Observations of climate change

1
Observations of climate change
Help!
Kevin E Trenberth NCAR
2
Global Warming is unequivocal

• Since 1970, rise in Decrease in
• Global surface temperatures NH Snow extent
• Tropospheric temperatures Arctic sea ice
• Global SSTs, ocean Ts Glaciers
• Global sea level Cold temperatures
• Water vapor
• Rainfall intensity
• Precipitation extratropics
• Hurricane intensity
• Drought
• Extreme high temperatures
• Heat waves

IPCC 2007
3
Global Warming
The climate is changing. We can and should take
mitigating actions that will slow and eventually
stop climate change. Meanwhile we must adapt to
climate change. But adapt to what? We do not have
predictions. We do not have adequate reliable
observations. We do not have the needed
information system!
4
Global mean temperatures are rising faster with
time
Period Rate Years ?/decade
IPCC 2007
5
Heat waves are increasing an example
Extreme Heat Wave Summer 2003 Europe 30,000 deaths
IPCC 2007
6
Surface Temperature 1901-2005
IPCC 2007
7
Drought is increasing most places
Mainly decrease in rain over land in tropics and
subtropics, but enhanced by increased atmospheric
demand with warming
The most important spatial pattern (top) of the
monthly Palmer Drought Severity Index (PDSI) for
1900 to 2002. The time series (below) accounts
for most of the trend in PDSI.
IPCC 2007
8
Extremes of temperature are changing! Observed
trends (days) per decade for 1951 to 2003 5th
or 95th percentiles From Alexander et al. (2006)
IPCC 2007
9
Increases in rainfall and cloud counter warming
Drought
Absence of warming by day coincides with wetter
and cloudier conditions
Trend in Warm Days 1951-2003
IPCC 2007
10
Regional climate change Hypothesis It is
impossible to address regional climate change
without fully addressing how patterns of climate
variability (modes) change, and thus how ENSO
El Niño Southern Oscillation NAO/NAM North
Atlantic Oscillation/Northern Annular Mode SAM
Southern Annular Mode PDO Pacific Decadal
Oscillation AMO Atlantic Multidecadal
Oscillation change!
11
El Niño - Southern Oscillation
SLP Surface temperature Precipitation
IPCC 2007
12
Pacific Decadal Oscillation SST pattern (above)
and time series (lower right) of 1st EOF of N
Pacific SSTs. NPI index of Aleutian Low Indian
Ocean SST (tropics)
IPCC 2007
13
Many observed climate anomalies can be simulated
in models with specified SSTs

• Sahel drought Hurrell et al 2004, Giannini et
  al 2003, Hoerling,
• US Dust Bowl Schubert et al. 2004, Seager et al.
  2005
• Drought (US, Europe, Asia) Hoerling and Kumar
  2003
• But we can not (yet) simulate the observed SSTs.

14

• Global increases in SST are not uniform. Why?
• Coupling with atmosphere
• Tropical Indian Ocean has warmed to be
  competitive as warmest part of global ocean.
• Tropical Pacific gets relief owing to ENSO?
• Deeper mixing in Atlantic, THC.
• This pattern is NOT well simulated by coupled
  models!
• Relates to ocean uptake of heat, heat content
  transport.

IPCC 2007
15
IPCC experience on observations

• Sorting out the climate signal from the noise in
  inadequate observations from a changing observing
  system is an ongoing continual challenge
• Space-based observations are a particular
  challenge

16
Temperatures
IPCC 2007
Annual anomalies of maximum and minimum
temperatures and diurnal temperature range (DTR)
(C) averaged for the 71 of global land areas
for 1950 to 2004. DTR 1979-2004

• Issues
• Missing data and treatment
• Quality control
• Max and Min T much more
• sensitive to inhomogeneities
• 4. Urban heat island
• 5. Need to continue to pressure
• countries to provide high
• frequency data (hourly and daily)

17
Radiation Top-of Atmosphere Wielicki et al.
2002

• Published Science
• Revised following comment
• Edition 2 (orbit decay correction)
• Edition 3 (SW filter dome)

IPCC 2007
18
Precipitation not a continuous variable
Large differences in amounts. Inability to
analyze characteristics intensity, frequency,
duration, type, as well as amount. Need hourly
data!
IPCC 2007
19
Tropical rainfall 30N-30S Land Total
Ocean
Issues Need much more complete and better data
on all hydrological variables set in a holistic
framework Precipitation hourly (intensity,
frequency, duration, type, amount) streamflow,
runoff, evaporation, drought indices, soil
moisture (incl ice), snow cover depth
Land systematic offset 3 Ocean no
relationship Total dominated by ocean
20
North Atlantic hurricanes have increased with SSTs
N. Atlantic hurricane record best after 1944 with
aircraft surveillance. Global number and
percentage of intense hurricanes is increasing
(1944-2005)
SST
21
Some issues Partial reprocessing of ISCCP data
has occurred for tropical storms (Kossin) Records
are far from homogeneous, even for satellite
era Records/practices are not comparable in
different regions, even now. We desperately need
an internationally coordinated reprocessing of
all satellite data for hurricanes, to get many
parameters of interest, such as size, intensity,
rainfall, integrated variables (0-100 km 0-400
km) etc.
Ivan 2004
22
Main Issues

• The in situ data are not global and have problems
• Satellites drift in orbit and instruments
  degrade the data generally do not provide a
  climate record. They could.
• The satellite record is in jeopardy, especially
  from demanifesting several climate instruments
  from NPOESS.
• A baseline transfer standard is essential in
  situ super sites (reference radiosonde plus
  network).
• Regional climate requires attention to modes of
  variability and model initialization

23
Why do we need an integrated Earth System
Analysis?

• We have a lot of observations from satellites
  and other remote sensing.
• The volumes are huge
• We use but a small fraction
• Most are not climate quality
• Inconsistencies exist across variables
• They do not make a climate observing system
• Reprocessing and reanalysis must be part of system

Goal Climate Data Records
24

• There is a need to better come to grips with the
  continually changing observing system.
• There is no baseline network to anchor the
  analyses or space observations.
• The radiosonde network is not it!
• The challenge is to improve continuity and be
  able to relate a current set of observations to
  those taken 20 years ago (or in the future).
• There is a need for more attention to data
  synthesis, reprocessing, analysis and re-analysis
  of existing data sets and
• 5. There must be a baseline set of measurements
  
• ? Sparse network (30-40) of reference sondes
  for satellite calibration and climate monitoring,
  UT water vapor co-located with regular sonde
  sites to replace them at appropriate times
  integrated with ozone sondes and/or GAW and BSRN
  GRUAN?
• ? GPS Radio Occultation.

25

• The challenge is to better determine
• how the climate system is changing
• how the forcings are changing
• how these relate to each other (incl. feedbacks)
• attribution of anomalies to causes
• what they mean for the immediate and more distant
  future (assessment)
• Validate and improve models
• seamless predictions on multiple time scales
• how to use this information for informed planning
  and decision making
• how to manage the data and reanalyze it routinely
• how to disseminate products around the world
• how to interact with users and stakeholders and
  add regional value
• From Trenberth et al 2002

? ? ? ? ? ? ? ? ? ? ?
Information System