Transcript Developing Decision-Support Modules for Climate Models: drawing

Developing Human System
Modules for Regional
Climate Models
Jessie Cherry
IARC/INE/ARSC@UAF
Peter Larsen
GSPP/LBNL@UC-Berkeley
Arctic System Modeling Workshop III,
University of Quebec, Montreal, July 2009
Presentation Outline
• “Old school” approach to the study of Human Dimensions
(HD) of Climate Change;
• Shortcomings with the “old school” approach;
• Some examples of HD modeling;
• Direct integration of HD into regional climate modeling
(i.e., “new school”);
• Implementation potential for particular sectors; and
• Benefits to developing an international HD working group
for the Arctic.
General Climate-related Modeling Approaches
Source: IPCC, 2007
Past Treatment of Human Dimensions
• Second (or third-order) modeling runs;
• Limited use of downscaled physical projections;
• Few examples of model comparison/testing
platforms and input/output sensitivity analyses;
• Weighted index, Delphi, and/or subjective
approaches are often employed; and
• Stakeholder feedback often occurs later on in the
development process, if at all.
Examples of Modeling HD: Alaska
Depr.
Matrix
Others
DRM
UAF GI
DCCED
Climate
Projections
APID
Denali
Infrastructure_DB_09_28_06.sas
NCAR
Import_Wx_UAF_NCAR_10_10_06.sas
DNR
Infrastructure Type
Agriculture
Airport
Bridges
Courts
Defense
Emergency Services
Energy
Grid
Harbor
Hospital
Law Enforcement
Misc. Building (govt)
Misc. Building (health)
Pipeline
Railroad
Roads
School
Sewer
Telecommunications
Telephone Line
Water
Replacement Cost
N/A
5,664,812
10,000
16,150,618
305,441
467,110
31,570
100,000
162,050
44,772,750
3,917,245
1,030,578
1,631,781
32,225,000
2,795,717
3,000,000
2,486,167
30,000,000
299,576
50,000
5,000,000
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
Units
N/A
Whole
Per foot
Whole
Whole
Whole
Whole
Per mile
Whole
Whole
Whole
Whole
Whole
Per mile
Per mile
Per mile
Whole
Whole
Whole
Per mile
Whole
Baseline Useful Life (years)
N/A
10
40
40
40
20
30
15
30
40
30
30
30
30
30
10
40
20
10
15
20
$
Graphs
Depreciator_10_10_06b.sas
Tables
HD Project:
“Estimating Risk to Alaska Public Infrastructure from
Climate Change” (Larsen et al, 2008)
Examples of Modeling HD: Alaska
Annual M aintenance Cost ($) or
D epreciation Rate ( lifespan)
$1
$2
$3
$4
$*
27º F
0 º c/32º F
Annual
Average
T emp.
Examples of Modeling HD: California
HD Project:
“Estimating Risk to California
Energy Infrastructure from
Climate Change” (Sathaye et al,
2009)
Examples of Modeling HD: California
Past/Current HD Modeling Concerns
The “old school” de-coupled HD approach:
• creates a strong disconnect between the
physical modeling and the climate impacts
communities;
• occasionally ignores stakeholder needs for
timely policy and decision making;
• often misses important feedbacks between
human agents and the climate system; and
• makes it difficult to compare and test
alternative modeling techniques.
Some Arctic Human Dimensions….
•
•
•
•
•
•
•
•
•
•
Resource Development
Hazard Response
Freshwater Supply
Renewable Energy (wind, hydro, geothermal)
Commercial and Sport Fishing/Hunting
Public and Private Infrastructure
Tourism
Subsistence Harvest
Marine Transport
Human Health
A “New School” HD Modeling Proposal….
1.
Develop Human System Modules directly into the
Arctic System Modeling platform;
2.
Make these modules portable and transparent
between different regional models;
3.
Encourage international collaboration;
4.
Focus on producing multiple socioeconomic impact
measures; and
5.
Facilitate model testing, scenario development,
stakeholder feedback, etc.
Some Thoughts on Decision
Support/Stakeholder Feedback
Turban defines decision support as "an interactive,
flexible, and adaptable computer-based information
system, especially developed for supporting the
solution of a non-structured management problem for
improved decision making. It utilizes data, provides
an easy-to-use interface, and allows for the decision
maker's own insights.” (Wikipedia, 2009)
• Decision support and ongoing stakeholder feedback
are very important factors to incorporate if the Arctic
System Model is going to be successful.
• What metrics will we use to gauge the overall
performance of this entire Arctic system?
Some Thoughts on Climate/HD Model
Interactions
• Need not occur at each model time step (e.g.,
hours vs. planning decades);
• One or two-way coupling may be appropriate
depending on the system (e.g., GHG
emissions); and
• Socioeconomic data collection and
dissemination will need to be substantially
improved;
•Quantifying coupled model uncertainty is very
important, but difficult to communicate.
Some Thoughts on Communicating
Uncertainty in HD Impacts…
Source: Larsen et al (2008)
Three different AOGCMs
Monte-carlo Simulation (varied inputs)
More Thoughts on Uncertainty in HD
Impact Estimates…
Harvard Economics Professor Martin Weitzman noted
in a seminal 2008 paper that fat-tailed structural
uncertainty about climate change, coupled with a lack
of information about high-temperature damages, can
potentially outweigh the influence of discounting in a
cost-benefit analysis framework.
What are the Challenges?
• Training and supporting interdisciplinary researchers
may be the biggest challenge;
• Pan-Arctic data collection and management is
another major challenge;
• Stakeholder engagement is time-consuming and
expensive;
• Some research disciplines are further along in the
evolution of systems modeling; and
• User-friendly “decision support tools” will need to be
developed in close collaboration with stakeholders.
Why include HD modules directly into
the ASM?
• There are (some) appropriate existing regional HD
models;
• We have the computing resources;
• We can attempt to minimize miscommunications
between the physical and social scientists across the
Arctic;
• It’s interesting and policy-relevant work at the
frontiers of research!!!
Benefits to Developing an International HD
Working Group
• The Arctic countries share many common HD because
of similar regional climate, geography, history, etc.;
• The Arctic countries also share common vulnerabilities;
• HD data is often disparate and difficult to find,
particularly at the Pan-Arctic level; and
• There is considerable experience within various Arctic
countries in the study of HD, but there is less
knowledge sharing occurring across countries.
Questions?
Reminder: HD Breakout
Session Later Today….
Additional Information

International Arctic Research Center at UAF: www.iarc.uaf.edu

Alaska Center for Climate Assessment and Policy (ACCAP):
www.uaf.edu/accap/

State of Alaska Climate Change Materials: www.climatechange.alaska.gov

E.O. Lawrence Berkeley National Laboratory: www.lbl.gov

Goldman School of Public Policy: www.gspp.berkeley.edu
Note: This presentation includes personal views of Peter Larsen.
Climate Change Planning
Walsh & Chapman:
PRISM downscaled
multi-model projections
of temperature and
precipitation for AK
under various scenarios
of Greenhouse Gas
emissions
Integrated Assessment
Definition: any model which combines
scientific and socio-economic aspects of
climate change primarily for the purpose
of assessing policy options for climate
change control (Kelly & Kolstad, 1998)
Integrated Assessment Modeling
McGuffie & Henderson-Sellers, 2005
Integrated Assessment Models
McGuffie & Henderson-Sellers, 2005
Example of
Human
System
Module
Goal is to be model
independent; work
with CCSM and
other models/
couplers
Cherry
Communicating uncertainty
New Scientific Methodology?
Funtowicz & Ravetz, in Ecological Economics, 1991
Arctic human dimensions
•
•
•
•
•
•
Oil and Gas Module (spill transport)
Rural Resilience (wind power potential)
Coastal Erosion (evolving coastline)
Freshwater (hydropower, water supply)
Marine Fisheries (Bering ecosystem)
Marine Transport (ice cover trajectories)
BSIERP
Lower Trophic Level
Ecosystem Model
Predation
Losses
Euphausiids
Detritus
14 component Model
NPZD-Benthos
Neocalanus
Pseudocalanus
Small
microzooplankton
Small
Phytoplankton
Nitrate
Large
microzooplankton
Large
Phytoplankton
Iron
Ammonium
Benthos
Benthic
Infauna
Benthic
Detritus
BSIERP Vertically Integrated models
BSIERP
Economic/ecological
model
NPZ-B-D
Lower trophic
level
ROMS
Physical
Oceanography
Climate scenarios
BEST
Nested models
FEAST Higher trophic
level model
Infrastructure
Impact of Climate Change on
Infrastructure study done for Alaska by
Peter Larsen and collaborators
Flow Chart of Model Processes
Depr.
Matrix
Others
DRM
UAF GI
DCCED
Climate
Projections
APID
Denali
Infrastructure_DB_09_28_06.sas
NCAR
Import_Wx_UAF_NCAR_10_10_06.sas
DNR
Infrastructure Type
Agriculture
Airport
Bridges
Courts
Defense
Emergency Services
Energy
Grid
Harbor
Hospital
Law Enforcement
Misc. Building (govt)
Misc. Building (health)
Pipeline
Railroad
Roads
School
Sewer
Telecommunications
Telephone Line
Water
Replacement Cost
N/A
5,664,812
10,000
16,150,618
305,441
467,110
31,570
100,000
162,050
44,772,750
3,917,245
1,030,578
1,631,781
32,225,000
2,795,717
3,000,000
2,486,167
30,000,000
299,576
50,000
5,000,000
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
Units
N/A
Whole
Per foot
Whole
Whole
Whole
Whole
Per mile
Whole
Whole
Whole
Whole
Whole
Per mile
Per mile
Per mile
Whole
Whole
Whole
Per mile
Whole
Tables
Baseline Useful Life (years)
N/A
10
40
40
40
20
30
15
30
40
30
30
30
30
30
10
40
20
10
15
20
$
Depreciator_10_10_06b.sas
Graphs
ISER Public Infrastructure
Study
Wind Farm Parameterization
for WRF
Adams & Keith
Modification of
the MYJ PBL
scheme
Similar work
being done
commercially
by 3TIER,
AER, others
MMS-WRF winds 1
MMS-WRF winds 2
MMS-WRF winds 3
MMS-WRF winds 4
AEA
Energy
Hydropower
Atlas, 2007
AEA
Ship track
Example of Climate-Related
Decision Support
https://rsgis.crrel.usace.army.mil/aedis/