Dan Cayan Scripps Institution of Oceanography, Climate Research Division, UCSD

andWater Resources Division, USGS

Thanks: Mike Dettinger, Noah Knowles, Mary Tyree

California’s Future ClimateLessons from Scenarios Assessments

Funding: PIER Program, California Energy Commission RISA Program, NOAA Office of Global Programs DOE

More info: http//:meteora.ucsd.edu/CAP/

Gov Swartzenegger’s June 2005 Executive Order

commissioned thisClimate Assessment, which

investigated potential climatechange impacts and formed key

scientific background for California’s greenhouse gasemissions legislation, AB-32

which was passed in fall 2006

Available on the web athttp://www.climatechange.ca.gov/biennial_reports/2006report/

California Climate Change Scenarios

Assessment

some lessons learned

Vulnerability:Vulnerability: response to a +3ºC warming response to a +3ºC warming

What fraction What fraction of each year’s of each year’s

precipitation precipitation historically fell historically fell

on days with on days with average average

temperatures temperatures just below just below freezing?freezing?

““Rain vs Snow”Rain vs Snow”

More vulnerableMore vulnerable

YOSEMITE

Computed by Mike Dettinger from gridded Computed by Mike Dettinger from gridded historical US weather data (from Bates et al, in historical US weather data (from Bates et al, in

rev)rev)

Less vulnerableLess vulnerable

Winter (Nov-Mar) SFE/P trends at western USweather stations: symbol area is proportional to study-period changes, measured in standard deviations as indicated; circles indicate high trend significance (p<0.05), squares indicate lower trend significance (p>0.05).

More RainLess SnowWY 1949-2004

Noah Knowles et al. 2006 in press J. Climate

We face significant losses of spring snowpack

By the end of the century California could lose half of its late spring snow pack due to climate warming. This simulation by Noah Knowles is guided by temperature changes from PCM’s Business-as-usual climate simulation. (a middle of the road emissions scenario)

• Less snow, more rain

• Particularly at lower elevations

• Earlier run-off

• More floods

• Less stored water

Knowles and Cayan 2001

since 1985 the number of large wildfires in western U.S. increased by 4X

Anthony Westerling et al. Science August 2006

Large wildfire

threat isaggravated by warmer

springs and summers

Anthony Westerling et al. Science August 2006

Ocean Beach , February 1983

Extreme storm-forced sea levels during an extreme tide

Observed SFO (left) and modeled Global (right). Sea level rise estimates based upon an envelope of output from several GHG emission scenarios

observed

Projected envelope of global s.l. rise

Climate models Only provide loose guidance onThe amount of sea level rise, butIt is very likely that rates will increase

San Francisco Bay/Delta Water Levels Confluence of sea level rise and increased flooding

~Jan 3 2006Nasa http://earthobservatory.nasa.gov/Newsroom/NewImages/Images/ncalifflood_amo_2006004_lrg.jpg

-2

0

2

4

6

8

1975 2000 2025 2050 2075 2100

Temperature Change (

oF)

A1 H ad C M 3 A1 P C M A2 H ad C M 3 A2 G F D L2.1

A2 P C M B 1 H ad C M 3 B 1 G F D L2.1 B 1 P C M

Uncertainty: Projected Warming Ranges Statewide annual average (°F)

14.4

10.8

9

3.6

0

-3.6

Lower Warming Range

Medium Warming Range

Higher Warming Range

Tem

pera

ture

Cha

nge

(°F

)

Meditterranean precipitation regime remains

Although models unanimously indicate a warmer climate, they are undecided if it will get wetter or dry out

6 different climate models 5 emission scenarios, IPCC SRES runs

Mike Dettinger, 2005San Fran. Estuary and Watershed Science

Projected patterns of precipitation changes2090-2099 versus 1980-1999

Globally, dry regions become drier?

Mike Dettinger, Jim Wells USGS and SIO record streamflow in Tuolumne Meadows

Challenges

Seasonally intensified warming?some models suggest amplified summer warming

Climate models project 1.5-2.C ocean surface warming by end–of-century.

Greater warming on land than oceans would amplify California coast-interior thermal gradient.

Summer land warming is accentuated

GFDL CM2.1 Jun-Aug air temp change2070-2099 minus 1961-1990

GFDL CM2.1 is a medium-high sensitivity model. Other models produce less (or more) warming

Modeling climate over California’s complex terrain: July 10m Wind Diurnal Variation

Hist means (inches):Div2 36.02Gfdl 43.13Pcm 29.63

Are modelscapable of producing realisticsuite of

wet and dryspells?

Distribution ofObs and simulated

5yr precip departures

obs

sim

sim

Sacramento drainage div

California needs a sustained modern Climate Observation Network

Douglas AldenScripps Institution of OceanographyInstalling met stationLee Vining, CA

Implications for Monitoring Primary snowpack loss is above ~1500m

Knowles and Cayan, 2004

Total number of stream gages below 1500m

(red)

Total number of HCDN gages below

1500m (red)

Total number of gages above 1500m (blue)

Total number of HCDN gages above

1500m (blue)

Rain-snow transition zoneNeeds more

Careful monitoing

Over all Stream gages, present-day elevational bias is small…

…but climate-quality HCDN subset is under-represented in crucial elevations.

Noah Knowles, USGS)

……on the other hand, it can be very weton the other hand, it can be very wetMay 16th 2005 A warm storm in the SierraMay 16th 2005 A warm storm in the Sierra Yosemite Valley floods from a 1” rainYosemite Valley floods from a 1” rain

need to understand event scale phenomena projected heat wave days

SRES A2 GHG Emissions Scenario

Slide 11Slide 11

Need observations of 3-d, upstream atmosphere

Hydromet testbed, NOAA ETL and collaborators

How to effectively work and learn across disciplines?

Scenarios, not forecasts

PCM. GFDL, HADModel-based

Climate Projections

Agriculture

Water Energy

Forests/Fire

Coping Capacity/Preparedness

Impacts-Physical-Ecological-Economic-Social

HumanHealth

Coasts

Susi Moser, NCAR

Recommendations (without much discussion) :Improved, sustained observations

higher spatial resolution, more coverage

3-d atmosphere and ocean upstream

California meso-micro climates including urban and agricultural settings

improve insitu network, real-time communications

scientific quality record keeping of economic, social measures

data archeology

Continued modeling at several scales, disciplinesOngoing climate simulation compute consortium Ongoing Consortium for

climate simulation/prediction to knit state/campuses/labs

End-to-end assessments, including eco, economic, social; close State participation

Study extreme events as well as secular changes

Study aerosol-clouds-precipitation

Link with other regional, national, international efforts

Support California State Climate Scenarios Assessments

Promote CEC-PIER Annual Climate Change Conference

Develop closer, better links to decision makers; strengthen outreach

Fellowship program for grad students, post docs to ensure continuity, new generation