Linking Weather and Climate S&TI

Randall M. Dole
NOAA Earth System Research Laboratory
July 2006

This presentation provides an overview of major phenomena and mechanisms linking

weather and climate variations, with a focus on a subset of major, recurrent phenomena that
impact extratropical wintertime weather and climate variations over the Pacific-North
American region. While progress in advancing understanding has been impressive, research
has also illuminated areas where significant future gains are possible. Emerging thrusts in
international and national research priorities suggest that over time artificial distinctions will
be removed between "weather" and "climate", as we begin to achieve a more unified
understanding of phenomena and processes that bridge time scales. We discuss these
research thrusts, which are likely to serve as increasingly vital components of an overall
research strategy in earth system science.
Introduction
Historically, the atmospheric sciences have tended to treat problems of weather and climate
separately. The real physical system, however, is a continuum, with fast weather processes profoundly
influencing climate variations and change, and, conversely, slower climate fluctuations and change
affecting the weather that we experience. While this past approach has served important purposes, it is
becoming increasingly apparent that in order to make progress in addressing many societally important
problems, from assessing risks of future hurricane activity to the possibility of abrupt climate change, we
require an improved understanding of the connections between weather and climate. As we move toward
a more unified view of phenomena and processes across time scales, the problem of understanding the
relationships between fast and slow processes - effectively, the links between weather and climate - is
emerging as an important new research priority. This presentation suggests some near-term directions
where prospects for advances appear most likely in building our understanding of the relationships
between weather and climate.
In approaching this broad problem, it is useful to consider three related sub-questions:
1) How do climate variations and change affect weather phenomena?
2) How do weather phenomena affect climate variations?
3) What are key phenomena and processes that bridge the time scales between synoptic-scale weather
(time scales of order a few days) and climate variations of a season or longer?
To give a specific example for the first question, we might ask how do climate variations (or change)
influence the probability of occurrence or potential intensity of hurricanes? This is a topic of enormous
societal importance, and an active focus for current scientific research. The reciprocal question is: what
are the effects of hurricanes on the climate system? Hurricanes clearly impact the atmospheric heat
budget through the transfer of heat from the tropical oceans into the tropical upper atmosphere, and
potentially also to higher latitudes. More subtly, by altering the upper ocean heat balance hurricanes can
change the ocean thermohaline circulation, and thereby induce longer-term climate variations. More
generally, many hydrological processes, such as clouds, moist convection, and water vapor transports
operate on fast "weather" time scales, and yet also profoundly impact climate variations and change. The
third question addresses the "no man's land" that lies between approximately a week and a season.
Tropical variations such as the Madden-Julian Oscillation (MJO) and air-sea coupling are examples of
________________
Correspondence to: Dr. Randall M. Dole, NOAA Earth System Research Laboratory, 325 Broadway,
Boulder, CO 80305
E-mail: Randall.M.Dole@noaa.gov
July 2006 http://www.nws.noaa.gov/ost/climate/STIP/

important phenomena and processes that occur on these scales.

While the presentation touches on the first two questions, the main emphasis is on this third question and,
especially, on identifying potential directions for near-term progress. In addressing this question, it is
important to recognize that the problems can and should be attacked from multiple directions, from
the weather side out (from days to a few weeks out), from the climate side in (how seasonal and
longer fluctuations affect weather variability), as well as through analyses of phenomena and processes
that bridge time scales.
High priority Issues
Issue 1: Organization of tropical convection
This is the mega-issue. It cuts across time scales from short and medium-range weather forecasts to
understanding and projecting regional climate variations and change. Studies suggest that beyond 7-10
days, variations in tropical heating are likely the major source for potential predictability. While there
have been recent improvements, both weather and climate models continue to show pervasive
deficiencies in representing the magnitudes, structures, and propagation of organized tropical convection.
The canonical (but by no means only) problem is that of predicting the evolution the MJO.
Issue 2: Tropical-extratropical interactions
Rossby wave propagation is a dominant feature that influences potential predictability of weather and
climate. So getting that right is fundamental. But how do slowly varying forced Rossby wave trains
affect weather variability, as manifested by changes in storm behavior and rainfall? Shapiro et al. (2001)
demonstrate how the life cycle of synoptic eddies display radically different behavior between the El
Nio winter of 1997-98 and the subsequent La Nia winter of 1998-99. These differences result in
striking differences in rainfall amounts and patterns, altering regions that are at greatest risk from high
impact events as well as the type of storm event (i.e., warm high snow level storms vs. cold low snow
level storms). The changes in the storm system structures between the two years imply large changes in
synoptic-scale eddy momentum and heat transports, which can feedback onto the more slowly varying
large-scale flow patterns. Bao et al. (2006) show that California flooding events are directly associated
with narrow, concentrated plumes of water vapor transported from the tropics to extratropics, with the
likelihood of tropical-extratropical interactions depending on whether tropical Pacific ocean conditions
were in El Nio, La Nia, or near-neutral states. The importance of these very narrow tropical plumes has
significant implications for climate models. It is unlikely that present generation climate models, as are
used, for example, in the IPCC assessment, are of sufficiently high horizontal or vertical resolution to
properly simulate these
narrow plumes, often termed
atmospheric rivers, in
which a substantial majority
of the poleward moisture
transports is apparently
accomplished.
Issue 3: Connecting
phenomena and processes
across time scales
Figure 1 (from the 2006 MJO
Workshop, Trieste, Italy)
illustrates the complexity of
interactions in tropical
convection across space and
time scales. The failure to
model properly processes that
Office of Science and Technology Climate & Weather 2
NOAAs National Weather Service Connection
July 2006 Science and Technology Infusion Climate Bulletin

occur on even diurnal to sub-diurnal processes can alter the time-mean convective heating, which has
direct implications for simulating climate. Ocean-atmosphere interactions present a rich interplay between
relatively fast and slow processes, and appear to be an important bridge across time scales. For example,
relatively fast westerly wind bursts associated with synoptic to intraseasonal time scales can trigger ocean
Kelvin waves that influence the evolution of ENSO events, including onset and decay.
Issue 4: Assessing Potential Predictability
Evidence is growing for potential predictability beyond the first forecast week. For example, 8-14 day
forecasts prior to the La Conchita Floods/Debris flow disaster (Fig. 2 left panel) showed above normal
(upper tercile) rainfall probabilities of around 70% for the region that was subsequently affected, with
good indications of a strong connection between the tropics and mid-latitudes consistent with flooding
scenarios. By 6-10 days in advance (Fig.2 right panel), the probability of upper tercile rainfall values
reached 80-90%. Such high probabilities in the upper tercile were strongly suggestive of a potential for
exceptional rainfall, as has been confirmed in other work carried out by Hamill and Whitaker looking at
shifts the forecast probability distribution functions for precipitation during this period.

Figure 2 Left panel: week-2 (8-14 day) forecast of the probability (in percent) of above-
average precipitation for a forecast initialized at 0000 GMT on 2 January 2005, using
calibrated probabilities based on reforecasts. (The La Conchita, CA landslide occurred on
12 January 2005, following massive rainfall along the California coast). Right panel: 6-10
day forecast of the probability of above-average precipitation for a forecast initialized at
0000 GMT on 3 January 2005.

Toward Bridging Weather and Climate

We are beginning to see real evidence for the convergence of weather and climate research. For weather
research, The Observations, Research, and Prediction Experiment (THORPEX) under World Weather
Research Program (WWRP) is focusing on improving the accuracy of forecasts of high impact weather
conditions out to two weeks in advance. There are initial plans for a THORPEX Pacific-Asian Regional
Campaign (TPARC) (2008-2009) that will focus significant efforts on improving understanding of
tropical-extratropical interactions. Within climate programs, the Coordinated Observation and Prediction
of the Earth System (COPES) under the World Climate Research Program (WCRP) proposes to address
the seamless prediction problem from weeks to centuries in advance. Thus, there are the beginnings of a
bridge between major weather and climate programs on time scales of a few weeks. A Year of
Coordinated Observing, Modeling and Forecasting is being discussed that would help link weather and
climate programs to focus on improving understanding of organized tropical convection and its impacts
on the extratropics.
Beyond these programs, the U.S. Climate Change Science Program (CCSP) Strategic Plan identifies as
one of its high priority questions how extreme events, such as droughts, floods, wildfires, heat waves and

Office of Science and Technology Climate & Weather 3

NOAAs National Weather Service Connection
July 2006 http://www.nws.noaa.gov/ost/climate/STIP/

hurricanes are related to climate variability and change. It is also noteworthy that over the past several
years, major modeling centers in the U.S. and abroad are beginning to move from separate weather and
climate models toward more unified weather-climate model architectures.
Science Challenges
Some of the key near-term science challenges are:
1. Modeling and prediction of organized tropical convection (MJO, ISO )
2. Tropical-extratropical interactions; storm tracks and moisture transports.
3. Role of atmosphere-ocean interactions in intraseasonal variability.
4. Systematic zonal flow variations: mechanisms and predictive implications.
5. Predictability of tropospheric wave guides and baroclinic wave packets.
6. Troposphere-stratosphere interactions; modeling and potential predictability.
7. Variability of climate modes beyond ENSO and MJO.
8. Effects of global ocean conditions, e.g., tropical Indian and Atlantic Oceans.
9. Effects of land surface processes.
10. Warm season climate system and its predictability.
11. Potential implications of climate change.
Other Challenges
Traditionally, different scientific communities have focused on weather prediction and climate
prediction. We need to move past this dichotomy and build stronger links between the two communities.
From a programmatic point of view, research programs have tended to be stove piped along disciplines.
This has resulted in putting priority on activities in the heart of the disciplines, leaving cross-cuts - such as
weather-climate linkages - as orphans. These barriers must be overcome, and increased emphasis
placed on activities that bridge weather and climate research goals. There is a need to develop more
scientists who are fluent in both weather and climate problems, and in the associated methodological
approaches.
Summary
Progress in our understanding and capabilities to predict phenomena at the interface between weather and
climate has been impressive. Research also illuminates areas where substantial further advances are
possible. Emerging thrusts in international and national research programs suggest that over time
artificial distinctions will be removed between weather and climate as we begin to achieve a more
unified understanding of phenomena and processes across time scales. Over the next decade, it is likely
that improving understanding and capabilities to predict the links between weather and climate will serve
as increasingly vital components of an overall research strategy in Earth system science.
Reference
Bao J, Michelson SA, Neiman PJ, Ralph FM, Wilczak JM, 2006: Interpretation of enhanced integrated
water vapor bands associated with extratropical cyclones: Their formation and connection to tropical
moisture. Mon. Wea. Rev. 134(4), 1063-1080
Shapiro, M.A., H. Wernli, N.A. Bond, and R. Langland, 2001: The influence of the 199799 El Nio
Southern Oscillation on extratropical baroclinic life cycles over the eastern North Pacific. Q. J. R.
Meteorol. Soc., 127, 331-342.

Office of Science and Technology Climate & Weather 4

NOAAs National Weather Service Connection