View
70
Download
0
Category
Tags:
Preview:
Citation preview
Climate change, Implications for
Hydrologic Extremes, and What can
we do about it? Tom Hopson
climate advisor
CSU Feed the Future Innovation Lab
“I have a very strong feeling that science exists to serve human welfare. It’s wonderful to have the opportunity given us by society to do basic research, but in return, we have a very important moral responsibility to apply that research to benefiting humanity.” Dr. Walter Orr Roberts (NCAR founder)
GCRS1950
1960
1970
1980
1990
2000
2010
GFDL
ANMRC
BMRC
NCAR 1
NCAR I1
NCAR II1
CCM 0-A
CCM 0-B
ECMWF
CCM 1
CCM 2
CCM 3
CSM
CCSM 3
CCSM 4
CESM
PCM
MPI
ECHAM3
ECHAM4
ECHAM5
UCLA
GISS
CM2.0
CM2.1
GISS II
GR
E
E2
MRI
CGCM1
CGCM2
CGCM3
CSIRO
UKMO
HadCM
HadCM2
HadCM3
HadGEM
HadGEM2
CGCM
CGCM2
CanESM2
CGCM3
2
3
3.6
CCSR
MIROC
MIROC5
BCC
CSM1.1
CNRM
CM5.1
INMCM
FGOALS
ACCESS
FGCM
Outline
• Climate change overview from IPCC AR5 and
Implications for hydrologic extremes
– A word of caution about regional uncertainty
• What can we do about it?
– Meningitis forecasting for Africa
– Case study: Bangladesh flood forecasting
Large-scale constraints on Mean
and Extreme Precipitation
What have we observed – “likely” (66-100%) that
human influence has affected:
•global surface specific humidity
•global zonal mean terrestrial precipitation
•Arctic precipitation
•global-scale intensification of heavy precipitation
Large-scale constraints on Mean
Precipitation
What we expect – mean precipitation and moisture
•distribution of relative humidity remain roughly
constant and thus water vapor mixing ratio to
increase (6-10%/degC)
•mean precipitation increasing at 1–3% / degC
(tropospheric net radiative cooling rate)
•“Wet-get-wetter”, “dry-get-drier”, but large regional
variability and margins unclear due to regional
circulation shifts
Large-scale constraints on
Extreme Precipitation
What we expect – extreme precipitation
•individual storms increase 6-10% /degC (scales
with available moisture)
•high confidence much greater than mean
precipitation, but varies with time-scale, location,
season
South Asian Monsoon
Precip increases in:
•average
•variance
•5-day seasonal max
•duration
%
Yr 2100
AR5
The Precip
Challenge
80 yr. Temp. Rise
CMIP
80 yr.
Precipitation
Trend ?
Covey et al. 2003
Climate Prediction:
Sources of Uncertainty Forcing (or Scenario Uncertainty)
– GHG emission scenarios (e.g., B1, A1B, A2, RCPs)
ozone, sulfate aerosols, land use, black carbon …
Response (or Model Uncertainty)
– Model sensitivity (different physics, parameterizations, resolution …) – 20+ different climate models in both IPCC AR4 and AR5
Internal (Natural) Variability
– Atmosphere and Ocean – Coupled atmosphere-ocean interactions – Multiple simulations
Jim Hurrell
(now using “Representative
Concentration Pathways”,
or RCPs, in AR5)
Projected Global SAT (IPCC AR4)
(Hawkins and Sutton 2009)
Natural Variability
Can be large & slow
(multiple decades)
} }
}
Forcing (or Scenario) Uncertainty
Response (or Model) Uncertainty
20th Century Simulations
Jim Hurrell
NCAR
Time
Forced Climate Change
The Challenge: Assessing Climate Change in the Presence
of Unforced Multi-decadal Variability
10 years
Jim Hurrell
Natural Variability over the Atlantic
All significant modes of natural variability
Variations and significant impacts from seasonal to multi-decadal time scales
Major Phenomena
• North Atlantic Oscillation (NAO)
• Atlantic Meridional Overturning Circulation (AMOC)
• Tropical Atlantic Variability
(Hurrell et al. 2006)
Jim Hurrell
NCAR
Unforced Multi-Decadal Variability
North Atlantic
SST
Jim Hurrell
1850 control
40 CESM Integrations
20th Century 21st Century
(2000-2060)
A1B GHG
40
different atmospheric initial states
same ocean, sea ice, land initial states
The NCAR Large Ensemble Project:
Year 541
Uncertainty Arising from Natural Variability
http://www.cesm.ucar.edu/working_groups/Climate/
i.e., spread is not
predictable
Jim Hurrell
NCAR
Projected European Precipitation
Uncertainty due to Natural Variability
DJF 2010-2060
Wettest
Run 3
Driest
Run 9
Forced }
}
Jim Hurrell
Outline
• Climate change overview from IPCC AR5 and
Implications for hydrologic extremes
– A word of caution about regional uncertainty
• What can we do about it?
– Meningitis forecasting for Africa
– Case study: Bangladesh flood forecasting
Humidity and Meningitis
in the Sahel
Abudulai Adams-Forgor, Anaïs
Columbini, Vanja Dukic, Mary Hayden,
Abraham Hodgson, Thomas Hopson,
Benjamin Lamptey, Jeff Lazo, Kristen
McCormack, Roberto Mera, Raj Pandya,
Jennie Rice, Fred Semazzi, Madeleine
Thomson, Sylwia Trazka, Tom Warner,
Tom Yoksas
NC STATE UNIVERSITY
19
Goal: improve WHO efficiency
of vaccine dissemination due
to humidity-reduced risk link
Context: Managing Meningitis in
the Sahel
• Meningitis is endemic in the Sahel in countries with a distinct wet-dry season – e.g. 1996-1997 epidemic resulted in 250,000 cases, 25,000
fatalities
• Infectious disease due to bacterium – Neisseria meningitidis – Dominant serogroup A
– 5-10% fatality rate
– 10-20% of survivors have permanent impacts, e.g. hearing loss, brain damage, leaning disabilities
• A reactive (polysaccharide) vaccine strategy had currently been used to manage epidemics – Doesn’t prevent transmission of the disease by the
individual vaccinated => no “herd immunity” – Only lasts two-to-three years – No immune response in children under two – Limited suppy => effective allocation strategies
Using Thorpex Tigge 1- to 2-week forecasts from 7 centers: JMA, KMA, ECMWF
NCEP, CMA, CMC, MeteoFrance, UK Met
Recent findings on impact of dust:
C. Perez et al. 2013 for Niger
A. F. Abdusalam et al. 2013 for N. Nigeria
Humidity
Dust
Temperature
Wind
Time Indoors
Irritation of
the mucosal
membrane
Greater
Interpersonal
Contact
Greater
transmission
opportunity
Impaired
mucosal
defenses Upper
respiratory
infection
New Nm.
strains
Migrating
individuals
Biologic
change
Environmental
Factors
Social Impacts
Biologic Impacts
Physiological Impacts
Meningitis:
Inflammation of the
protective membranes
covering the brain and
spinal cord. In Africa,
often caused by bacteria
Person-to-person
transmission through
respiratory and throat
secretions – between 10-
25% of population may
carry bacteria at any time;
higher during epidemics
Effects:
can be life-threatening,
but also lead to brain
damage and deafness.
Tens of thousands of
people
Possible causes:
Relative humidity improves prediction
• Knowing the RH two weeks ago improves accuracy in predicting an epidemic by ~25%1
• Coupled with a two week forecast, this indicates an improved ability to anticipate a roll-off in epidemic 4 weeks in advance
1It turns out other variables (air temp, winds, NE winds) also help,
but less than relative humidity
Without RH
Using RH
Archive Centre
Current Data Provider
NCAR NCEP
CMC
UKMO
ECMWF MeteoFrance
JMA KMA
CMA
BoM CPTEC
IDD/LDM
HTTP
FTP
Unidata IDD/LDM
Internet Data Distribution / Local Data Manager
Commodity internet application to send and receive data
NCDC
Unique Datasets/Software Created
Thorpex-Tigge
Countries provided forecasts
6/ 18/ 13 7:52 AMmap of africa - Google Maps
Page 1 of 1https:/ / maps.google.com/ maps?q= map+ of+ africa&ie= UTF8&hq= &hne…93,73.828125&t= m&z= 4&vpsrc= 6&ei= xGXAUZbACeTuwQGYioCQDA&pw= 2
Address
National Center for Atmospheric Research (NCAR)
Relative humidity
across the Meningitis
Belt forecast 2 week
in advance …
… converted to
probability of an
epidemic alert
occurring
across the Belt
4 weeks in advance
Within dashed region,
risk has doubled
(above background levels)
Relative Humidity linkage
Locations where end-of-season
forecasts would have saved vaccine
Using “Perfect” Forecasts
- 18 epidemics identified
Using Climatological Information
-- 3 epidemics identified
• WHO districts where end-of-season forecasts predict no need for follow-up
vaccination campaign due to changing weather conditions
• ~2.6 million doses of vaccine used elsewhere more effectively
• Expense ~ US$ 1 million
Outline
• Climate change overview from IPCC AR5 and
Implications for hydrologic extremes
• A word of caution about regional uncertainty
• What can we do about it?
• Meningitis forecasting for Africa
• Case study: Bangladesh flood forecasting
Operational Flood Forecasting for Bangladesh:
Tom Hopson, RAL-NCAR
Peter Webster, Georgia Tech
A. R. Subbiah and R. Selvaraju, Asian Disaster
Preparedness Centre
Climate Forecast Applications for Bangladesh (CFAB):
USAID/CARE/ECMWF/NASA/NOAA
Bangladesh Stakeholders: Bangladesh Meteorological Department, Flood Forecasting and Warning Center,
Bangladesh Water Development Board, Department of Agriculture Extension, Disaster Management Bureau,
Institute of Water Modeling, Center for Environmental and Geographic Information Services, CARE-
Bangladesh
Context: CFAB Project PI Peter Webster, Georgia Tech (pjw@eas.gatech.edu)
Partners: USAID, CARE, ECMWF, Bangladesh’s Meteorology Dept and Flood
Forecasting Warning Centre (FFWC), NASA-TRMM, NOAA-CMORPH
Purpose: provide flood forecasts of the Ganges and Brahmaputra rivers for
Bangladesh, operational 2003-ongoing
(World Food Program)
Damaging Floods:
large peak or extended duration
Affect agriculture: early floods in May, late floods in September
Recent severe flooding: 1974, 1987, 1988, 1997, 1998, 2000, 2004, and 2007
1998: 60% of country inundated for 3 months, 1000 killed, 40 million homeless, 10-20% total food production
2004: Brahmaputra floods killed 500 people, displaced 30 million, 40% of capitol city Dhaka under water
2007: Brahmaputra floods displaced over 20 million
River Flooding
Forecast Trigger: ECMWF forecast files
Updated TRMM-
CMORPH-CPC
precipitation estimates
Updated distributed
model parameters
Updated outlet
discharge estimates
Above-critical-level
forecast probabilities
transferred to Bangladesh
Lumped Model Hindcast/Forecast
Discharge Generation
Distributed Model Hindcast/Forecast
Discharge Generation
Multi-Model Hindcast/Forecast Discharge Generation
Discharge Forecast PDF Generation
Calibrate model
Statistically correct
downscaled forecasts
Generate forecasts Generate hindcasts Generate forecasts Generate hindcasts
Update soil moisture
states and in-stream flows
Generate hindcasts
Calibrate AR error model
Calibrate multi-model
Generate forecasts Generate hindcasts
Generate model error PDF
Convolute multi-model forecast
PDF with model error PDF
E O
F
M
Q P
B D
L
F
C
Generate forecasts
Five Pilot Sites chosen in
2006 consultation
workshops based on
biophysical, social criteria:
Rajpur Union
-- 16 sq km
-- 16,000 pop.
Uria Union
-- 23 sq km
-- 14,000 pop.
Kaijuri Union
-- 45 sq km
-- 53,000 pop.
Gazirtek Union
-- 32 sq km
-- 23,000 pop.
Bhekra Union
-- 11 sq km
-- 9,000 pop.
Average Damage (Tk.) per Household in Pilot Union
7,255
28,745
60,99364,000
4058
0
10,000
20,000
30,000
40,000
50,000
60,000
70,000
Uria Gazirtek Kaijuri Rajpur Bekra
Union
Av
era
ge
Da
ma
ge
(T
k)
pe
r
Ho
us
eh
old
2007 Brahmaputra Ensemble Forecasts and
Danger Level Probabilities
7-10 day Ensemble Forecasts 7-10 day Danger Levels
7 day 8 day
9 day 10 day
7 day 8 day
9 day 10 day
Community level decision
responses for 2007 flood
forecasts (Low lands)
“… on 25th July we started communicating the information to as many
people as possible … especially those people living in river islands
(“chars”)...”
“On the 28th and 29th, meetings were organized in villages near Rangpur
… they perceived that the river water level would fall, but our forecasts
showed a rising trend…[with] significant chance of overflow and
breaches [of weak] embankments ... We engaged … an evacuation plan
urgently”
“We communicated the forecast to another pilot union … on July 26th …
to mobilize resources for evacuation ... All the six villages in the union
were later flooded to a height of 4-6 feet on July 29th… about 35% of the
people in the union were evacuated in advance.”
“The communities in Rajpur Union … were able to use the forecast for …
mobilizing food, safe drinking water for a week to 10 days, protecting
their … rice seedlings, fishing nets, and … fish pods.”
Summary
• Climate change impacts on the hydrologic cycle
have already been observed, most strongly in the
Arctic
• Globally, mean precipitation is expected to
increase 1-3%/degC, while the strength of
extreme storms increase 6-10%/degC
• However, much local and regional variation so
use care in estimating impacts (use multiple
models and ensembles)
• However, ensemble (probabilistic) weather
forecasts continue to improve and can be used to
mitigate some of the impacts
Climate Prediction:
Sources of Uncertainty Forcing (or Scenario Uncertainty)
– GHG emission scenarios (e.g., B1, A1B, A2, RCPs)
ozone, sulfate aerosols, land use, black carbon …
Response (or Model Uncertainty)
– Model sensitivity (different physics, parameterizations, resolution …) – 20+ different climate models in both IPCC AR4 and AR5
Internal (Natural) Variability
– Atmosphere and Ocean – Coupled atmosphere-ocean interactions – Multiple simulations
Projected Global SAT (IPCC AR4)
(Hawkins and Sutton 2009)
Natural Variability
Can be large & slow
(multiple decades)
} }
}
Forcing (or Scenario) Uncertainty
Response (or Model) Uncertainty
20th Century Simulations
NCAR
Linear Trend of Surface Temperature
1901 – 2012 ( C over period)
Mixture of internal variability and forced climate change
IPCC (2013)
°
NCAR
Change in Winter Sea Level Pressure (1980-2009)
(hPa) Dec-Mar
Pressure Falls
Pressure Rises
Decadal Climate Variability
1900-2009
1900-2009
Natural Variability over the Atlantic
All significant modes of natural variability
Variations and significant impacts from seasonal to multi-decadal time scales
Major Phenomena
• North Atlantic Oscillation (NAO)
• Atlantic Meridional Overturning Circulation (AMOC)
• Tropical Atlantic Variability
(Hurrell et al. 2006)
NCAR
Unforced Multi-Decadal Variability
North Atlantic
SST
NCAR
Time
Forced Climate Change
The Challenge: Assessing Climate Change in the Presence
of Unforced Multi-decadal Variability
10 years
NCAR
The Community Earth System Model www.cesm.ucar.edu
60
Figures 1
2
3
4
5 6
Figure 1. Schematic of the different component models in CESM1. 7
8
(Hurrell et al. 2013)
Community Climate System Model (CCSM)
60
Figures 1
2
3
4
5 6
Figure 1. Schematic of the different component models in CESM1. 7
8
60
Figures 1
2
3
4
5 6
Figure 1. Schematic of the different component models in CESM1. 7
8
60
Figures 1
2
3
4
5 6
Figure 1. Schematic of the different component models in CESM1. 7
8
60
Figures 1
2
3
4
5 6
Figure 1. Schematic of the different component models in CESM1. 7
8
60
Figures 1
2
3
4
5 6
Figure 1. Schematic of the different component models in CESM1. 7
8
NCAR
The Community Earth System Model
• CESM is used to:
– Explore Earth’s climate history and
processes responsible for variability
and change
– Estimate future of environment for
policy formulation
Modeling the Earth System
www.cesm.ucar.edu
• Developed by NCAR NSF, DOE, Universities, National Laboratories
• Fully documented, frequently and freely
distributed, fully supported releases
• Capacity Building (e.g., tutorials and workshops)
• CESM: a set of different geophysical
component models that exchange
boundary data via a coupler
• Code base developed over 20+ yrs:
runs on multiple platforms, resolutions
and model configurations
The NCAR Large Ensemble Project:
Uncertainty Arising from Natural Variability
http://www.cesm.ucar.edu/working_groups/Climate/
40 member ensemble from 2000-2061
http://www.cesm.ucar.edu/working_groups/Climate/experiments/ccsm3.0/
Deser, Phillips, Bourdette, Teng (2012): Climate Dynamics
(also articles by Branstator, Teng, Meehl and others)
30+ member ensemble from 1920-2080
http://www.cesm.ucar.edu/experiments/cesm1.1/LE/
Kay et al. (2014): Bulletin of the American Meteorological Society
1850 control
40 CESM Integrations
20th Century 21st Century
(2000-2060)
A1B GHG
40
different atmospheric initial states
same ocean, sea ice, land initial states
The NCAR Large Ensemble Project:
Year 541
Uncertainty Arising from Natural Variability
http://www.cesm.ucar.edu/working_groups/Climate/
i.e., spread is not
predictable
Winter Surface Air Temperature Trends 2010-2060
Projected European Climate
Uncertainty due to Natural Variability
Hurrell, Deser and Phillips (2014), Geophysical Research Letters
NCAR
2010 2010-2060 Trends (°C/51 yrs)
NCAR
Projected European SAT
Uncertainty due to Natural Variability
DJF 2010-2060
Forced
Warmest
Run 26
Coolest
Run 13
• Forced and unforced amplitudes similar over Europe
• Unforced component has large spatial scales
Natural (Total – Forced)
Warmest
Coolest
NCAR
Projected European SAT
Uncertainty due to Natural Variability
DJF 2010-2060
Forced
Warmest
Run 26
Coolest
Run 13
NCAR
Projected European Precipitation
Uncertainty due to Natural Variability
DJF 2010-2060
Wettest
Run 3
Driest
Run 9
Forced }
}
NCAR
Dominant Source of Noise: Dynamics
SLP and Air Temperature Trends 2010-60
NCAR
Uncertainty due to NAO
% chance warming
Projected European SAT (DJF)
2010-2035 2010-2060
# of ensemble members with trend > 0
total # of ensemble members
NCAR
Uncertainty due to NAO
% chance increase
Projected European Precipitation (DJF)
2010-2035 2010-2060
# of ensemble members with trend > 0
total # of ensemble members
NCAR
Projected European SLP and SAT (DJF)
Total Natural Forced = +
Run 1
0
Run 2
6
Two of 40 realizations
Dominant Source of Noise: Dynamics
NCAR
Uncertainty due to Natural Variability
Projected European SLP and Precipitation (DJF)
• Natural component can be larger than forced
• Natural component has large spatial scales
Two of 40 realizations
Total Natural Forced = +
Run 1
0
Run 2
6
NCAR
2*
Leading EOF (67%) of 40 SLP Trend Maps
(Dominant pattern of SLP trend uncertainty)
Regressions on SLP PC1
SLP and Surface Air T (DJF)
SLP and Precipitation (DJF)
North Atlantic Oscillation
NCAR
2*
Range of Outcomes
(Due to uncertainty introduced by NAO)
Forced + NAO Forced – NAO
Recommended