A YEAR OF EXTREMES: 12 BILLION-DOLLAR DISASTERS
2011 is already in the record books as a year of historic extreme events.
One of those new records is the number of events totaling at least $1Billion in damages.
Today, NOAA announces in 2011, there have now been 12 extreme weather events each totaling at least $1B. The previous record was 9, set in 2008. These 12 events are depicted on the slide.
Earlier this year, we announced there were 10 to date; today we announce numbers 11th and 12th:
The aggregate damage from these 12 events is approximately $52 billion.
We have not finished tallying damages caused by additional extreme events, such as the pre-Halloween winter storm that impacted the Northeast and the wind/flood damage from Tropical Storm Lee, so stay tuned for the final total # of $1B events and the aggregate damage total.
Please note that damages totaling less than a billion dollars individually are not included in this tally, even though many of them represent additional significant financial losses.
And the economic losses are far from the full picture. More than 1,000 people died from these disasters. Deaths this year are almost double the yearly average (~600).
Each of these events is a huge disaster for victims who experience them; collectively, they are an unprecedented challenge for the Nation – for the safety of citizens, the bottom line for businesses, and the societal stresses they engender.
Timely, accurate, and reliable weather warnings and forecasts are essential to our collective well-being, but also to the Nation’s ability to recover and prosper.
Now, I’ve emphasized how unusual this year has been, but a single year can just be an anomaly. Is that the case here? What are we documenting across years? And what might we expect in the future?
Globally, according to Munich Re, the frequency of extreme events has risen steadily over the past 20 years. The number of meteorological and hydrological events each tripled in that time.
IPCC Managing Extreme Events Report (SREX)
The IPCCrecently released a special reporton the Risks of Extreme Events and Disasters. In short, this report says that we can expect more of many of these extreme events. Here’s what the report says about 5 types of extreme events, ordered by how certain we are of the prediction:
In a separate study, a white paper from the White House Office of Science and Technology Policy tells us that it’s very likely that large-scale changes in climate have influenced – and will continue to influence – many different types of extreme events, such as heavy rainfall, heat waves, and flooding. Large-scale climate change is also likely to affect small-scale phenomena like severe thunderstorms and tornadoes, but the nature and the degree of that influence are very uncertain, particularly for tornadoes.
These patterns only underscore the importance of enhancing our ability to predict and manage these events.
If you examine the extreme events on this map, you’ll see they run the gamut – from highly localized and brief events like tornadoes to regional-scale weather events like hurricanes, snow and flooding where we can provide a longer lead time to prepare … to climate-scale events like the drought in the Southern Plains of Texas, Oklahoma and Louisiana where we can watch the conditions develop over multiple weeks.
These events have different underlying physical drivers.
Therefore, to observe, monitor, predict and manage the impacts of these events requires understanding everything from the vertical wind profile at one individual location (in order to predict favorable tornadic conditions) to large-scale weather patterns which stretch around the globe (blocking patterns which can bring on heat/cold waves, drought).
This means we need diverse observations from weather balloons, radar, satellites, networks of soil moisture sensors, ocean temperature sensors, and on and on.
Understanding, predicting, and managing extreme events requires an extraordinary amount of information about the physical state of the earth system, and how it’s changing from moment to moment and decade to decade.
I believe that one essential key to meeting this challenge is critical environmental intelligence. Just like ‘intelligence’ in the security world combines data, information, analysis, modeling, and assessment, so too does ‘intelligence’ in the environmental arena.
ENVIRONMENTAL INTELLIGENCE ACROSS THE WEATHER-CLIMATE CONTINUUM
NOAA is America’s trusted broker for weather and climate data, information, and warnings. NOAA provides weather and climate products and services on every temporal scale – from minutes to decades – and on every spatial scale from neighborhoods to global.
Environmental intelligence is needed on every time and space scale in which decisions are made.
Let me highlight 3 categories of services NOAA provides.
How do we continue to improve our environmental intelligence? By improving environmental observations, research and modeling.
Improving weather and climate forecasts will require:
And, intelligence is only as useful as the capacity to use it.
Improving effective response to forecasts will require:
With that framework, let me now focus on what we are doing to enhance both critical environmental intelligence and its effective use and where we are facing challenges.
Environmental intelligence starts with observation systems – satellites, ground-based monitors, planes, ships, buoys, moorings, weather stations, tall towers, weather balloons, underwater gliders, and so on.
Our ability to reduce vulnerability to extreme events is dependent on many factors, but most certainly on our ability to observe and monitor changes in the frequency and intensity of these events.
Major overarching challenges are to improve ability to detect extremes reliably, maintain continuity of observations, and improve the interdependence of observing systems.
For example, many of our in-situ observation networks were built to observe weather, and don’t have the long-term stability and accuracy required for climate.
During extremes, it’s difficult to get consistent and robust observations. This is especially true for smaller scale events, such as hurricane precipitation, tornado and severe thunderstorm winds, and hail.
I believe we are in need of new, cost-effective technologies to monitor extreme events.
Satellites must have a sufficiently long overlapping period of record or else there’s a discontinuity in the data. Achieving this goal is increasingly challenging in light of the expense of satellites coupled with tighter budgets.
To inform our decisions on which observing networks to invest in, we need to understand the current predictive value of each system at various levels of investment.
For example, how much better are the predictions with twice as many observing points? How much worse are they with half as many? To get this information, we need to link observing systems with modeling experiments.
Observing Systems Simulation Experiments (OSSEs) do just that. OSSEs are a powerful tool at our disposal. They are typically aimed at assessing the impact of hypothetical, simulated data on a model or forecast system.
To ensure that modelers get the data they need, modeling requirements must be better communicated to the folks who design observing systems.
Comparing multiple independent observing systems allows the scientific community to increase their confidence in longer term trends.
Above and beyond having good observations, their analysis is critical. What are some of the challenges we face in the data analysis arena?
One opportunity that looms large is to make better use of multiple independent data sets – akin to using model ensembles.
PURPOSEFUL REDUCTION OF STRUCTURAL AND STATISTICAL MONITORING ERRORS
This slide depicts the use of multiple data sets, in this case, indicators drawn from Essential Climate Variables (ECV).
This graphic is based on the Bulletin of the American Meteorological Society’s State of the Climate report from 2009, with some modifications by NOAA’s National Climatic Data Center, NCDC.
It shows 10 key physical indicators of the earth system:
Each indicator looks at multiple data independent data sets. This technique helps to increase the confidence in the data and mitigate the effect of structural or statistical error.
Looking at multiple indicators is also helpful. Since the physical properties that the indicators are measuring are connected, it’s not surprising that the measurements themselves are consistent with one another. This “suite of indicators” approach lends additional confidence the measurement, understanding, and ultimately prediction of a vast and complex system.
These graphs depict trends in annual averages. Of course, the means include the extremes, but next I’d like to show you a way we can look directly at trends in the extremes themselves.
MONITORING AND COMMUNICATING ABOUT EXTREMES: NOAA U.S. CLIMATE EXTREMES INDEX
The “indicator” approach produces a useful tool for both monitoring and communicating the diverse and highly variable nature of extremes.
This is a graphic of NOAA’s U.S. Climate extremes index.
The index includes the variables shown in the bullets, and it broadly covers extreme heat, cold, rain, drought, and land-falling hurricanes.
This kind of index can help us answer questions from industry, from emergency managers, from infrastructure planners, and from the public, all of whom have strong interest in the current state of extremes, and what we might expect in the future.
For example, notice the upward trend in recent decades. While the index has been high in the past, the change from about 1970 to the present day is quite pronounced and qualitatively different from the preceding years.
What’s driving this index up? What observations are contributing to the trend? Decision-makers use this kind of information to help plan and manage their response to extreme events. Let’s look a little closer.
DECONSTRUCTING CLIMATE TRENDS: What’s driving the increase since the 1970s?
Deconstructing the index reveals what is driving the trend: in this case it is extremes in maximum and minimum temperature, too little and too much water, and 1-day heavy precipitation.
These graphs are individual elements of NOAA’s U.S. Climate Extremes Index. The vertical axes continue to be the percent of the country affected by the given extreme. All graphs are for the January to October period, each year from 1910 to 2011.
These are the primary components that are driving the changes in the Climate Extremes Index: extremes in maximum and minimum temperature, too much and too little soil moisture, and 1-day heavy precipitation events.
Extremes in maximum temperature mean unusually warm or cold daily high temperatures – either in the top 10% or bottom 10% of maximum temperatures. You can visually see that since 1970, more of the country is experiencing unusually warm highs (the red bars) and less of the country is experiencing unusually cool highs (the blue bars).
But, more striking is the change in the extremes in minimum temperature related to unusually warm or cold nighttime lows. More of the country is experiencing unusually warm nights, and less of the country is experiencing unusually cool nights. Warm overnight lows are related to heat stress in both people and plants and animals – they never get a chance to cool off – so this data set is of particular importance to those managing the response to the extremes.
What does the Index tell us about too much or too little soil water? The green bars show the percent of the country in extremely wet conditions, and the brown bars show extremely dry conditions.
While the country has had periods of more severe drought – for example, during the “Dust Bowl” years of the 30’s – we’ve never seen the country get both drier and wetter at the same time as this graph shows in the past decade.
And finally, the 1-day heavy precipitation graph shows more single days with precipitation much above normal (top 10%). This is important for hydrological engineers, water resource managers, and emergency managers in flood-prone areas, among others.
A variety of observations, taken over time, and used in well-crafted ensembles, can improve understanding and management of extremes.
These observational data – our historical record – are the vital input to models critical to predicting and projecting the future state of the Earth system.
Extremes occur on multiple scales. Modeling should match the time and space scales.
Here we’re looking at the observed versus modeled results for the severity of summer heat waves.
The model was run by NOAA’s Geophysical Fluid Dynamics Laboratory (GFDL).
As you can see, the climate models are beginning to resolve the key features of regional-scale events such as heat waves.
The output of global climate models often needs to be “downscaled” to be of greater use at regional and smaller scales. When global climate models are run on grids with large spaces between the points, they sometimes can’t resolve features of interest to decision-makers.
Downscaling historically involved either nesting a regional climate model into a global climate model, called “dynamical downscaling,” or using various statistical techniques to project the state of the system in between grid points.
However, newer high-resolution models running on supercomputers, can resolve much finer detail.
Combined with established downscaling techniques, these new models allow modelers to perform “right scaling”: generating results at the right scale for the phenomenon of interest or the application of the results.
As we saw earlier with observations, analyzing the results of ensembles of several models together allows modelers to improve predictive skill, and also to evaluate the different characteristics of different models.
An overarching challenge is to focus modeling activities on topics of particular societal interest, like extreme events, and at timescales which are highly relevant to decision-making. The seasonal timescale is particularly important to agriculture, for example, and to water managers who need to plan their resources for the upcoming seasons. The decadal timescale is critical for infrastructure planners, and for the construction and some insurance sectors, who need accurate pictures of the likely climate in 10-30 years.
As I pointed out earlier, many sectors of society are vulnerable to changes in weather and climate – particularly the destructive effects of extremes. In order to make sound decisions, people need to know what the trends in these types of events are, and the extent to which human activity may be affecting the trends.
Next let’s take a look at what influences our ability to detect trends in extreme events and attribute any changes to various causes.
DETECTION AND ATTRIBUTION
The current state of the science for detection and attribution of changes in extremes varies greatly by type of extreme event.
This graph is the result of a series of workshops that NOAA convened to investigate the state of the science for detection and attribution of trends in extreme events. The graph is based on the papers summarizing the findings of the workshops, some of which are already in journal review, for example, in the Bulletin of the American Meteorological Society.
The circles represent the various types of extremes considered at the workshops: extreme storms, heat, cold, drought and floods.
Each type of extreme event was evaluated for the adequacy of our current level of physical understanding of what might be driving changes. Each event was also evaluated for the adequacy of the available data to detect changes, including trends.
So for example, we have a relatively good understanding of what can lead to changes in heat waves, and our data are fairly adequate to detect changes in heat waves over time.
However, with ice storms, both our physical understanding of what factors would affect a trend, and our data on their frequency of occurrence and intensity, are much less adequate.
Dr. Ken Kunkel of our Cooperative Institute for Climate and Satellites reported earlier this week on our present understanding and ability to detect trends in heat waves and extreme precipitation events.
DECISION SUPPORT TOOLS
Climate or long-range forecasts look ahead weeks and months; these are the forecasts that people, communities, and businesses need to prepare for many extreme conditions. These are the tools critical for communicating uncertainty and risk.
We know how very useful these tools have been recently. Two examples:
And we know that more and more communities are looking for and using these tools to plan, for example to adapt to a climate-influenced future. Again, two examples:
Businesses also use climate forecasts and data strategically. Three examples here:
Another decision-support tool to facilitate long-range planning/climate adaptation is NOAA’s Digital Coast tool, shown here depicting the Sea Level Rise Impacts Viewer tool.
THE PATH FORWARD
What’s likely down the road on the weather-climate front? increased overall warming; amplified water cycle; and more extreme events and more variability, i.e. more wild weather and wild swings in weather.
As the nation's requirements for environmental intelligence continues to grow, two things happen:
Our experiences with budgets last year and this year are likely a harbinger of challenging times ahead for sustaining and improving our ability to predict and therefore manage extreme events. The road ahead has great uncertainty.
Observing systems, research, and high performance computing are all absolute prerequisites to producing weather and climate forecasts: and all are at risk.
We need to make sure that the current economic and political landscapes don’t erode our ability to provide accurate, reliable forecasts.
The Nation’s need to understand, predict and manage our response to weather and climate extremes exceeds the scope of any individual organization or government agency. Collaboration is vital.
Our partnerships with academia, industry, other federal agencies, and the international community are the cornerstones for successfully securing the environmental intelligence we need in the future.
A significant role exists for AGU in promoting and supporting partnerships in the prediction and management of extreme events and in making sure that scientific integrity, a core value of science and democracy, is protected.
NOAA’s mission is to understand and predict changes in the Earth's environment, from the depths of the ocean to the surface of the sun, and to conserve and manage our coastal and marine resources. Join us on Facebook, Twitter and our other social media channels.