NOAA Climate Research: Climate Variability - PowerPoint PPT Presentation


PPT – NOAA Climate Research: Climate Variability PowerPoint presentation | free to download - id: 15bba6-ZDc1Z


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation

NOAA Climate Research: Climate Variability


CLIVAR Atlantic research examples. Coupled modeling - TAV, NAO. P. ... EPIC. NAME. MESA. North American Monsoon Experiment. Phenomena of Interest. I. II. III ... – PowerPoint PPT presentation

Number of Views:29
Avg rating:3.0/5.0
Slides: 34
Provided by: jamesk
Learn more at:


Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: NOAA Climate Research: Climate Variability

NOAA Climate Research Climate Variability
NOAA Research Overview
Climate Variability
Drought in Great Plains, ca. 1934
California floods during 1998 El Nino
Mission-critical research
Climate variability research is critical to
NOAAs mission To understand and predict
changes in the Earths environment and conserve
and manage coastal and marine resources And,
specifically, Strategic Goal 2 Understand
climate variability and change to enhance
societys ability to respond.
NOAA Strategic Plan Performance Measures
CV Overall Strategy
  • Four components
  • Monitor and Observe
  • Understand and Describe
  • Assess and Predict
  • Engage, Advise, and Inform
  • For the success of the overall program, it is
    vital that these components be linked together.

Measures of success
Measures of Success
  • Understand and describe
  • Increased number of new research findings and
    progress toward their implementation into NOAA
  • Decreased degree of uncertainty of climate system

CCSP priority
U.S. Climate Change Science Plan Chapter 4
Climate Variability and Change
  • Major Research Questions
  • To what extent can uncertainties due to climate
    feedbacks be reduced?
  • 2. What are limits of climate predictability, and
    how can climate predictions and climate change
    projections be further improved?
  • What is the likelihood of abrupt climate changes?
  • 4. How do extreme events respond to climate
    variability and change?
  • 5. How can information on climate variability and
    change be most efficiently developed and
    communicated to serve societal needs?

CV Research Priorities
Research Priorities
  • NOAA CV programs emphasize priority areas
    described in the CCSP and NOAA SP.
  • Increase understanding of climate feedbacks
  • 2. Clarify limits to climate predictability
    improve climate predictions (CLIVAR CDC other
    NOAA labs).
  • Increase understanding mechanisms for abrupt
    climate change (CLIVAR ATL, SEARCH).
  • 4. Response of extreme events to climate
    variability and change (Weather-Climate
    Connection CDC, ETL, AOML).
  • 5. Develop climate information to serve societal
    needs (CDC overall program, especially where
    linked to RISAs, CDEP, NWS, and IRI).

CV Resource allocation
CV Resource Allocation (FY03 est.) (in M)
.9 AOML 4.2 CDC 2.5 ETL .2 FSL
.2 PMEL 2.4 ENSO obs (PMEL) 4.7
35.3 CLIVAR amount includes 6.1 M for
sustained ocean obs., and OGP-sponsored research
funding for CLIVAR Pacific, CLIVAR Atlantic, and
CLIVAR Pan-American Climate Studies (PACS).
Support for Observing Systems
NOAAs global and regional observing systems are
crucial in supporting monitoring,
interpretations, and predictions of climate
Total Climate Variability contributions to the
Climate Observation Program are 13.1M (CLIVAR
Obs 5.1M, OAR Labs 8M).
Core Activities
NOAA Base Components Climate Variability
  • Office of Global Programs
  • Climate Variability and Predictability (CLIVAR)
  • - CLIVAR Atlantic
  • - CLIVAR Pacific
  • - CLIVAR Pan-American Climate Studies (PACS)
  • Climate Observations and Services Program (COSP)
  • Study of Environmental Arctic Change (SEARCH)
  • Weather-Climate Connection
  • NOAA Research labs

  • CLIVAR Atlantic
  • CLIVAR Pacific
  • CLIVAR Pan-American Climate Studies (PACS)

Climate Variability and Predictability
  • Overall Objectives
  • Develop an understanding of of natural climate
    variations and their global and regional
  • Assess predictability of these climate modes
    through observational and modeling studies.
  • Foci
  • El Niño - Southern Oscillation (ENSO), Pacific
    Decadal Variability (PDV), Arctic Oscillation
    (AO), North Atlantic Oscillation (NAO), Tropical
    Atlantic Variability, and North American monsoon
  • Abrupt climate change (Atlantic thermohaline
  • Method
  • Sponsor PI research/field experiments in key
    regions CLIVAR-Pacific, CLIVAR-Atlantic,
  • Support interagency national and international
  • Implement Climate Model Process Teams (CPTs) to
    develop and improve climate model representations
    of physical processes.

CLIVAR Atlantic aims to describe changes in, and
assess predictability of, three major climate
  • Tropical Atlantic Variability (TAV)
  • North Atlantic Oscillation (NAO)
  • Meridional Overturning Circulation (MOC)

Figure courtesy of Science (2002)
CLIVAR Atlantic research examples
  • Coupled modeling - TAV, NAO
  • P. Chang and R. Saravanan
  • S.-P. Xie
  • J. Marshall
  • Coupled modeling - tropical teleconnections
  • M. Hoerling and J. Hurrell
  • Ocean modeling - TAV (including subtropical
    cells), MOC
  • G. Halliwell and R. Weisberg
  • Atmospheric analysis - NAO
  • J. Hurrell et al.
  • R. Miller et al.
  • M. Baldwin and T. Dunkerton
  • Data set development and analysis - TAV, MOC, NAO
  • P. Niiler
  • L. Yu and R. Weller

CLIVAR Pacific Objectives
  • Improve understanding of Pacific basin-scale
    atmosphere-ocean variability, its predictability
    on seasonal and longer timescales, and
    anthropogenic impacts. Particular foci include
    ENSO and Pacific Decadal Variability. This
    requires further comprehensive analysis, testing
    and improvement of coupled models.
  • Document time-varying T, S, currents in the upper
    ocean at 300 km, 10 day resolution over the
    entire basin north of 40o S for a 15 year period,
    with higher resolution in boundary currents and
    near the equator. Apply ODA to provide a
    three-dimensional time-dependent analysis based
    on this data.
  • Document time-varying vertical and lateral fluxes
    and air-sea exchange of heat, fresh water, and
    momentum over the corresponding period.
  • Improve physical parameterizations in OGCMs,
    AGCMs and NWP models via process studies and via
    ODA, which as a by-product identifies systematic
    errors in the atmospheric forcing of the ocean or
    the assimilating ocean model.

PACS Science Objectives
  • PACS seeks to extend the scope and improve the
    skill of climate prediction over the Americas on
    subseasonal to interdecadal time scales with an
    emphasis on summer precipitation. Specific
  • Improve understanding and provide more realistic
    simulations of coupled ocean-atmosphere-land
    processes, with emphasis on
  • the response of planetary-scale atmospheric
    circulation and precipitation patterns to
    potentially predictable surface boundary
  • the mechanisms that couple climate variability
    over ocean and land
  • the seasonally varying climatological mean state
    of the ocean, atmosphere, and land surface
  • The effects of land surface processes and
    orography on the variability of seasonal rainfall
  • Determine the predictability of warm-season
    precipitation anomalies over the Americas on
    seasonal and longer time scales.
  • Advance the development of the climate observing
    and prediction system for seasonal and longer
    time scales.

  • PACS focuses on the phenomena that are crucial
    for organizing seasonal rainfall patterns
  • - the oceanic ITCZs
  • - the continental scale monsoon systems,
  • - the tropical and extratropical storm tracks
  • For implementation, three regional process study
    domains are defined
  • - Eastern Pacific (EPIC and VEPIC)
  • - North American Monsoon System (NAME)
  • - South American Monsoon System (MESA)

North American Monsoon ExperimentPhenomena of
PACS Achievements
  • Established enhanced pibal upper air sounding
    network and auxiliary raingauge networks to
    augment existing networks over Pan-America for
    studying low-level flow and precipitation
  • Enhanced observing system in eastern Pacific with
    extension of 95W TAO line and stratus mooring
  • Initiated the PIRATA array of moored buoys in the
    tropical Atlantic

PACS Deliverables
  • Measured improvements in coupled climate models
    capability to predict North and South American
    climate variability months to seasons in advance
  • Infrastructure to monitor and predict the North
    and South American monsoon systems
  • More comprehensive understanding of Pan-American
    summer climate variability and predictability
  • Contributions to assessments of climate
    variability and long-term climate change for
    regions within North and South America
  • Strengthened multinational scientific
    collaboration across Pan-America

CLIVAR deliverables
CLIVAR milestones/deliverables
  • Improved climate predictions for global climate
    variability on S/I and longer time scales.
  • Contributions to the development of the sustained
    global climate observing system, esp. ocean
  • Data sets from process field campaigns.
  • Improved physical understanding of climate
  • Assessments of predictability of climate modes.
  • Accelerated improvements in modeling of physical
    processes through the CPTs. Initiate a CPT
    focusing on deep atmospheric convection (Q4,
  • Conduct the South American Low Level Jet
    Experiment (Q2, OGP)
  • Enhance observations in Mexico and the southwest
    U.S. for NAME experiment (Q4, OGP)

Study of Environmental Arctic Change (SEARCH)
  • Objectives
  • Identify causes for observed multi-decadal trends
    of interrelated changes in the Arctic
    (atmosphere, ice, ocean, land).
  • Clarify potential for feedbacks (albedo, fresh
    water export,
  • release of carbon from permafrost/methane
  • Determine implications for abrupt changes.
  • Assess impacts to ecosystem and society.
  • Foci
  • Interannual to decadal time scales.
  • Arctic/subarctic ocean fluxes relationship to
  • thermohaline variability.
  • Expand on limited observations to track key
  • incorporate into models.
  • Method
  • Implement and sustain environmental observations.
  • Data analysis and process research.

SEARCH Products
Examples SEARCH Products
Atmosphere-Ocean-Land Surface Interactions
and Feedbacks over Arctic
Prototype Observing Array
SEARCH Products/deliverables
SEARCH deliverables
  • Temperature, radiation and ice data to support
    analyses of ice/albedo feedback, ocean
    thermohaline circulation, Arctic shipping routes,
    marine mammal management.
  • Atmospheric data to enhance model physics and
    improve prediction of Arctic Oscillation, US
    temperature and hydrologic forecasts
  • Long-term data to detect decadal changes,
    demonstrate links to mid-latitudes

Weather-Climate Connection
Weather-Climate Connection
  • Objectives
  • Improve understanding and predictions of links
    between climate variations and high impact
    weather phenomena
  • Improve regional observing capabilities
  • Develop stronger link between climate research
    and user needs
  • Infuse new science and technology into NOAA
  • Foci
  • Improve predictions on weekly to seasonal time
  • Initial focus on tropical-midlatitude
    interactions over the Pacific and their regional
    impacts on U.S.
  • Method
  • Observational, diagnostic, and modeling studies
    at regional scales to assess predictability and
    realize the potential for operational prediction.
  • Research coordinated with services (NWS) and end

Research example Wx-Clim. Connection
Weather-climate Connection
40 of rain/ 7 days
Pineapple Express

  • Where will storm track be for the next few weeks?
  • When will an arctic outbreak affect the east
  • When will the rain (drought, heat wave, etc.)
  • How will a climate shift affect the weather in a
    particular region?

Week 2 reliability and skill score derived from
an Ensemble MRF Reforecast Experiment (CDC) (23
years of training data, cross validated)
Bottom line Big gains can be made in forecast
skill by statistically correcting forecasts from
a frozen model. Validations over last two years
indicate that skills of U.S. T, p week two (8-14
day) forecasts derived by this method are
superior to official 6-10 day forecasts.
How many years of training data are needed?
Results suggest that most of the gains can be
achieved by conducting reforecasts for 5-10
years, with forecasts run every 4-5 days.
Weather-Climate Connection deliverables
Weather-Climate Connection Milestones/Deliverable
  • Improved regional forecast capabilities of U.S.
    temperatures and precipitation from a week to a
  • Climate prediction capabilities for high-impact
    events, including droughts and major floods.
  • Enhanced data sets and analyses to identify and
    interpret weather-climate connections between the
    tropics and mid-latitudes.
  • Develop modified and improved practices for
    biweekly and/or monthly U.S. temperature and
    precipitation outlooks (Q2 CPC/CDC).

Research Laboratories
NOAA Research
  • Objectives
  • Carry out long-term research central to NOAAs
  • Provide sustained support for NOAA climate
    observations and services (e.g., NWS Climate
    Prediction Center)
  • Deliver products for decision support
  • Regular and timely provision of climate obs. and
  • Foci
  • Develop national capabilities to describe,
    interpret, and predict climate variations,
    emphasizing major climate phenomena such as ENSO,
    droughts, and floods.
  • Provide and interpret ocean data
  • Develop capabilities to monitor and predict the
    ocean environment on time scales from days to

Research Example AOML
PMEL Recent Accomplishments
  • Monitor and observe The TAO array has provided
    accurate, high resolution, real-time data for
    tracking the evolution of the 2002-2003 El Niño.
  • Understand and describe TAO data and related
    shipboard measurements have supported
    approximately 50 refereed journal publications
    per year on climate variability and change.
  • Assess and predict The TAO array is the backbone
    of the ENSO observing system providing real-time
    data essential for accurate analyses and
    forecasts of evolving climatic conditions in the
    tropical Pacific. TAO data were fundamental to
    the successful NCEP forecast in January 2002 that
    an El Niño was developing.
  • Engage, advise, inform a) PMEL scientists serve
    on many national and international committees
    promoting awareness of climate science and NOAAs
    climate mission. c) TAO web pages, providing
    valuable information to the general public,
    educators, government policy makers and private
    businesses, received nearly 25 million hits in
    the past year.

Test of Bridge Hypothesis in GFDL Model (CDC,
GFDL) - obs. vs. model correlations - 1950-99
Observed SSTs (FMA) correlated with Niño Index
Mixed Layer Model correlations - SSTs
(FMA) specified in Niño region, MLM elsewhere
The Perfect Ocean for Drought (CDC, CPC)
Observed Temperature and Precipitation
anomalies (June 1998 - May 2002)
Model-simulated Temperature and Precipitation
Anomalies given SSTs over this period
Research lab deliverables
NOAA Research FY03 Milestones/Deliverables
  • Determine the origins and assess the
    predictability of the 1998-2002 U. S. drought,
    leading to improved drought forecasts (Q2, CDC)
  • Continue internationally coordinated studies to
    determine the role of the Tropical Atlantic on
    global climate (Q3, AOML)
  • Develop a website dedicated to ongoing, real-time
    predictions of tropical convection associated
    with the MJO (Q3, CDC).
  • Provide data for operational forecasting and
    analyses of the 2002-2003 El Niño and for
    comparisons with previous events (Q4, PMEL).