esa climate change initiative climate modelling user group cmug

ESA Climate Change Initiative

Climate Modelling User GroupCMUG

www.cci-cmug.org

Overview

• What are the issues for climate modelling?

• What is the role of CMUG?

• What is the added value of CMUG?(requirements, errors, data formats, exploitation,

lessons learnt)

• Some examples..

Uses of satellite data for climate• To ascertain decadal and longer term changes in the climate• Detection & attribution of observed variations to natural and

anthropogenic forcings

• Evaluate the physical processes most relevant to reducing uncertainty in climate prediction

• To develop, constrain and validate climate models thus gaining confidence in projections of future change

• Input or comparison to reanalyses (e.g. ERA-CLIM, EURO4M) • Seasonal and decadal model initialisation (ocean, land surface,

stratosphere)• To identify biases in current and past in situ measurements (e.g.

radiosondes, buoys)

Issues for climate modelling• Higher resolution (horiz, vertical, time)• Regional climate prediction (e.g. UKCP)• More physical processes• Seasonal to decadal prediction• Use of reanalyses for climate• Seamless prediction - weather prediction to climate

change using same model

• Metrics developed to evaluate models – CCI datasets can help here

• The way we use observational data is evolving

Model resolutions are increasing…

E.g. the new Met Office model, HadGEM3, will have a horizontal resolution of ~ 60 km and 85 vertical levels

Climate models are becoming increasingly complex…

A fully coupled Earth System Model includes:

• Atmosphere, ocean, sea-ice, land surface

• Land ecosystems: vegetation, soils

• Ocean ecosystems: plankton

• Aerosols: sulphate, black carbon, organic carbon, dust, sea salt

• Tropospheric chemistry: ozone, methane, oxidants

W. Collins et al., 2008

© Crown copyright Met Office

Decadal prediction:Global mean surface temperature anomaly

Observations

Forecast

Forecast from 2007

Requires data for both initialisation and verification of forecasts.

D. Smith et al., Science 2007

Issues for climate modelling• Higher resolution (horiz, vertical, time)• Regional climate prediction (e.g. UKCP)• More physical processes• Seasonal to decadal prediction• Use of reanalyses for climate (ERA-CLIM)• Seamless prediction - weather prediction to climate




The way we use data for model evaluation is evolving

Model

CloudSat

• Forward modelling of measured quantities (radiances, radar reflectivities) rather than high-level products

• Increased focus on using observations to investigate physical processes in greater detail

• Aim is to improve the representation of these processes in climate models

from A. Bodas-Salcedo et al., 2009

CMUG Consortium

ECMWFIFS, ERA, MACC

Dick Dee

MPI-MeteorologyECHAM, JSBACH

MétéoFranceArpege, MOCAGE, CNRM-CM, Mercator

Met Office Hadley CentreHadGEM, FOAM, HadISST

Thierry Phulpin

Paul Van Der Linden

Alex Loew Serge PlantonDavid Tan

Roger Saunders Mark Ringer

Silvia Kloster Stefan Kinne

Met Office Hadley CentreClimate Modelling

NWPHadGEM3, FOAM, HadSST

ECMWFReanalyses

NWPIFS (ERA-Interim)

MACC

MPI-HamburgClimate ModellingECHAM6, JSBACH

MétéoFranceClimate Modelling

NWPArpege, MERCATORCNRM-CM, MOCAGE

Climate Modellers

Reanalyses

ESA CCI projects

Sea-level

Sea surface temperature

Ocean Colour

Glaciers and ice caps

Land Cover

Fire disturbance

Cloud properties

Ozone

Aerosols

Greenhouse Gases

CMUG Consortium and models

Main Activities of CMUG1. Refining of scientific requirements derived from

GCOS for climate modellers.2. Provide technical feedback to CCI projects3. Assess the global satellite climate data records

(CDRs) produced from the 10 CCI consortia4. Look specifically at required consistencies across

ECVs from a user viewpoint. 5. Promote and report on the use of the CCI

datasets by climate modellers 6. Interact with related climate modelling and

reanalysis initiatives.


GCOS for climate modellers.2. Provide technical feedback to CCI projects3. Provide reanalysis data to CCI projects4. Assess the global satellite climate data records



datasets by modellers 7. Interact with related climate modelling and


Implications for requirements

• The new ECV datasets must have added value over existing ones and future proof for model evolutions

• Start from GCOS Tables as much has been done there• Be clear about applications for specific dataset as this drives

the required accuracy:– Climate monitoring high stability, precision and accuracy– Change detection high stability, precision – Evaluate processes in model high precision and accuracy– Model validation high stability, precision – Assimilation high precision

• Datasets must be globally complete (spatially and temporally) • Uncertainty estimates are as important as product itself for all

applications. Correlation of errors in space/time also important

Lessons learnt from past• Recognise move of modellers to using lower level of

products (e.g. level 1 radiances). This is especially true for reanalyses (N.B. Importance of GSICS and CLARREO)

• It took more than 15 years to get ISCCP cloud and ATSR SST datasets used for climate

• Observation simulators are important for some satellite products to compare apples with apples (e.g. clouds ..)

• Good statistical summaries of TCDRs help• CCI projects should provide colocated datasets• Essential to include error characteristics• Easy access to data and simple format to read

Observation simulators

• Ensures comparison of equivalent model variable with observations

• This was the key for use of ISCCP clouds• Note additional source of error from simulator in

comparisons

Observationsimulator

COSPCFMIP Observational Simulator Package

MODEL WORLD

OBSERVATIONS

Radar Reflectivity

Alt

itu

de

(km

)

COMMON GROUND

COSPCloudSatCALIPSOISCCPMISRMODIS

STATS

STATS

Users are being consulted

Reanalysis centres:ECMWF, JMA, NCEP, GMAO, CIRES

Climate modelling centres consulted:Hadley, UEA,MPI, IPSL, MF CNRM, Rossby Centre, GFDL, GISS, NCAR, JPL, JAMSTEC, NCEO, MRI-JMA, CMA, CAWCR, NCMWRF, KMA, ….

On-line questionnaire is available at:http://survey.euro.confirmit.com/wix/p416267727.aspx until 4th July.• To date replies from about 25 respondents• Also meeting at EGU General Assembly to gather inputs• IS-ENES to provide input to questionnaire and help analyse results• BADC providing advice on data format issues

Speaking the same language

• Definition of variables has in the past been top-down

• New communities are more bottom up via internet fora

• We need to bridge the gap between EO data providers and climate modellers

• CMOR NetCDF is an example from the climate world

Satellite Climate world world

Data Format Issues(inputs from EGU meeting)

• Access: FTP, Web browser, OpenDAP,..• Level of processing:

– Level 1 (swath) for model assessment (N.B. needs model observation operator ideally in COSP)

– Level 2 (swath) for model process studies and inferring trends– Level 3 (gridded) for generic model evaluation

• Format: CF compliant NetCDF (but what about swath data?)

• Projection: Lat/Long preferred• Tools for reading: Dataset producers should provide

these• Consistency for all products produced in CCI

Examples of CMUG input

• Ensure CDRs proposed are useful for climate or reanalysis aplications

• Ensure proposed datasets are consistent with requirements

• Provide a consistent framework for specification of errors

• Assess the need for observation simulators or other tools for exploitation

Error characterisation of CDRs

• An estimate of the errors for each CDR produced is essential for use in climate applications

• There are several types of errors– Precision – Accuracy – Stability – Representativeness

• The importance of specifying each depends on the application

• Errors should be specified on a FOV basis. Aggregated error estimates are not sufficient

• Single sensor products are simpler than merged products• Error correlations are also important to document

See next slide for definitions

Errors associated with CDRs• Accuracy is the measure of the non-random, systematic error, or

bias, that defines the offset between the measured value and the true value that constitutes the SI absolute standard

• Precision is the measure of reproducibility or repeatability of the measurement without reference to an international standard so that precision is a measure of the random and not the systematic error. Suitable averaging of the random error can improve the precision of the measurement but does not establish the systematic error of the observation.

• Stability is a term often invoked with respect to long-term records when no absolute standard is available to quantitatively establish the systematic error - the bias defining the time-dependent (or instrument-dependent) difference between the observed quantity and the true value.

• Representativity is important when comparing with or assimilating in models. Measurements are typically averaged over different horizontal and vertical scales compared to model fields. If the measurements are smaller scale than the model it is important. The sampling strategy can also affect this term.

CMUG specific assessments

??

Integrated view of ECVs1. Through ensuring common input datasets are used for

CDR creation and in some cases common pre-processing (e.g. geolocation, land/sea mask, cloud detection)

2. Through comparisons of CDRs for different ECVs (e.g. SST, sea-level, sea-ice and ocean colour)

3. Through comparisons of CDRs with model fields (e.g. GHG and Ozone CDRs and MACC model profiles/total column amounts) CMUG will be involved in development of observation simulators for some ECVs Pre-cursors of ECVs will be used for preparation.

4. Through studying teleconnections (e.g. El-Nino SST shows consistent impact on cloud fields, fires).

5. Through assimilation of CDRs and to assess impact on analyses and predictions (e.g. SST in ERA-Interim)

Landcover

• ECV landcover will provide land cover information, but no land surface parameters associated with it.

Model parametersSurface paramters per grid cell and PFT: e.g.-Albedo -Background albedo-LAI-faPAR-Forest ratio

-Soil parameter-Roughness length

Land cover product

Landcover

• ECV landcover will provide land cover information, but no land surface parameters associated with it.

• Objective– Generation of a consistent land

surface parameter data set to be used in climate models

• Variable in time (no pure climatology)

• Additional information about variability of surface parameters per PFT at the model grid scale

– Evaluate the impact of the new surface parameter set on climate model simulations using ECHAM6

– Provision of the data set to the climate modelling community for assessment in their models

Model parametersSurface paramters per grid cell and PFT: e.g.-Albedo -Background albedo-LAI-faPAR-Forest ratio

-Soil parameter-Roughness length

Land cover product

Understanding the effect of changes in the observing systems is key to understanding reanalysis quality

Data used in ERA-Interim

Total ozone content over the period 2000-2005

(Tesseydre et al, 2007)

MOCAGE-Climat

NIWA climatology

Use of ISCCP to evaluate modelsLow level cloud: CTP < 680 hPa

Thick “Stratus”

OD > 23

Medium “Stratocu”

3<OD<23

ISCCP HadGEM1 HadCM3

Martin et al. (2006)

Related Activities1. GCOS and GSICS2. EU (IS-ENES, EUGENE, ..) 3. EUMETSAT (CM-SAF) and SCOPE-CM4. NOAA-NASA initiatives (e.g. JPL CMIP5)5. WCRP Observation and Assimilation Panel6. Reanalyses (ERA-CLIM, MACC) + EURO4M7. Coupled Model Intercomparison Project and

follow-on activities8. IPCC AR-5 and AR-6

Proposed CMIP5 model runs

AR-5

CCI datasets couldstart to be used in the evaluation of theseresults

Summary

• CMUG has now started to support the ESA CCI.

• We are seeking input from the climate modelling and reanalysis communities.

• It is crucial the products produced are ‘fit for purpose’ otherwise this will be a lost opportunity (and wasted money).

• If interested please get in touch via [email protected]

Any questions?www.cci-cmug.org

[email protected]

esa climate change initiative climate modelling user group cmug

Documents

climate models

climate applications

climate modelling community

climate system perspective

model metrics

observational data

use of cci data sets

model evaluation