date: Fri, 19 Nov 2004 11:14:00 +0000 from: Sarah Raper subject: Fwd: IPCC AR4: Projections using EMICs to: Tom Wigley --Apple-Mail-1-149073115 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Dear Tom, Please see this draft call for EMICs. I think it will be circulated on Monday so please don't make any comment until then, but here is a pre-view. It is good for us I think since after some discussion they decided to use our SCM officially as the reference. Please forward to me the PDMCI emails you got recently (Tim O sent them to me at UEA but now they are inaccessible on my PC there). I think I will try to register for data, then perhaps I will get emails direct. I don't know what all those stabilization scenarios are going to be from Bern - partly why I send this to you.... Sarah Begin forwarded message: > From: Thomas Stocker > Date: 19 November 2004 06:59:35 GMT > To: "Raper, Sarah " , "Raper, Sarah " > > Subject: IPCC AR4: Projections using EMICs > > Dear Sarah > > Below is the nearly final letter to the community at large. It > cristalized that we do not want to call for SCM simulations in this > letter but focus on EMICs solely. The SCM simulations should be done > exclusively by you using your model to provide the important link > across all ARs. We call those "reference SCM simulations". > > As to the EMICs the process is open, as long as there is a citable > model description in the referenced literature. > > Please have a look at the scenarios, the letter and the formulations. > I would like to send this out no later than Monday. > > Best regards > > Thomas > > > Dear Colleague > > As coordinators of Chapter 10 "Global Climate Projections" of the > Fourth > Assessment Report of the Intergovernmental Panel on Climate Change > (AR4), IPCC, we invite you to contribute long-term model simulations > using EMICs, Earth System Models of Intermediate Complexity. > > By EMIC we understand models that are dynamically simpler than > comprehensive AOGCMs, although they might well be more "complete" in > terms of climate system components that are included. Typically, EMICs > are some composite of simplified versions of atmospheric (e.g. energy > balance, or quasi-geostrophic), and/or ocean model components (e.g., > zonally averaged), and a suite of parameterizations. In order to > qualify > for participation, a technical description of the EMIC must have > appeared in an international, refereed journal. > > In Chapter 10 of AR4 we will present, in addition to the projections to > year 2100 by the most comprehensive AOGCMs, long-term projections out > to > year 2500 or 3000 using EMICs. The traditional reference simulations > with a simplified climate model (SCM) will also be included. > > The new EMIC simulations have a number of goals: > > 1. We wish to estimate the range of uncertainty of long-term climate > evolution beyond 2100, such as global mean temperature change, sea > level > rise, ocean heat uptake, Atlantic meridional overturning, etc. > > 2. Scenarios in addition to SRES should be simulated with models more > complex than SCMs, in order to estimate concentration and emissions > commitments, i.e. what long-term global changes are expected when > either > concentrations are kept constant (concentration commitment), or > emissions are set to zero (emissions commitment). > > 3. EMICs ensemble simulations should be used to estimate model-PDFs of > certain quantities; > > 4. Some EMICs allow for the equilibrium climate sensitivity to be > tuned. > With these, the dependence of the simulated quantities on climate > sensitivity should be estimated. Such EMICs must, in addition to > simulations with the preferred equilibrium climate sensitivity, present > results for equilibrium climate sensitivities of 1.5 degC, 3 degC, and > 4.5 degC for a doubling of CO2. > > 5. Some EMICs exhibit multiple equilibria and transitions to different > climate states (e.g., a shut down of the Atlantic MOC). This feature is > not simulated by SCMs and therefore, the opportunity exists to > investigate the dependence of climate projections on such nonlinear > changes. > > EMICs are highly simplified and parameterized models. They are > heterogeneous and no published intercomparison experiments are > available > currently. Therefore, we will admit only globally averaged quantities > in > these projections. In order to ensure comparability of different > models, all need to simulate a set of well-defined scenarios as a > minimum requirement to be included in Chapter 10. > > Furthermore, we encourage all groups who have different versions of > their EMICs (e.g., with different ocean mixing parameterisations, or > different atmospheric parameter settings) to submit results for each > model version. This allows us to assess the robustness of results with > respect to different model parameterisations. Due to space limitations, > we cannot guarantee that all results submitted to us will be included > in > Chapter 10. > > This solicitation of EMIC results will give us, for the first time, to > assess the range of uncertainty across the entire model hierarchy. We > trust that this will be a novel and unique contribution to the IPCC > process. > > If you have any queries to this call, or identify issues which require > clarification, or any other comments, we look forward to hear from you. > > Thank you for your efforts. > > Best wishes > > Thomas Stocker & Gerald Meehl > Coordinating Lead Authors Chapter 10 > > > *********************************************************************** > * > > INSTRUCTIONS FOR THE CONTRIBUTION OF EMIC MODEL RESULTS TO THE FOURTH > ASSESSMENT REPORT OF IPCC WORKING GROUP 1 (Chapter 10) > > *********************************************************************** > * > > Documentation > ------------- > > Describe your model in one paragraph, adressing the different climate > system components it includes. Also provide information according to > the > entries of Table 8.1 (http://www.grida.no/climate/ipcc_tar/wg1/316.htm) > of the Third Assessment Report of IPCC. > > Required: give reference to papers in refereed journals which describe > the model set up and characteristics. > > > Control Simulations > ------------------- > > In order to quantify climate drift, we require a control run from all > models with CO2 held constant at 280 ppm, with an integration time of > 1000 years. Output as specified below. > > > Scenarios > --------- > > 1. Idealized stabilization scenarios to 2xCO2 > ............................................. > > > Note: all CO2 concentrations are referred to "pre-industrial" with pCO2 > = 280 ppmv. > > Integration time 0 to 3000 years > 1a. IS2x0.5: Idealized Stabilization Scenario 0.5%/yr to 2xCO2, then > constant > 1b. IS2x1.0: Idealized Stabilization Scenario 1%/yr to 2xCO2, then > constant > 1c. IS2x2.0: Idealized Stabilization Scenario 2%/yr to 2xCO2, then > constant > > > 2. Idealized stabilization scenarios to 4xCO2 > ............................................. > > Integration time 0 to 3000 years > 2a. IS4x0.5: Idealized Stabilization Scenario 0.5%/yr to 4xCO2, then > constant > 2b. IS4x1.0: Idealized Stabilization Scenario 1%/yr to 4xCO2, then > constant > 2c. IS4x2.0: Idealized Stabilization Scenario 2%/yr to 4xCO2, then > constant > > > 3. Idealized reduction scenarios after increase to 4xCO2 > ........................................................ > > Integration time 0 to 3000 years > 3a. IS4x0.5R1.0: Idealized Reduction Scenario: 0.5%/yr to 4xCO2, then > -1%/yr to 1xCO2, then constant > 3b. IS4x1.0R1.0: Idealized Reduction Scenario: 1%/yr to 4xCO2, then > -1%/yr to 1xCO2, then constant > 3c. IS4x2.0R1.0: Idealized Reduction Scenario: 2%/yr to 4xCO2, then > -1%/yr to 1xCO2, then constant > > > 4. Smooth stabilization profiles > ................................ > > profiles will be available from www.climate.unibe.ch/emicAR4. They are > updates of WRE Scenarios, with observed CO2 until 2003 > > Integration time 1765 to 3000 > > 4a. SP450 > 4b. SP550 > 4c. SP650 > 4d. SP750 > 4e. SP1000 > 4f. DSP450: Stabilization profile to 450 ppm with delayed turning point > 4g. DSP550 > 4h. OSP350: Stabilization profile to 350 ppm with overshoot > 4i. OSP450 > > > 5. Stabilisation profiles for comparison with PCMDI results > ........................................................... > > For comparison with AOGCM results three additional stabilisation > profiles are run: A1B, A2, and B1 which are those considered in CMIP. > Data for these scenarios are given in IPCC TAR, Appendix II. Those > EMICs > with no radiation code will use values from Table II.3.11 > > Integration time 2000 to 3000 > > 5a. A1B: following SRES scenario A1B to year 2100, then constant > 5b. A2: following SRES scenario A2 to year 2100, then constant > 5c. B1: following SRES scenario B1 to year 2100, then constant > > > > 6. Emission commitment > ...................... > > These simulations are only for EMICs which include a dynamic carbon > cycle component. As a requirement, this component must be published in > the refereed litereature also. > > Emissions are calculated based on SP scenarios and the Bern carbon > cycle > model. They will be available from www.climate.unibe.ch/emicAR4. > > Time 1765 to 3000 (for EMICs with Carbon Cycle model component only) > > 6a. SP450EC2100: Emission Commitment Scenario: CO2 Emissions prescribed > according to SP450, set to 0 after 2100 > 6b. SP550EC2100: Emission Commitment Scenario: CO2 Emissions prescribed > according to SP550, set to 0 after 2100 > 6c. SP750EC2100: Emission Commitment Scenario: CO2 Emissions prescribed > according to SP750, set to 0 after 2100 > 6d. SP1000EC2100: Emission Commitment Scenario: CO2 Emissions > prescribed > according to SP1000, set to 0 after 2100 > > > > Conversion to radiative forcing > ------------------------------- > > For those models without a radiation code the radiative forcing of CO2 > should be calculated following TAR, chap 6: > > RF(CO2) = 5.35 W/m2 ln(CO2(t)/CO2(t0)) > > where CO2(t0=1765) = 280 ppm, or as specified in the input file > > To convert atmospheric concentration into inventories use > 1 ppm = 2.123 GtC > > following Enting et al., 1994 > > > Output Variables > ---------------- > > We only admit global mean averages of quantities, with the exception of > Atlantic meridional overturning circulation (MOC), if available. All > variables are given as annual mean values from year 1765 AD to year > 3000 > AD, or from year 0 to year 3000. Data spacing is 1 year, only. By > annual mean of year 2013 we mean the average from 1.1.2013 to > 31.12.2013. > > All output which does not conform with our requirements will be > rejected. > > 1) Radiative forcing in W/m^2 > > 2) Global mean surface air temperature > > 3) Globally integrated ocean heat uptake > > 4) Global sea level rise due to thermal expansion only. Thermal > expansion is to be calculated from in situ density which depends on in > situ temperature, salinity, and pressure. > > 5) Atlantic meridional overturning circulation, defined as the maximum > of the annual mean of the Atlantic meridional overturning, north of the > equator and below 500 m depth. > > In addition, some EMICs include a carbon cycle model. For those EMICs, > we consider: > > 6) pCO2 (atmosphere) > > 7) total carbon inventory in model system > > 8) carbon inventory of the ocean > > 9) carbon inventory on land (litter, soils, vegetation) > > > > Output Convention > ----------------- > > Output is accepted in ASCII format only. > > comment lines: > > The first lines are comments. Each comment line start with the sign # > > The first line is a number which indicates the number of the following > text lines at the beginning of the file (e.g.: # 32). > > Name, address, e-mail, phone, fax, > any explanation of the model, model references and other comments > should > be included. The last few comment lines hold headers with the variable > names and units > > data: > > Data are then given in ascending years and then organized the following > way; the last columns are for EMICs including a carbon cycle component > > Missing values should be flagged using the number -0.999E99 as flag > values. > > Sufficient digits should be provided to allow for the calculation of > the > first derivative (e.g., ocean heat uptake from ocean heat content) > > Data should be provided in the units indicated below > > The arrangement of columns should be as indicated below. If a variable > is not calculated, please fill the column with flags > > example: > > # 7 > # In this example the first 8 lines are comment lines > # The first line is followed by 7 comment lines > # The last two comment lines indicate variable names and > # units > # > # year radforcing Tair HEATocean(anomaly) RISE MOC > pCO2 CARBONtotal CARBONocean CARBONland > # (year) (W/m^2) (degree C) (Joule) (meters) (Sverdrup) (ppmv) > (GtC) (GtC) (GtC) > > 2020. 7.39 2.345 2.452E22 0.221 0.999E99 12.52 41258. 425.3 > 38.5 > 2021. 7.48 2.567 2.472E22 0.235 -0.999E99 11.73 41265. 430.2 > 40.8 > > > Time Line > --------- > > Submissions of results, including relevant references and descriptions > of the model used should be done until 31. March 2005. You will be able > to submit your results through the internet on > > www.climate.unibe.ch/emicAR4 > > This website will be installed by mid December 2004. > > > -- > ------------------------------------------------------------------ > Thomas Stocker > Climate and Environmental Physics stocker@climate.unibe.ch > Physics Institute, University of Bern phone: +41 31 631 44 62 > Sidlerstrasse 5 fax: +41 31 631 87 42 > 3012 Bern, Switzerland http://www.climate.unibe.ch/~stocker > ------------------------------------------------------------------ > > > > > > > > > > > > > > > > > --Apple-Mail-1-149073115 Content-Transfer-Encoding: 7bit Content-Type: text/enriched; charset=US-ASCII Dear Tom, Please see this draft call for EMICs. I think it will be circulated on Monday so please don't make any comment until then, but here is a pre-view. It is good for us I think since after some discussion they decided to use our SCM officially as the reference. Please forward to me the PDMCI emails you got recently (Tim O sent them to me at UEA but now they are inaccessible on my PC there). I think I will try to register for data, then perhaps I will get emails direct. I don't know what all those stabilization scenarios are going to be from Bern - partly why I send this to you.... Sarah Begin forwarded message: 0000,0000,0000From: Thomas Stocker < 0000,0000,0000Date: 19 November 2004 06:59:35 GMT 0000,0000,0000To: "Raper, Sarah " <, "Raper, Sarah " < 0000,0000,0000Subject: IPCC AR4: Projections using EMICs Dear Sarah Below is the nearly final letter to the community at large. It cristalized that we do not want to call for SCM simulations in this letter but focus on EMICs solely. The SCM simulations should be done exclusively by you using your model to provide the important link across all ARs. We call those "reference SCM simulations". As to the EMICs the process is open, as long as there is a citable model description in the referenced literature. Please have a look at the scenarios, the letter and the formulations. I would like to send this out no later than Monday. Best regards Thomas Dear Colleague As coordinators of Chapter 10 "Global Climate Projections" of the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (AR4), IPCC, we invite you to contribute long-term model simulations using EMICs, Earth System Models of Intermediate Complexity. By EMIC we understand models that are dynamically simpler than comprehensive AOGCMs, although they might well be more "complete" in terms of climate system components that are included. Typically, EMICs are some composite of simplified versions of atmospheric (e.g. energy balance, or quasi-geostrophic), and/or ocean model components (e.g., zonally averaged), and a suite of parameterizations. In order to qualify for participation, a technical description of the EMIC must have appeared in an international, refereed journal. In Chapter 10 of AR4 we will present, in addition to the projections to year 2100 by the most comprehensive AOGCMs, long-term projections out to year 2500 or 3000 using EMICs. The traditional reference simulations with a simplified climate model (SCM) will also be included. The new EMIC simulations have a number of goals: 1. We wish to estimate the range of uncertainty of long-term climate evolution beyond 2100, such as global mean temperature change, sea level rise, ocean heat uptake, Atlantic meridional overturning, etc. 2. Scenarios in addition to SRES should be simulated with models more complex than SCMs, in order to estimate concentration and emissions commitments, i.e. what long-term global changes are expected when either concentrations are kept constant (concentration commitment), or emissions are set to zero (emissions commitment). 3. EMICs ensemble simulations should be used to estimate model-PDFs of certain quantities; 4. Some EMICs allow for the equilibrium climate sensitivity to be tuned. With these, the dependence of the simulated quantities on climate sensitivity should be estimated. Such EMICs must, in addition to simulations with the preferred equilibrium climate sensitivity, present results for equilibrium climate sensitivities of 1.5 degC, 3 degC, and 4.5 degC for a doubling of CO2. 5. Some EMICs exhibit multiple equilibria and transitions to different climate states (e.g., a shut down of the Atlantic MOC). This feature is not simulated by SCMs and therefore, the opportunity exists to investigate the dependence of climate projections on such nonlinear changes. EMICs are highly simplified and parameterized models. They are heterogeneous and no published intercomparison experiments are available currently. Therefore, we will admit only globally averaged quantities in these projections. In order to ensure comparability of different models, all need to simulate a set of well-defined scenarios as a minimum requirement to be included in Chapter 10. Furthermore, we encourage all groups who have different versions of their EMICs (e.g., with different ocean mixing parameterisations, or different atmospheric parameter settings) to submit results for each model version. This allows us to assess the robustness of results with respect to different model parameterisations. Due to space limitations, we cannot guarantee that all results submitted to us will be included in Chapter 10. This solicitation of EMIC results will give us, for the first time, to assess the range of uncertainty across the entire model hierarchy. We trust that this will be a novel and unique contribution to the IPCC process. If you have any queries to this call, or identify issues which require clarification, or any other comments, we look forward to hear from you. Thank you for your efforts. Best wishes Thomas Stocker & Gerald Meehl Coordinating Lead Authors Chapter 10 ************************************************************************ INSTRUCTIONS FOR THE CONTRIBUTION OF EMIC MODEL RESULTS TO THE FOURTH ASSESSMENT REPORT OF IPCC WORKING GROUP 1 (Chapter 10) ************************************************************************ Documentation ------------- Describe your model in one paragraph, adressing the different climate system components it includes. Also provide information according to the entries of Table 8.1 (http://www.grida.no/climate/ipcc_tar/wg1/316.htm) of the Third Assessment Report of IPCC. Required: give reference to papers in refereed journals which describe the model set up and characteristics. Control Simulations ------------------- In order to quantify climate drift, we require a control run from all models with CO2 held constant at 280 ppm, with an integration time of 1000 years. Output as specified below. Scenarios --------- 1. Idealized stabilization scenarios to 2xCO2 ............................................. Note: all CO2 concentrations are referred to "pre-industrial" with pCO2 = 280 ppmv. Integration time 0 to 3000 years 1a. IS2x0.5: Idealized Stabilization Scenario 0.5%/yr to 2xCO2, then constant 1b. IS2x1.0: Idealized Stabilization Scenario 1%/yr to 2xCO2, then constant 1c. IS2x2.0: Idealized Stabilization Scenario 2%/yr to 2xCO2, then constant 2. Idealized stabilization scenarios to 4xCO2 ............................................. Integration time 0 to 3000 years 2a. IS4x0.5: Idealized Stabilization Scenario 0.5%/yr to 4xCO2, then constant 2b. IS4x1.0: Idealized Stabilization Scenario 1%/yr to 4xCO2, then constant 2c. IS4x2.0: Idealized Stabilization Scenario 2%/yr to 4xCO2, then constant 3. Idealized reduction scenarios after increase to 4xCO2 ........................................................ Integration time 0 to 3000 years 3a. IS4x0.5R1.0: Idealized Reduction Scenario: 0.5%/yr to 4xCO2, then -1%/yr to 1xCO2, then constant 3b. IS4x1.0R1.0: Idealized Reduction Scenario: 1%/yr to 4xCO2, then -1%/yr to 1xCO2, then constant 3c. IS4x2.0R1.0: Idealized Reduction Scenario: 2%/yr to 4xCO2, then -1%/yr to 1xCO2, then constant 4. Smooth stabilization profiles ................................ profiles will be available from www.climate.unibe.ch/emicAR4. They are updates of WRE Scenarios, with observed CO2 until 2003 Integration time 1765 to 3000 4a. SP450 4b. SP550 4c. SP650 4d. SP750 4e. SP1000 4f. DSP450: Stabilization profile to 450 ppm with delayed turning point 4g. DSP550 4h. OSP350: Stabilization profile to 350 ppm with overshoot 4i. OSP450 5. Stabilisation profiles for comparison with PCMDI results ........................................................... For comparison with AOGCM results three additional stabilisation profiles are run: A1B, A2, and B1 which are those considered in CMIP. Data for these scenarios are given in IPCC TAR, Appendix II. Those EMICs with no radiation code will use values from Table II.3.11 Integration time 2000 to 3000 5a. A1B: following SRES scenario A1B to year 2100, then constant 5b. A2: following SRES scenario A2 to year 2100, then constant 5c. B1: following SRES scenario B1 to year 2100, then constant 6. Emission commitment ...................... These simulations are only for EMICs which include a dynamic carbon cycle component. As a requirement, this component must be published in the refereed litereature also. Emissions are calculated based on SP scenarios and the Bern carbon cycle model. They will be available from www.climate.unibe.ch/emicAR4. Time 1765 to 3000 (for EMICs with Carbon Cycle model component only) 6a. SP450EC2100: Emission Commitment Scenario: CO2 Emissions prescribed according to SP450, set to 0 after 2100 6b. SP550EC2100: Emission Commitment Scenario: CO2 Emissions prescribed according to SP550, set to 0 after 2100 6c. SP750EC2100: Emission Commitment Scenario: CO2 Emissions prescribed according to SP750, set to 0 after 2100 6d. SP1000EC2100: Emission Commitment Scenario: CO2 Emissions prescribed according to SP1000, set to 0 after 2100 Conversion to radiative forcing ------------------------------- For those models without a radiation code the radiative forcing of CO2 should be calculated following TAR, chap 6: RF(CO2) = 5.35 W/m2 ln(CO2(t)/CO2(t0)) where CO2(t0=1765) = 280 ppm, or as specified in the input file To convert atmospheric concentration into inventories use 1 ppm = 2.123 GtC following Enting et al., 1994 Output Variables ---------------- We only admit global mean averages of quantities, with the exception of Atlantic meridional overturning circulation (MOC), if available. All variables are given as annual mean values from year 1765 AD to year 3000 AD, or from year 0 to year 3000. Data spacing is 1 year, only. By annual mean of year 2013 we mean the average from 1.1.2013 to 31.12.2013. All output which does not conform with our requirements will be rejected. 1) Radiative forcing in W/m^2 2) Global mean surface air temperature 3) Globally integrated ocean heat uptake 4) Global sea level rise due to thermal expansion only. Thermal expansion is to be calculated from in situ density which depends on in situ temperature, salinity, and pressure. 5) Atlantic meridional overturning circulation, defined as the maximum of the annual mean of the Atlantic meridional overturning, north of the equator and below 500 m depth. In addition, some EMICs include a carbon cycle model. For those EMICs, we consider: 6) pCO2 (atmosphere) 7) total carbon inventory in model system 8) carbon inventory of the ocean 9) carbon inventory on land (litter, soils, vegetation) Output Convention ----------------- Output is accepted in ASCII format only. comment lines: The first lines are comments. Each comment line start with the sign # The first line is a number which indicates the number of the following text lines at the beginning of the file (e.g.: # 32). Name, address, e-mail, phone, fax, any explanation of the model, model references and other comments should be included. The last few comment lines hold headers with the variable names and units data: Data are then given in ascending years and then organized the following way; the last columns are for EMICs including a carbon cycle component Missing values should be flagged using the number -0.999E99 as flag values. Sufficient digits should be provided to allow for the calculation of the first derivative (e.g., ocean heat uptake from ocean heat content) Data should be provided in the units indicated below The arrangement of columns should be as indicated below. If a variable is not calculated, please fill the column with flags example: # 7 # In this example the first 8 lines are comment lines # The first line is followed by 7 comment lines # The last two comment lines indicate variable names and # units # # year radforcing Tair HEATocean(anomaly) RISE MOC pCO2 CARBONtotal CARBONocean CARBONland # (year) (W/m^2) (degree C) (Joule) (meters) (Sverdrup) (ppmv) (GtC) (GtC) (GtC) 2020. 7.39 2.345 2.452E22 0.221 0.999E99 12.52 41258. 425.3 38.5 2021. 7.48 2.567 2.472E22 0.235 -0.999E99 11.73 41265. 430.2 40.8 Time Line --------- Submissions of results, including relevant references and descriptions of the model used should be done until 31. March 2005. You will be able to submit your results through the internet on www.climate.unibe.ch/emicAR4 This website will be installed by mid December 2004. -- ------------------------------------------------------------------ Thomas Stocker Climate and Environmental Physics stocker@climate.unibe.ch Physics Institute, University of Bern phone: +41 31 631 44 62 Sidlerstrasse 5 fax: +41 31 631 87 42 3012 Bern, Switzerland http://www.climate.unibe.ch/~stocker ------------------------------------------------------------------ --Apple-Mail-1-149073115--