Transcript cutting-edge climate science and services

1
CUTTING-EDGE CLIMATE
SCIENCE AND SERVICES
Geoff Love
An essential task of
the climate researcher
is to increase our
understanding of the
climate we experience.
In climate science
there is a history of
researchers using data
sets to identify
statistical relationships
between variables. The
science challenge is to
understand why there
is a correlation and to
use that information in
a predictive sense.
2
A fundamental conundrum in meteorology is that
we have good predictability for about 5 to maybe
7 days in the mid-latitudes. But beyond that not a
lot of skill in forecasting – and what we do have is
probably due to tropical forcing.
Whereas, in the tropics, short term predictability
is low. Forecasters rarely beat persistence on 1 - 3
day time-scales unless there is a major synoptic
event like a tropical cyclone. However, on longer
timescales (of the order of a month and beyond),
low, but positive skill is provided by a variety of
tropical features.
3
Seasonal forecasts, on
timescales of 1 month to
1 year have lower skill
than weather forecasts
and have higher skill in
the tropics.
Scenarios for climate
change on timescales of
greater than 1 year are
informed by changing
composition of the
atmosphere – a feature
that is reasonably well
predicted.
PREDICTABILITY
Mid-latitude weather
forecasts have reasonable
accuracy out to a
maximum of two weeks
(currently somewhat
less).
Mid-latitudes
Tropics
Global
4
5
6
7
8
The cutting-edge science questions this arrangement of the
atmosphere (clouds, clear bits, moisture, thunderstorms,
winds, etc) poses are:
• What are the time and space scales of organisation?
• Why are they like that?
• How can we represent these in models of the atmosphere?
• What do we have to measure to: (1) specify the significant
parts of the atmosphere properly, and (2) how should we
analyse these observations so that our predictive models can
make best use of our observations?
9
Matt
Wheeler’s
answer is
that the
atmosphere
(and the
tropical bit in
particular) is
made up of a
zoo of
propagating
waves.
10
11
CHANGING LARGE-SCALE PATTERNS
Temperature anomaly
Over the period 1950 to
2003 there appears to
have been a systematic
change in temperature and
rainfall
12
ANOTHER LARGE-SCALE FEATURE IS THE
WALKER CIRCULATION
The Walker Circulation – the
schematic circulation between
South America and Southeast
Asia/North Australia
13
So, is the Walker Circulation responding to these
changes in global temperature?
It seems to me to be a cutting edge science question
to answer this.
An El Nino event is a weakening of the Walker
Circulation, a La Nina event a strengthening. There
are many systems, both statistical and dynamical that
attempt to model, in a predictive sense, this
Circulation.
To succeed in modelling the Walker Circulation, so
that we could predict its changes, would be a cuttingedge scientific achievement.
14
The Bureau of
Meteorology
operates the
POAMA
system, and
using it
routinely
produces an
ensemble of
predictions for
the strength of
the Walker
Circulation.
15
ENSO AND EXTREME EVENTS
Average annual
number of tropical
cyclones in 2o X 2o
grid boxes (1969 –
1998 climatology).
In the La Nina years
there are more
tropical cyclones,
and they are closer
to the Australian
coastline, than in the
El Nino years.
16
RAINFALL AND ENSO
Variability of Annual rainfall
20
18
16
Coefficient (%)
The story of
Australian climate,
and particularly
rainfall, is closely tied
to what happens with
the El Nino / La Nina
cycle.
14
12
10
8
6
4
2
0
Australia S. Africa Germany France
NZ
India
UK
Canada
China
USA
Russia
Country
La Nina
El Nino
17
RAINFALL AND EL NINO
23 El Ninos in the past 100 years, 11
to 1950 and 12 since.
Composite pictures of the summer
and winter rainfall anomalies for the
twelve most significant El Ninos.
summer
THE EL NINO YEARS
1902-1903, 1905-1906, 1911-1912, 1914-1915
1918-1919, 1923-1924, 1925-1926, 1930-1931
1932-1933, 1939-1940, 1941-1942, 1951-1952
1953-1954, 1957-1958, 1965-1966, 1969-1970
1972-1973, 1976-1977, 1982-1983, 1986-1987
1991-1992, 1994-1995, 1997-1998
winter
18
RAIN AND LA NINA
18 La Ninas in the past 100 years,
9 up to and including 1950, and 9
since.
summer
Composite pictures of the summer
and winter rainfall anomalies for
the twelve most significant La
Ninas
THE LA NINA YEARS:
1904-1905 , 1909-1910, 1910-1911, 1915-1916
1917-1918, 1924-1925, 1928-1929, 1938-1939
1950-1951, 1955-1956, 1956-1957, 1964-1965
1970-1971, 1971-1972, 1973-1974, 1975-1976
1988-1989 , 1995-1996
winter
19
SCENARIOS FOR RAIN AND
TEMPERATURE
SUMMER - NORTH
The yellow ellipse encompasses
95% of the 1000 year unforced
runs projections of
temperature and precipitation
for the Northern Australian
region of the Canadian
CGCM2 model.
About 1oC warming and 15% more
rainfall in the north in summer.
Northern Australia extends from the north
coast to 30oS.
The blue ellipse is from the UK
Hadley Centre HadCM3
model.
Source: Ruosteenoja, Carter, Jylha and
Tuomenvirta, “ Future climate of world
regions: an intercomparison of modelbased projections for the new IPCC
emission scenarios”. Pub: Finnish
Environment Institute. Helsinki 2003.
Blue – HadCM3
Yellow - CGCM2
20
SCENARIOS FOR RAIN AND
TEMPERATURE
WINTER - NORTH
About 1oC warming
and no change in
rainfall in the north
in winter.
Blue – HadCM3
Yellow - CGCM2
21
SCENARIOS FOR RAIN AND
TEMPERATURE
SUMMER - SOUTH
About 1oC
warming and
maybe slightly
more rainfall in the
south in summer.
Blue – HadCM3
Yellow - CGCM2
22
SCENARIOS FOR RAIN AND
TEMPERATURE
WINTER - SOUTH
About 1oC warming
and no change in
rainfall in the south
in winter.
Blue – HadCM3
Yellow - CGCM2
23
For broad scale scenarios to be truly useful they need to be
down-scaled so as to inform private individuals, farmers,
fishers and small- and large-business operators how climate,
climate change and climate variability will affect them.
Accurate,
verifiable
downscaling is
cutting-edge
science
24
After developing scenarios for climate change and
downscaling these they must be extended to better
understand impacts on biological and economic systems
(most particularly on global trade and national energy
policy).
High quality
Earth-system
simulation is
cutting-edge
science
25
CLIMATE SERVICES
Must be
what the
user
community
needs and
wants
Million
Annual Hits on the Bureau of Meteorology's
Climate Web Pages
80
70
60
50
40
30
20
10
96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05
26
THE MOST POPULAR PRODUCTS
27
THE MOST POPULAR PRODUCTS #1
28
29
THE MOST POPULAR PRODUCTS #2
30
THE MOST POPULAR PRODUCTS #3
Temperature
Temperature anomaly
31
THE MOST POPULAR PRODUCTS #4
Other charts; eg NDVI and various
seasonal timescale forecast products
32
WHAT ELSE ?
Climate service users want:
• More data – radar, observations, predictions;
• Local scale information;
• Simple explanations or explanations of likely
uncertainties; and
• Forecasts at the scale of their enterprise with
long lead times and high accuracy.
33
THE FUTURE CUTTING EDGE
SERVICE ?
Cutting edge services will continue to make good use of the
most up-to-date science to develop products and the best
technologies to deliver these services.
Cutting edge services will inevitably follow the delivery of
cutting edge science
The cutting-edge service will develop partnerships to form
the link between scientists and users throughout the
community.
THANK YOU
34