Transcript Understanding and Applying the Science

Understanding and Applying
the Science
Anthony Chen
Climate Studies Group Mona
University of the West Indies
The Science
• Weather, Climate, Climate Variability and Climate Change
– What can we predict?
• How do we predict ?– Models
– Model Limitations
• Dealing with uncertainties in climate change
– Dealing with uncertainty in causes of climate change
– Dealing with limitations in models
– Dealing with climate change scenarios
• Sources of information
– Predictions of climate variability
– Scenarios of climate change
Weather and Climate
• Weather
– day by day occurrence
• Climate
– average over season or
longer
Climate Variability – regular, irregular, long term
Climate Change - trendy
Temperature changes since the
industrial revolution~ 1750
Trend – Climate Change
What can we predict?
• Weather can be predicted but not beyond
10 days:
– Butterfly Effect (small changes in initial
conditions can cause large changes farafield)
• Climate Variability can be predicted
• Not climate change prediction but climate
change scenarios
Rainfall Amount
Weather
Observed Precipitation vs Predicted Precipitation
14
Observed
12
Predicted
10
8
mm
Obs Precip
Pred Precip
6
4
2
0
1
2
3
4
5
6
Day
Days
7
8
9
10
11
Climate
Observed Precip vs Predicted Precip
14
12
10
Obs Precip
Pred Precip
Avg Obs
Avg Predict
mm
8
6
4
2
Days
39
37
35
33
31
29
27
25
23
21
19
17
15
13
11
9
7
5
3
1
0
Why is short term climate
(climate variability) predictable?
• Regularity of behaviour of energy input
– Sun
• Slow variation of conditions influencing
climate over the season
–
–
–
–
Sea surface temperature
Land vegetation
Soil Moisture
Ice cover
The Science
• Weather, Climate, Climate Variability and Climate Change
– What can we predict?
• How do we predict ?– Models
– Model Limitations
• Dealing with uncertainties in climate change
– Dealing with uncertainty in causes climate change
– Dealing with limitations in models
– Dealing with climate change scenarios
• Sources of information
– Predictions of climate variability
– Scenarios of climate change
Models
• Dynamic
– General Circulation Models (GCM)
• Atmosphere-Ocean Coupled GCM (AOGCM)
– Regional Models (RCM)
• Statistical Models
Science Used in Dynamic Models
• Meteorology
–
–
–
–
–
Application of Physics to the Atmosphere
Equations of motion
Thermodynamics
Radiation
Etc.
• Climate Change Science
– Application of all sciences to the land, sea and atmosphere
– Chemistry of the atmosphere
– Atmosphere – Land interaction
• Effect of Forests, trees, soil moisture on the atmosphere
– Atmosphere – Ocean interaction
• Winds
• Fluid Dynamics
• Thermodynamics
Equations and Processes in Dynamic
Computer Models
Solves for/calculates and steps forward in time
• Equations of motion
• First law of thermodynamics
• Physics of water vapor and clouds
•Chemical processes in atmosphere
• Land - atmosphere interactions
Biological processes
• Land - ocean interactions
General Circulation Model Process
• Equations of motion, First Law of
Thermodynamics, etc., can be solved by
numerical analysis
• Solved for grid boxes around the globe
– 300 to 500 km x 300 to 500 km
• Solved at various height levels
• Average value for various parameters
(pressure, wind, humidity) solved for each
box, each height level
Parameterization
• Some processes are not solved by equations, but are
governed by the values of designated parameters
• Calculation over a grid box gives an average for the grid
box
– Some processes are too small to be resolved
• Representation of clouds is a particular problem
– Average condition over a grid box may not be convective
(necessary condition for cloud formation), but in reality some
clouds form since part of the box may be convective.
– One solution is to relate cloud formation to the humidity in the
atmosphere (parameterization)
– Leads to uncertainty in rainfall amounts
Large Grids Required for GCMs
• Vary from model to model
• Capacity of computer would be overloaded
if smaller grids were calculated over the
globe
– Exception is a Japanese 20 x 20 km GCM
Regional Climate Models (RCM)
• Dynamic climate models run over smaller
region with smaller grids
– 50 x 50 km or 25 x 25 km, for example
• Uses the GCM as boundary conditions to
drive the model
• Give more details
• Process of getting fine details is called
downscaling
GCM, RCM Downscaled and Observed Precip over
England
GCM
RCM
RCM
Observations
Statistical Models
• Determine statistical relationships
between:
– what we want to predict or forecast
• e.g. temperature and rainfall
– large scale atmospheric parameters
• e.g., pressure, winds, humidity
• Can be used to downscale values of
temperature and rainfall at a station in
terms of values of large scale parameters
obtained from GCMs
Uncertainty, especially
Precipitation, in Dynamic Models
• Cannot replicate exactly all processes
• Cannot simulate climate at every point
– Grid boxes
• Butterfly Effect
– Initial conditions
• Structure of models differ
Uncertainty in Statistical Models
• Uncertainty that the regression equations
remain the same in the future climate.
– However no surprizes are expected since the
major driving forces are well understood, even
if we cannot completely model them
• Climate depends on too many changes in
parameters, not all captured by regression
equations
• Butterfly effect
The Science
• Weather, Climate, Climate Variability and Climate Change
– What can we predict?
• How do we predict ?– Models
– Model Limitations
– Dealing with limitations
• Dealing with uncertainties in climate change
– Dealing with uncertainty in causes climate change
– Dealing with limitations in models
– Dealing with climate change scenarios
• Sources of information
– Predictions of climate variability
– Scenarios of climate change
How does climate change occur?
I. Natural variations in solar radiation
affect temperature changes
How does climate change occur?
II. Cooling
• Natural blocking of
the sun’s radiation
– volcanic ash
– Aerosols (dust)
• Man made aerosols
blocking out the sun’s
radiation
How does climate change occur?
III. Natural greenhouse gases (CO2, water
vapour, etc) in the atmosphere traps heat
If there were no natural atmospheric
greenhouse effect the temperature of the
earth would be 30º C colder.
IV. Since the industrial Revolution,
Man-made Green House Gases have
been added:
• CO2 from
fossil fuel
power plants
• N2O for
automobile
exhaust
Water Vapour from
airplane exhaust
(very effective
greenhouse gas)
Methane,
before
and since
Chlorofluro-carbons CFC’s
How does climate change occur?
V. Land use changes affect temperature
and hydrological cycle
Founded 1988 by the World Meteorological Organization
(WMO) and the United Nations Environment Programme
(UNEP)
Working Group I assesses the scientific aspects of climate
change.
Working Group II assesses impacts, vulnerability and
adaptation
Working Group III assesses options for mitigating climate
change.
Fourth Assessment (AR4) 2007
Attribution: Comparisons of climate models and
observations of global mean temperature
Agree only when both natural and anthropogenic
forcing are included IPCC 3rd assessment
Attributing climate change regionally, 4th Assessment,
to anthropogenic and natural causes
Black line – observation 1900-2000
Red – all forcing (natural and man made)
Blue – natural forcing
Similarly for the rest of the world
Bands give the range of model values
IPCC Statement
• From new estimates of the combined
anthropogenic forcing due to greenhouse
gases, aerosols and land surface
changes, it is extremely likely (> 95%
probability) that human activities have
exerted a substantial net warming
influence on climate since 1750.
Dealing with limitations in the models
• Use average of as many models as possible
• Use different initial conditions with same model
– Ensemble mean
• Use different models with same initial conditions
– 21 models used for IPCC 4th assessment
– Get mean
• Use other supporting material
Need supporting evidence to
complement model results
– Strong Physical basis or explanation
• Is there a physical basis for temperature
increase
– Science of Global Warming is sound
• Is there a physical basis for precipitation
increase or decrease
– Agreement of as many of the following
• Observed historical trends in climate
• General Circulation Models of Climate
• Downscaling of global models
– Regional climate models (dynamic)
– Statistical downscaling
Expressing Uncertainties
• Probability statement, e.g., in terms of
probability distribution functions
• Statistical distributions, more easily
undestood
• Likelihood statements
– Very likely (> 90% probability)
– Likely (> 66% probability)
– Extremely likely (> 95% probability)
Table 11.1, Chapter 11, IPCC Working Group 1, 4th
Assessment: Changes (2080s vs. 1980s) in Caribbean
temperature and precipitation from a set of 21 global
models for the A1B scenario. (All regions of world given)
Methodology used for 4th IPCC Assessment
Averaged
Average of GCM
results, not
downscaled
Outputs from several GCMs
Output from one GCM
Regional
Model
based on one GCM
SDSM
Output from one GCM
RCM Downscaled
Scenario
SDSM Downscaled
Scenario
based on one GCM
What we would like to see for 5th Assessment
Averaged
Downscale
scenario based on
several RCM’s
RCM Outputs using several
GCM’s
Averaged
SDSM outputs using
several GCMs
Downscale
scenario based on
several SDSMs
Climate Change Scenarios
• Besides inherent uncertainties in models
there are uncertainties in the changes in
the main driving forces changing climate
• Not possible to accurately predict the
changes in the main driving forces in
climate change
• Instead use scenarios of plausible futures
– Special Report Emission Scenario (SRES)
Special Report on Emission Scenarios
(SRES) schematic and storyline summary
Some SRES
Scenarios for 5th Assessment
% Change in emission relative to 1990
IPCC
emission
scenarios
+50%
Developing countries
0%
(1990)
-50%
World as a
whole
Developed countries
2020
2050
Which scenarios to use?
• It is up to you
– Justify use
• A1B is in-between high
and low emission and
is used by IPCC for
illustration
• Always quote
emissison scenario
used
Final Word on Modelling & Scenarios
• Recall the state of the art
– Exact Predictions not possible
• What is important in using GCMs
– The sign of the change (increasing or decreasing)
– The agreement among models and scenarios about
increasing or decreasing trends
• Use supporting evidence if possible
– Observations
– RCMs
– Statistical downscaling
• Use statistical distribution (mean, median, min, max,
25%, 50% and 75% quartile) if multi-model values are
available
• Use a range of scenarios if necessary, or use a scenario
between high and low, e.g., A1B
More Information
• Your Regional Climate Centres
– May be able to give downscaled results (RCM or
Statistical)
• Your National Communication to UNFCCC
• Climate Variability
– International Research Institute for Climate and
Society (http://portal.iri.columbia.edu)
• Climate forecast
• El Nino forecast
• Climate Change
– IPCC Working Group 1 Report
• Climate Change 2007, Chapter 11, Regional
Projections (http://www.ipcc.ch/about/workinggroup1.htm)
• GCM results mainly
– Climate Research Unit, CRU (www.cru.uea.ac.uk/ )
Another Climate Change Scenario Tool:
MAGICC/SCENGEN
www.cgd.ucar.edu/cas/wigley/magicc/
Greenhouse gas emissions
MAGICC
Global mean temperature
change
SCENGEN
Regional Scenarios of Climate
Change
GCM patterns
Climatologies
Applications - Need to know:
• Impacts of Climate Change based on scenario
– Example of Jamaica
• Vulnerability to Climate Change
– Example of Jamaica
• Adaptation to Climate Change
– Example of Jamaica
• UNDP VRA
• Tools for vulnerability and adaptation
assessment
• More information
Scenarios Available
•
•
•
•
•
Temperature
Precipitation
Evapotranspiration
Run-off
Soil moisture
Best Estimates based on RCM,
SDSM and GCMs – Jamaica’s 2nd
National Communication
Temperature
2050s
2080s
1.05 degree C
2.45 degrees C
Changes from 1961-1990 average, based on
Average of A2 and B2 scenarios
2050s
2080s
% change from average of 19611990 values
Precipitation:
Region 1
Region 2
Region 3
0
-10
-10
-30
-30
-20
No estimate
No estimate
-40
-20
Region 5
Region 6
-10
-10
-40
-30
Region 7
-10
-30
Region 4:
Portland
St. Thomas
5
1
6
7
2
3
4
Wet and Dry Spells
• Wet day %: The percentage of days that exceed
a wet-day threshold limit of 0.3 mm
• Mean wet spell length: The average length of
continuous wet-days with amounts greater than
or equal to the wet-day threshold
• Mean dry spell length: The average length of
continuous dry-days with amounts less than the
wet-day threshold
• Obtained from SDSM
– Wet day % and Mean wet spell length decreases
– Mean dry spell length increases
2050s
2080s
Wet-day%
-24
-44
Wet spell length
-7
-10
Dry spell length
32
80
Wet-day%
-2
-7
Wet spell length
-3
-6
Dry spell length
1
4
% changes:
Region 5:
Region 3:
Best Estimate other parameters
• By the end of the century sea levels are also
expected to rise by 0.23 to 0.47 metres under an
A1B scenario, but the models exclude future
rapid dynamical changes in ice flow.
– A recent study suggests that the rate of rise may
actually double (Science Daily, Feb. 12, 2008).
• Evaporation is also projected to increase by
approximately 0.3 mm/day over the sea. As
noted before, the changes over land may be
less.
• One model has projected more intense
hurricanes in the Atlantic.
Impacts on Jamaica
•
•
•
•
•
•
•
•
Sea level rise
Spread of diseases like dengue
Bleaching and death of coral reefs
Possible more intense hurricanes
Water resource shortfall
Agricultural drought
Reduction in Tourist arrival
Depletion of coastal resources
– Death, Migration of fishes to cooler waters
• Endangered human settlement
• Possible extinction of some species in biodiversity
Danger of sea level rise
Dengue
• Previous epidemiology research showed
dengue transmission linked to higher
temperatures
– 2ºC rise leads to 3-fold increase in
transmission
• Under average of A2 and B2 scenario
temperature in Jamaica will rise 2.45ºC
– Expect 3-fold increase in transmission
Applications
• Impacts of Climate Change based on scenario
– Example of Jamaica
• Vulnerability to Climate Change
– Example of Jamaica
• Adaptation to Climate Change
– Example of Jamaica
• UNDP VRA
• Tools for vulnerability and adaptation
assessment
• More information
Vulnerability
to Dengue
Some Vulnerability Indicators for
Health
Based on indicators identified in the literature:
• Immunity
• Knowledge of symptoms and vectors of disease.
• Use of protective measures.
• Measures of resilience and stress – education,
employment, income, female household headship, room
densities, coping strategies, integration into the
community.
• Source of water, water storage.
• Distance from the nearest health facility
Characteristics of the Most and Least vulnerable groups
Characteristics
Group 5
(%)
Group 1
(%)
1.Female Household headship
94
40
2.Unskilled
88
13
3.Primary education or none.
77
7
4.Minimum wage or less
57
0
5.Not Coping
88
0
6.Water storage in drum
88
20
7. Clinic distance
64
13
8. No protection
88
53
9.No knowledge of dengue
transmission
94
7
10. No social integration
53
47
11. No knowledge of dengue
symptoms
88
13
12. No personal acceptance for
dengue control
77
7
Community with highest proportion
Johns Hall -53%
Granville/Pitfour – 73%
Applications
• Impacts of Climate Change based on scenario
– Example of Jamaica
• Vulnerability to Climate Change
– Example of Jamaica
• Adaptation to Climate Change
– Example of Jamaica
• UNDP VRA
• Tools for vulnerability and adaptation
assessment
• More information
Adaptation Strategies Matrix For Dengue
Table 3. Adaptation Strategies Matrix (Continued)
Measures
Cost
Effectiveness
Social
Acceptability
Friendly for
Environment
Neighbor
Effects
Technical
Challenges
and SocioEconomic
Change
Score
Long Term
1. Surveillance for vector or
larval/pupal control and
environmental sanitation
H
M
M
M
L
L
16
2. Community education and
involvement
M
H
H
H
H
M
26
3. Chemical control
H
M
M
L
M
L
16
4. Biological control
H
H
M
H
M
M
20
5. Adult Control
- Physical-mesh windows
- Personal protection
M
M
H
M
H
M
H
M
H
M
H
H
24
16
6. Use of physical control-lowcost secure drums
H
H
M
H
H
H
20
7. Granting security of tenure to
squatters
H
H
H
M
H
H
20
8. Early warning system
M
H
H
H
H
H
24
Columns 2 through 7 indicate assessment criteria. Column 8 gives a composite score based on the
ranking in columns 2–7. Assessments are on the basis of high, medium, and low. In compiling the
composite score, High is given a score of 5, medium a score of 3 and low a score of 1, except for
columns 2 and 7, where the scoring allocation is reversed. The maximum possible score is 30.
Expert Judgment
• Expert judgment is an approach for soliciting informed
opinions from individuals with particular expertise.
• This approach is used to obtain a rapid assessment of
the state of knowledge about a particular aspect of
climate change.
• It is frequently used in a panel format, aggregating
opinions to cover a broad range of issues regarding a
topic.
• Expert judgment is frequently used to produce position
papers on issues requiring policy responses and is
integral to most other decision-making tools.
• This approach is most useful either in conjunction with a
full research study or when there is insufficient time to
undertake a full study.
Applications
• Impacts of Climate Change based on scenario
– Example of Jamaica
• Vulnerability to Climate Change
– Example of Jamaica
• Adaptation to Climate Change
– Example of Jamaica
• UNDP VRA
• Tools for vulnerability and adaptation
assessment
• More information
UNDP Vulnerability Reduction
Assessment (VRA)
• Use Expert Judgement here
• Local experts know what the community
needs to know
– Effects of flood and drought, storm surge,
increased temperatures, etc
• Get climate change scenarios from sources
above
– Be aware of the limitations
How does the VRA Work?
•The VRA is composed of four indicators, based on the
UNDP Adaptation Policy Framework approach.
•These four indicators become four questions – tailored to
the community and posed in community meetings before,
during and after project implementation.
•VRA meetings yield quantitative and qualitative data:
• Useful in aggregating and assessing programmatic impact
• Useful in guiding project design and management
• Useful in capturing lessons learned
67
The VRA in theory
APF Step
VRA Indicator
VRA Question
In these examples, we consider the
case of a community facing
increasing drought risks
Assessing current
vulnerability
1. Vulnerability of
livelihood/welfare to existing
climate change and/or climate
variability.
Example: What happens when
there is drought? How does this
affect you and your community?
Assessing future
climate risks
2. Vulnerability of
livelihood/welfare to
developing climate change
risks.
Example: What would happen if
drought was twice as frequent?
How would this affect you and your
community?
Formulating an
adaptation
strategy
3. Magnitude of barriers
(institutional, policy,
technological, financial, etc)
barriers to adaptation.
Example: What stands in the way
of adapting to increasing drought?
What means do you or your
community have to manage events
occurring more frequently?
Continuing the
adaptation
process
4. Ability and willingness of
the community to sustain the
project intervention
Example: Rate your confidence that
the (project activity) will continue
after the project period.
68
Applications
• Impacts of Climate Change based on scenario
– Example of Jamaica
• Vulnerability to Climate Change
– Example of Jamaica
• Adaptation to Climate Change
– Example of Jamaica
• UNDP VRA
• Tools for vulnerability and adaptation
assessment
• More information
Vulnerability Methods
Compendium on methods and
tools to evaluate impacts of,
and
vulnerability and adaptation to,
climate change
(Final, 1/2/2005)
http://ncsp.undp.org/report_detail.cfm?Projectid=151
Livelihood Sensitivity Exercise
• Livelihood sensitivity mapping exercise is a means of
integrating existing knowledge of climate vulnerability
with livelihood analysis.
• It commonly involves stakeholder participation.
• The exercise involves developing a matrix with three
blocks of rows — beginning with ecosystem services
(e.g., soil moisture), then livelihood activities (such as
crop production) and finally a synthesis based on
livelihoods themselves.
• Climatic stresses (e.g., drought) are listed as columns.
• Users then fill in the cells — rating the sensitivity of
ecosystem services, activities and livelihoods to a
range of hazards and stresses.
•
Drought
Soil Moisture Negative
Crop
Production
Negative
Livelihoods
affected
Farmer
Shop Keeper
House holder
Flood Temperature
Increases
Vulnerability Indices
• Vulnerability is defined by the IPCC as the
combination of sensitivity to climatic
variations, the probability of adverse
climate change, and adaptive capacity.
• For each of these components of
vulnerability, formal indices can be
constructed and combined.
– c/o dengue vulnerability exercise
Adaptation Methods
Adaptation Decision Matrix (ADM)
• Multicriteria assessment techniques to evaluate the relative
effectiveness and costs of adaptation options.
• Users are asked to specify criteria that will be used to
evaluate options and weight the criteria.
• Users are asked to give a score (e.g., 0 to 5) on how well
each criterion is met under a particular scenario for each
option.
• The scoring can be based on detailed analysis or expert
judgment.
• Scores can be multiplied by weights and summed up to
estimate which options best meet the criteria.
• Detailed research and analysis or expert judgement are
needed to provide a basis for the evaluation; otherwise the
scoring may be mainly subjective.
• c/o dengue adaptation exercise
More Information
• Regional/National Centres for Risk
Reduction
• Country’s National Communication to
UNFCCC
• National adaptation programmes of action
(NAPAs) submitted to UNFCCC
• IPCC Working Group 2
(http://www.ipcc.ch/about/workinggroup2.htm)