DOE Climate modeling strategy

G. L. GeernaertDOE/SC

• Why climate modeling

• Who does what across the federal government

• Climate modeling capabilities in DOE

• What the data show

• Who does what across the federal government

• Roadmap for DOE supported climate modeling research

Atmospheric CO2 at Mauna Loa Observatory

Modern CO2 concentrations are increasing The current concentration is the highest in 800,000 years, as determined by ice

1960 1970 1980 1990 2000 2010

320

340

360

380

Par

ts p

er m

illio

n (p

pm)

Concentration now ~400 ppmConcentration now ~400 ppm

• Antarctic ice core data show that concentration did not exceed 280 ppm for about the last 800,000 years

• Concentration prior to 1800 was 280 ppm

Under a “business-as-

usual” scenario, concentrations could rise to 1,000 ppm

Under a “business-as-

usual” scenario, concentrations could rise to 1,000 ppm

Carbon emissions by energy sector

Coal-fired power plants emit more CO2

(~ 33% of the U.S. total) than any other source,

including surface transportation.

Sources: Energy Information Administration (EIA), 2008, Annual Energy Outlook 2008;EIA, 2007, Emission of Greanhouse Gases in the United States 2006

Global temperature anomalies from the 1890–1919 average

5

Climate processes and components

ENSO and climate

USG climate modeling platforms for research

Platform Agency* Mission Assets and strengths Leveraging

GFDL NOAA Public safety, fisheries, forecasting

in-house, co-located; seasonal to centennial

DOE HPC; Community

GISS NASA Satellite sensor design/deploy

In-house, co-located; data assimilation

Community

Community NSF Basic science, workforce development

NCAR DOE labs

Community DOE Energy sector vulnerability, resilience, security

National Labs, NCAR, LCF / ASCR; UQPCMDI, ESGF/other, dedicated field research

NSF, GFDL, GISS

*each agency also has a science mission.

Department of Energy • Office of Science • Biological and Environmental Research

10 BER Climate – OMB 2015

Evolution of Earth System Research Modeling at DOEAtmosphere/

Land

Ocean

Sulfate aerosols

Carbon cycle

Dynamic Ice sheet-ocean, SLR

Atmosphere/Land

Atmosphere/Land

Atmosphere/Land

Atmosphere/Land

OceanSea-ice

OceanSea-ice

Ocean, Sea-iceOceanSea-ice

Sulfate, dust, sea-salt, carbon

aerosols

Aerosol size, mixtures, cloud

effects, chemistry

Carbon cycleInteractive vegetation

Biogeochemical cycles

Dynamic vegetation

Biogeochemical cycles

Carbon-water coupling

Human-water/energy/

atmosphere/land

Variable meshAtmosphere/ice-

sheets/ocean

Mature development

Active development

Challenge:UQ distribution within and among climate components

Integrated Earth System Prediction

(w/IA, IAV)

Department of Energy • Office of Science • Biological and Environmental Research

11 BER Climate – OMB 2015

Metrics and questions drive validation exercises

•Standard measures: temperature, precipitation, cloudiness..

•Scale: global, regional, surface, troposphere, stratosphere., …

•Questions: El Nino; Monsoonal; Sea ice; Sea level rise, (climate sensitivity) ...

Spatial resolution of climate models is increasing

25 km

150 km

75 km

300 km

Department of Energy • Office of Science • Biological and Environmental Research

13 BER Climate – OMB 2015

Metrics – what’s missing from last slide?

EXTREMES

Ice storms

Snow stormsHurricanes

Severe drought

Heat wavesCold waves

Flood

Tornadoes

Storm surge

Energy and climate change

Energy issues•Grid reliability•Energy supply•Energy extraction•Pipelines•Infrastructure design•Technology design•Security vulnerability

Climate signatures of changing…•Extreme temperature•Drought extent and duration•Sea level rise•Severe precipitation•Changing precipitation patterns•Jet stream dynamics•Sea ice coverage•Permafrost thaw rates•Surface hydrology•Subsurface hydrology

USGCRP strategic plan (released in 2012)

– Goal 1: Advance Science• Extremes, thresholds, tipping points

– Goal 2: Inform Decisions

– Goal 3: Conduct Sustained Assessments

– Goal 4: Communicate and Educate

– Crosscut: Provide knowledge on scales appropriate for decision making

– Cross-cut: Incorporate social and biological sciences

– Cross-cut: enable response to global change via iterative risk management

– Under discussion: Risk Modeling Framework

CESD Goals• Process knowledge and innovative computational methods

advancing next-generation, integrated models of the human-Earth system.

• Process-level understanding of atmospheric systems and terrestrial ecosystems, extending from bedrock to the top of the vegetative canopy.

• Coupled biogeochemical processes in complex subsurface environments to enable systems-level environmental prediction and decision support.

• Enhance the unique capabilities and impacts of the ARM and EMSL scientific user facilities and other BER community resources to advance the frontiers of climate and environmental science.

• Address science gaps that lead to solutions for DOE’s most pressing energy and environmental challenges.

Department of Energy • Office of Science • Biological and Environmental Research16 US European Workshop

Department of Energy • Office of Science • Biological and Environmental Research

17 BER Climate – OMB 2015

DOE roadmap to advance predictability

ObservationalInfrastructure

Community Modeling

Community DataInfrastructure

Uncertaintycharacterization

ComputingNumerics

Resolution

ExtremesThresholds

Tipping points

ARM

IFRCEMSL

ESGF

Ameriflux

CESM and components

Systemintegration

PCMDI

CDIACRASM

IAM

Department of Energy • Office of Science • Biological and Environmental Research18 OMB briefing Sept 2012

Predictability Research Roadmap

UQ

Spatial resolution (km)40020010050251052 800

1990

2000AR3

2007AR42013

AR52015

2020AR6?

2023

TODAY

General circulation

models

General circulation

models

Earth system models

Earth system models

Regional climate models

Regional climate models

Integrated assessment

models

Integrated assessment

models

GCMs are becoming more highly coupled like ESMs

GCMs are being

dynamically downsized to regional

scales

Regional Models and IAMs share a regional focus

IAMs will ultimately be included in ESMs to form Integrated ESMs

Relationships among model types

ESMs can be

dynamically downsized to

regional scales

What are the major knowledge gaps in climate models?

ARM Research Sites – FY14 vs FY13Testbeds for high resolution models

Southern Great Plains (1993) North Slope of Alaska: Barrow (1998) and Atqasuk (1999) Tropical Western Pacific: Manus (1996), Nauru (1998), and Darwin (2002) First ARM Mobile Facility (2005); Second ARM Mobile Facility (2010) ARM Aerial Facility (2007) Eastern North Atlantic and Third ARM Mobile Facility (2013)

°

°

Future: three major challenges for DOE

• Major upgrade in climate model platform capabilities

• Big data analytics – interoperability, diagnostics, modularity

• Shift to a new class of science problems

Department of Energy • Office of Science • Biological and Environmental Research

23 BER Climate – OMB 2015

Modeling towards high resolution, reduced uncertainty

Sources of uncertainty in climate modeling

Computational system errors and limitations

Computational system errors and limitations

Limits to theory and understanding

Limits to theory and understanding

Limits to the ability to mathematically describe

the earth system

Limits to the ability to mathematically describe

the earth system

Software ready for Exascale Computing

• Upgrade CESM codes, interoperable and modular components at petascale

• Exascale computing (1018 flops) will increase the speed of computation by three orders of magnitude over today’s state-of-the-art petascale computers

• Exascale computing will enable:– Simulation of clouds over their natural

range of scales for global climate

– Modeling fully turbulent exchange of heat and gases between the atmosphere and ocean

– Robust climate models for early warning, adaptation, and mitigation

– Higher resolution“Challenges in Climate Change Science and the Role of Computing at the Extreme Scale.” DOE BER and ASCR, 2009.

The Future: Big Data for climate/earth sciences

• Science tools, testbeds, ease of access for community

• Unification of metadata involving– ESGF, CDIAC (Ameriflux), NGEE, ARM

– Kbase

• Common formats with other agencies

Big Science Questions

• Extreme phenomena– Storms: ice, snow, hurricanes, etc.

– Storm surge

– Hydrology

• Abrupt Climate Change – Large-scale climate shifts that occur quickly,

persists for decades to millennia, and can cause substantial disruptions in human and natural systems

– 4 types of ACC would pose a major challenge to society:

• Rapid change in glaciers, ice sheets, permafrost, and sea level

• Widespread changes to the hydrologic cycle, including droughts

• Abrupt change in the Atlantic Meridional Overturning Circulation

• Rapid release of methane to the atmosphere