Breakout Groups

1
Breakout Groups

• ONE
• Francis Zwiers
• Stewart Cohen
• Bryan Tugwood
• Rick Lee
• Ray Desjardins
• Seong Heon Kim
• Bill Richards
• Katja Woth
• Georges Desrochers
• FOUR
• Rob Wilby
• Dieter Riedel
• Mark Barton
• John Charlery
• Juno Song
• Kim Logan
• Paulin Coulibaly
• Suzan Lapp

• TWO
• Jean Palutikof
• Don Lemmen
• Julia James
• Elaine Wheaton
• Bill Taylor
• Jim Byrne
• Nicole McKechnie
• Carrie Holcapek
• Mohammad Matiur Rahmen
• FIVE
• Alain Bourque
• Clare Goodess
• Sam Gameda
• Denise Neilsen
• André St. Hilaire
• Gary Lines
• VTV Nguyen
• Adam Fenech

THREE Hélène Côté Eric Taylor Ben
Kangasniemi Vanessa Egginton Abdel Maarouf John
Anderson Shusen Wang Jim Helbig SIX Dave
Sauchyn Xuebin Zhang Bing Rong Henry
Hengeveld Ross Herrington Jiafeng Wang Ge
Yu Martin Lacroix Juraj Cunderlik
2
Extremes Scenarios Workshop Breakout Group 1

• Cohen, Zwiers, McKinnon, Lee, Desjardins,
  Richards, Kim, Woth, Desrochers, Tugwood

3
Q1most important aspects of variability/extremes

• Frequency, duration, severity, timing of events
• some elements will have to be derived outside of
  the GCM (e.g. soil moisture)
• others can be obtained directly (e.g. freeze-thaw)

4
Q2 which variables?

• Soil moisture
• water precipitation IDF
• atmospheric stability mixed layer depth
• sea level rise 6-hour event from RCM

5
Q3 what should a scenario consist of?

• Change in risk of a particular event
• RCM--scenarios of internally consistent multiple
  variables (e.g. warm-dry spells)

6
Q6 past climates

• What is more reliance?
• Can learn about undisturbed conditions
• can complement instrumental record
• possible use as temporal analogue (with caveats)

7
Q7 uncertainty

• Range of plausible outcomes, so specific language
  may not be desired (uncertain about use of IPCC
  terminology--needs further work)
• for impacts--scenario comes from a known source
  with unknown probability

8
Q8 downscaling

• Yes, impacts researchers want to use this
• could use either RCM or statistical methods (with
  appropriate caveats for each)

9
Q13 national strategy

• Hard to find another facility like CCIS goes
  beyond what UK-LINK does
• impacts community well served by CCIS better
  then what existed before this reduces gap
  between GCM output production and its use for
  impacts/adaptation research

10
Group 2
National Workshop on the Development of
Scenarios of Climate Variability and Extremes

• Data and Knowledge Gaps
• more data available starting with dew point,
  radiation
• major gaps on ocean parameters
• El Nino / NAO need Canadian research capacity
• time scales focus on relevance to stakeholders
  (nearer term)
• paleo needs to be integrated into all aspects of
  scenario development
• For IA work key variables need to be stakeholder
  defined need process to have this lead

11
Group 2
National Workshop on the Development of
Scenarios of Climate Variability and Extremes

• Strategy
• increased collaboration between govt
  departments
• regional expertise in regional centres
• driven by impacts and adaptation needs
• needs to be supported by research program e.g.
  uncertainties
• scenario facility needs to reside within govt
  as need is driven by policy, ensure long term
  stability
• model of collocation of govt and university
  researchers
• RCM output could contribute significantly, need
  more runs and greater access to data

12
Group 2
National Workshop on the Development of
Scenarios of Climate Variability and Extremes

• Uncertainty
• critical to communicate but very difficult to do
  define
• level needed depends on decisions that are being
  made
• uncertainty of when generally less important
  than if
• assessment of models limit access to less
  reliable model data
• place emphasis on probability distribution
  functions

13
Group 3

• Question 1.
• a)      Need to define extreme for specific
  sector or location.
• b)      More analysis needed of historical
  climate extremes and variability.
• c)      More information needed on event-driven
  historical extremes - e.g. hot spells in
  conjunction with warm low temperatures and high
  humidity.
• d)      Historical data from different agencies
  needed to be made available to all agencies and
  researchers.
• e)      Monitoring network needs strengthening.
  Also, remote sensing of climate elements could
  enhance surface observations.

14
Group 3

• a)      More stakeholder involvement needed to
  identify climate information gaps.
• b)      More information on disagreements between
  different climate analyses.
• c)      C-CIARN Regions and Sectors need to
  pro-actively get information from stakeholders
  and decision-makers on
• a.      Current vulnerability based on historical
  climate extremes.
• b.      Stakeholder knowledge of historical
  climate extremes
• c.      Climate-related datasets that could be
  made available widely
• d.      Needs for future climate information

15
Group 3

• Question 3.
• a)      Expanded and improved interface between
  Scenarios Facility and users.
• b)      Scenarios should consist of information
  useful to stakeholders as well as modellers.
• c)      Need to identify whether existing
  communities or sectors will experience extremes
  that have never been experienced before.
• d)      Scenarios need to
• a.      Provide plausible information to allow
  stakeholders to assess the change in risk of
  future impacts.
• b.      Focus on multiple variables.
• c.      Provide output focussed on user needs.

16
Group 3

• Ensure extreme scientists improve their
  interscientific communication
• Question 6.
• a)      Paleo climate is necessary to increase
  our knowledge of regional climate and its
  variability.
• b)      Paleo climate useful in increasing user
  confidence in GCM and RCM predictions.
• Question 7.
• Uncertainty information is needed.

17
Group 3

• a)      Since some impacts models are more
  sensitive than others to climate scenario
  impacts, the amount of certainty needs will
  vary with the user.
• b)      Need some measure of the attribution of
  uncertainty in a scenario.
• Question 8.
• a)      Downscaling is important, but deriving
  extremes that are useful for individual users
  from these outputs is problematic.
• b)      Information on tools and methodologies
  for downscaling need to be provided also, along
  with consultation on how to use downscaled data.

18
Group 3

• c)      Downscaled data should also include
  information from analogues and adjustment of
  historical variability and extremes.
• Question 13.
• a)      Scenarios facility should continue.
• b)      Scenarios Facility needs to immediately
  engage stakeholders to define its needs.
• c)      Scenarios Facility should provide access
  to paleoclimate information.
• d)      Scenarios Facility should provide
  qualitative assessment GCM suitability for
  different areas of Canada, or perhaps which ones
  should be used for ensemble forecasting.       
   
•  

19
Group 3

• e)      Scenarios Facility should post output
  from RCMs from Canada and other countries when
  available.
• f)        A comparison of RCMs is needed for
  Canada.
• e)      Scenarios Facility should post output
  from RCMs from Canada and other countries when
  available.
• g)      Information on strength and weaknesses of
  NCEP and other re-analyses.
• h)      A standard evaluation protocol is needed
  for all scenario tools.

20
Group 4

• Chair Gary Lines (EC Atlantic)
• Adam Fenech (EC)
• Rapporteur Andre St-Hilaire (INRS-ETE,
  Ouranos).Alain Bourque (Ouranos)Clair Goodess (UK
  Stardex)Van Nguyen (McGill, Brace Centre)Bill
  Girling (Manitoba Hydro)Sam Gameda (Ag Can
  Ottawa)Denise Neilsen (Ag Can BC)Trevor Murdoch
  (CICS-floater) 

21

• 1a. Most important aspects of climate variability
  and extreme events
• Point-scale information
• Micro-climate of island states, sub-RCM grid
• Extreme topography of BC (250-3500 mm)
• Extreme wind speeds building design/ fire risk
• Wind direction eg fire spread, air quality
• Extremes of precipitation (snow accumulation)
• Sequencing of wet/dry-days and spells
• Multi-season droughts
• Disaggregation to sub-daily eg. PMP
• Sea-level rise and saltwater intrusion

22

• 1b. Level of understanding of current
  vulnerabilities to climate change
• Limited data availability restricts knowledge
• Long-term, homogeneous records
• Sparse networks and getting sparser
• Ungauged watersheds
• Meta information on land-use, management
• Qualitative understanding in many sectors
• Operational rules for water plants
• Engineering design

23
1c. Relevance of long-term climate
variability Lack of confidence in model
variability Historic ranges useful Best to have
tool-box full of techniques
24

• 2. Which variables should be the focus
• Not really possible to short-list but if pushed
• Precipitation
• Temperature
• Wind speeds
• Local sea-level rise

25
3. Understanding of scenarios of climate
variability and extremes Format time-series,
return periods, thresholds Diverse users, diverse
needs Time-series can give all of the
above Ensembles help express uncertainty of
model Some indices can downscaled directly
26
6. Should we place more reliance on information
from past climates? Not at the expense of other
activities Merits more effort, provided in usable
format Past not neccesarily a useful analogue but
may Help understanding of large-scale processes
27
7. Should information on uncertainty be
provided? Yes. IPCC classes helpful eg Highly
likely Regional skill scores for various GCMs,
RCMs
28
8. Downscaling a viable option? Yes. No better
alternative for many applications Scenarios
(liability!!!) 1/9 Tools 8/9 Downscaling can
incorporate local knowledge Opportunity to build
capacity Support and training critical
29
13. National strategies to develop scenarios of
extremes and variability More investment in
monitoring/data networks National coordination,
research at province level Networks could mimic
EU groups Single scenarios centre providing data
training More capacity to develop/critique
tools in-house Broader network of local
trainers Partnership of Federal Government,
Provincial and stakeholders Concerted/sustained
links to international groups WMO endorsement of
tools/techniques
30
Group 5 Breakout Group Questions
31
Group 5

• 1.        Current methods for analysing extremes
  (useful feedbacks from international projects)
  Is it necessary to use in an integrated approach
  between dynamical and statistical downscaling
  tools to analyse extremes? 
• 2.      How do we deal with extremes in a
  non-stationary climate?
• 3.        Do extremes in a nordic climate require
  special attention?

32
Group 5

• 1.      What recommendations are suggested for a
  national strategy to develop scenarios of
  extremes and climate variability (e.g.,
  national/international collaboration,
  collaborative projects, tool development, bridges
  between historical-paleo data groups, modelling
  groups, statisticians, and IA users)?
• Impact res. Projects may reveal weaknesses in
  current climate modelling approaches as well as
  uncertainty in impact models. Climate model may
  be improved by feeding back these weaknesses to
  the modellers.

33
Group 5

• Practical users guide to GCM and RCM outputs.
  Some of it exists on CCIS site but more may be
  required.
•  
• A lot of piecemeal approach thus far. A more
  integrated approach is required with a national
  scope.
•  
• There should be a national body to establish a
  link between scientific community and IA.
• (CCAF and Adaptation groups have initiated such
  work (CCIARN)).

34
Group 5

• There is a submission on national centre of
  excellence on CC (research focus but CCIARN and
  Ouranos are involved)
• Regional efforts in networking. 
• Should there be an opportunity to organize a
  national body as funding will be restructured in
  the near future?
• Note that NSERC has a international fund. This
  is an opportunity. There is a new project on
  Ensembles

35
Group 5

• There are two types of users science but also
  easy transfer of data for IA. Must add non
  complex info sheets for IA. Not necessarily a
  CCIS job. Perhaps three levels or pre-digested
  info (Scenarios/IA Managers, etc.).

36
Group 5

• Impact res. Projects may reveal weaknesses in
  current climate modelling approaches as well as
  uncertainty in impact models. Climate model may
  be improved by feeding back these weaknesses to
  the modellers.
• Practical users guide to GCM and RCM outputs.
  Some of it exists on CCIS site but more may be
  required. A lot of piecemeal approach thus
  far. A more integrated approach
• is required with a national scope.

37
Group 5

• There should be a national body to establish a
  link between scientific community and IA.(CCAF
  and Adaptation groups have initiated such work
  (CCIARN)). There is a submission on national
  centre of excellence on CC (research focus but
  CCIARN and Ouranos are involved)Regional efforts
  in networking. Should there be an opportunity to
  organize a national body as funding will be
  restructured in the near future?Note that NSERC
  has a international fund. This is an
  opportunity. There is a new project on
  Ensembles

38
Summary - Group 5

• 1. Info availableLong term time series /
  homogeneity problem2. Vulnerability Known
  NO. Not enough

39
Group 6 Report
40
Q1. Most important aspects

• Need for more consultation with IA community on
  this suggest use of C-CIARN to do this
• Will likely be regionally variable not national
  in scope.
• Need to include Little Ice Age and MWP in long
  term database to complement obs data ensuring
  that we capture 2 key periods
• New data types mean that proprietary nature of
  data needs to be respected in development of
  scenarios (i.e. allow publication before
  constucting scenarios)
• CCIS may have a role to play in provision of
  current climate data (to complement current
  sources)

41
Variability Extreme Scenarios

• Scenarios are plausible pictures of future
  climate, its variability and extremes I.e.future
  weather scenarios.
• Need for data not only for IA but also to
  validate model performance on variability
  extremes
• XT scenarios will be spatially and temporally
  different from current mean scenarios.
• Selected variables depend on user need and hence
  require consultation
• Recognition of the limitations in generating
  scenarios of desired parameters and work towards
  finding new ways of using climate scenarios to
  address questions

42
Q8. Downscaling

• Need to make available a suite of several
  scenarios and different types of tools to develop
  scenarios to address the varying needs of users
• Also provides a means of ensuring that the nature
  of the uncertainties are well explained as they
  will be different than for current scenarios of
  means.
• The need to check on proper use of the tools
• Need mechanisms to ensure that tools developed
  under grants come into the public domain.

43
Recommendations National Strategy

• Should produce the best scenarios we can now and
  also continue research to refine them over time.
• ID needs of IA community C-CIARN CCIS to
  take leadership in process to prioritize needs
• CCIS C-CIARN should put together a proposal
  asap
• Develop better understanding of current
  vulnerabilities
• Improved paleo data will be needed to fill in
  gaps
• Also provides better understanding of scenario
  needs
• Both suite of extreme scenarios and tools for
  developing such scenarios should be provided to
  users
• Recommend renewal of funding for next phase of
  CCIS

44
Recommendations Continued

• In developing research program, should use
  results focused strategic approach (including
  competition within this framework where
  appropriate)
• Include role for CCIS advisory committee within
  this strategy
• Preference is for an integrated funding system
  that encourages investment from industry/others