What data do we require for extremes analysis and what is available?

1

• What data do we require for extremes
  analysisand what is available?
• (an intro to the BOG on data)
• Albert Klein TankKNMI, The Netherlands

Warning no attempt to be comprehensive bias
towards land-atmosphere bias towards Europe
2
Outline

• Russian heat wave (yes, one more time)
• GCOS IP-2010
• Data Policy white paper from the Exeter workshop
  oncreating surface temperature datasets
• Other relevant initiatives
• Further issues for BOG discussion

3
Example Russian heat wave July 2010
4
Example Russian heat wave July 2010
Enough data for statistical modelling, event
attribution, or studying physical processes, but
Courtesy John Christy (top), Adrian Simmons
(bottom)
5
Example Russian heat wave July 2010
6
Example Russian heat wave July 2010
31 days with Txgt25C normal is 9.5 days
Data available from http//eca.knmi.nl
7
Example Russian heat wave July 2010
16 nights with Tngt20C normal is 0.5 night
Data available from http//eca.knmi.nl
8
Example Russian heat wave July 2010
No significant trend at Moscow
16 nights with Tngt20C normal is 0.5 night
Data available from http//eca.knmi.nl
9
Daily griddedproduct (E-OBS)

• Based on station records
• Daily fields
• 1950 now
• 0.25 deg resolution
• Matching RCM grids
• Associated error fieldsexist, but rarely used!

Haylock et al., J. Geophys. Res., 2008 data
available from http//eca.knmi.nl
10
Daily griddedproduct (E-OBS)

• Based on station records
• Daily fields
• 1950 now
• 0.25 deg resolution
• Matching RCM grids
• Associated error fieldsexist, but rarely used!

Haylock et al., J. Geophys. Res., 2008 data
available from http//eca.knmi.nl
11
Traceability

• About 60 ispublic, i.e.available fromthe
  ECADwebsite
• For the otherstations, onlythe metadataand
  derivedproducts canbe released

Data available from http//eca.knmi.nl
12
GCOS IP-10
13
GCOS IP-10

• Observations are required forinformed decisions
  on prevention, mitigation, and adaptation
  strategies to support research to initialise
  predictions to develop the models to assess
  social and economic vulnerabilities
• Observations are essential public goodsbenefits
  of global availability of data exceed any
  economic or strategic value to individual
  countries from withholding national data
• Observations underpin all efforts by Parties to
  the United Nations Framework Convention on
  Climate Change (UNFCCC) to mitigate, and adapt
  to, climate change

14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
Some GCOS IP-10 issues

• Increasing need for local, high-frequency surface
  atmospheric data on climate to characterise
  extremes
• Satellite remote-sensing systems become more
  important, but surface-based and airborne in situ
  and remote-sensing systems will always remain
  essential
• Metadata (i.e., information on where and how the
  observations are taken) are absolutely essential
• Limited progress in developing countries, and
  support for capacity-building is small in
  relation to needs
• Flow of data to the user community and to the
  international data centres is inadequate
• Parties should produce national plans on their
  climate observing system

20
Exeter workshop
21
Exeter workshop

• Several white papers, e.g. on Data Policy
• Gap exists between theory (including GCOS and GEO
  requirements) and practice
• Partly due to data policy, but not alone
• Also lack of engagement, lack of resources, and
  inadequate data-system infrastructures
• Data policy issues are persistent and unlikely to
  go away in the near future

See http//www.surfacetemperatures.org
22
Exeter workshop

• Some recommendations
• Accept trade off between traceability and data
  completeness
• Acknowledge that involvement of data providers
  from countries throughout the world is essential
• Involves more than simply sending the data to an
  international data centre
• Scientific community to deliver information to
  support local climate services ( return of
  investment important in particularly for
  developing countries)
• Support digitization of paper archives

23
Other relevant initiatives

• Global
• coordinated inter-callibration (GSICS) and
  reprocessing (SCOPE-CM) of satellite data
• ICOADS version 3 released (almost)
• several reanalyses datasets released (MERRA,
  CFSR, JRA)
• 20th Century reanalysis (based on surface
  pressure data only)
• new global reanalysis project by ECMWF (ERACLIM)
  with much attention for improving reanalysis data
  input
• new project (CLIMDEX) for updating the global
  dataset of extremes indices and developing global
  daily gridded datasets (building on GHCN-Daily)

24
Other relevant initiatives

• Regional
• regional reanalysis for North America (NARR)and
  Europe (http//www.euro4m.eu )
• daily station collections ( daily gridded
  datasets) for Asia (APHRODITE) and South America
  (CLARIS-LPB)
• several new national high resolution datasets

More in WOAP4 Meeting Report Hamburg,
Germany March 2010 GCOS Publication No. 142
25
Index for heavy falls
What about the blank regions in the map?
Alexander et al., JGR, 2006 also in IPCC-AR4
26
ETCCDI Regional Workshops(complemented by APN)
Peterson and Manton, BAMS, 2008
27
Regional workshops

• Organised by the Expert Team on Climate Change
  Detection and Indices (ETCCDI)
• ETCCDI is a group of scientists jointly sponsored
  by several international agencies
  (WMO-CCl/WCRP-CLIVAR/JCOMM)
• Environment Canada provides, maintains, and
  further develops the R-based workshop software
  (freely available from http//cccma.seos.uvic.ca/E
  TCCDI)

28
ETCCDI Regional Workshops(complemented by APN)
Working together
Regional workshops successful concept, but
often no access to original data!
Peterson and Manton, BAMS, 2008
29
Further issues for BOG (1 of 2)

• Tension between traceability (access to the
  primary sources)and data completeness (use
  whatever available)
• Need high density, high frequency, sharing, and
  long records
• Adaptive strategies for dealing with extreme
  events place even higher demands on observations
• Datasets need continuous work, both for updating
  and improving quality/homogeneity
• Including scientist developing datasets in
  research projects is a good idea, e.g. in CMIP5
  climate model evaluation
• Met Services are not keen if their only role is
  providing data application relevant products are
  a necessary return of investment

30
Further issues for BOG (2 of 2)

• Need to close the gap between rapid IT
  developments and actually implementing modern
  distributed database management systems
• Reprocessing of data and reanalysis important
• Some (satellite) datasets are becoming so large
  that it is difficult for many users to acquire
  them
• Users ask for products that meet their
  requirements, often through integration of data
  from different sources (in situ, satellite,
  reanalysis)
• Work needs to comply with WMO/GCOS/GEO ideas on
  systems/standards, but in the end the actual
  delivered datasets count rather than nice words

31

• mailto Albert.Klein.Tank _at_ KNMI.nl