Table of Contents
Go to ⇒ [ PI ] - [ 6ka ] - [ 21ka ] - [ LM ] - [ 8.2ka ] - [ LI ] - [ Plio ]
Last Interglacial Experimental Design
Make sure you use the final version of the LI BC!
Goal
Run early Eemian, mid-Eemian and glacial inception snap shots. This will allow us to model the high latitude response to obliquity maximum (130ka) and minimum (115ka), and near precession optimum for ITCZ shift, ENSO and tropical monsoons (125ka)
Core experiments
130ka, 125ka and 115ka time slices with ocean-atmosphere models. If only one Last Interglacial experiment is carried out, 130ka is the top priority.
Extra experiments
- 128ka
- repeat core and 128ka experiments with equivalent ocean-atmosphere-vegetation models
- repeat 130ka experiment with a higher value of CO2, to account for possible dating uncertainties and to assess the impact of the relatively low CO2 in the core experiment. (contact Bette Otto-Bliesner if interested)
- Ice-sheet reconstruction sensitivity
- Reduced Greenland ice-sheet, possibly using the model reconstruction of Otto-Bliesner et al (2006) (contact Bette Otto-Bliesner if interested).
- Reduced West Antarctic ice-sheet (contact Bette Otto-Bliesner if interested).
- transient interglacial simulations: 130-125ka (or 130-115ka). Either low resolution model or GCM with accelerated boundary conditions, using transient orbit and trace gases (contact Dan Lunt if interested).
- transient holocene simulations: 8-2ka (or 8-0ka). Either low resolution model or GCM with accelerated boundary conditions, using transient orbit and trace gases (contact Dan Lunt if interested).
- transient and snapshot simulations with variable solar forcing (contact Hans Renssen and/or Steven Phipps if interested)
Boundary conditions
130ka (CORE - TOP PRIORITY)
| PMIP3 | Alternative solution | |
|---|---|---|
| Orbital parameters | [ ecc = 0.038209 ] - [ obl = 24.242° ] - [ peri-180° = 228.32° ] | |
| Date of vernal equinox | March 21 at noon | |
| Trace gases | [ CO2 = 257 ppm ] - [ CH4 = 512 ppb ] - [ N2O = 239 ppb ] - [ CFC = 0 ] - [ O3 = Same as PI ] | |
| Aerosols | Same as PI | |
| Solar constant | 1365 W/m2 | As in PI |
| Vegetation | Pre-industrial fixed | |
| Ice sheets | Same as PI | |
| Topography and coastlines | Same as PI |
Please use the discussion panel to comment this table!
128ka (EXTRA)
| PMIP3 | Alternative solution | |
|---|---|---|
| Orbital parameters | [ ecc = 0.039017 ] - [ obl = 24.131° ] - [ peri-180° = 259.65° ] | |
| Date of vernal equinox | March 21 at noon | |
| Trace gases | [ CO2 = 275 ppm ] - [ CH4 = 709 ppb ] - [ N2O = 266 ppb ] - [ CFC = 0 ] - [ O3 = Same as PI ] | |
| Aerosols | Same as PI | |
| Solar constant | 1365 W/m2 | As in PI |
| Vegetation | Pre-industrial fixed | |
| Ice sheets | Same as PI | |
| Topography and coastlines | Same as PI |
Please use the discussion panel to comment this table!
125ka (CORE)
| PMIP3 | Alternative solution | |
|---|---|---|
| Orbital parameters | [ ecc = 0.040013 ] - [ obl = 23.798° ] - [ peri-180° = 307.14° ] | |
| Date of vernal equinox | March 21 at noon | |
| Trace gases | [ CO2 = 276 ppm ] - [ CH4 = 640 ppb ] - [ N2O = 263 ppb ] - [ CFC = 0 ] - [ O3 = Same as PI ] | |
| Aerosols | Same as PI | |
| Solar constant | 1365 W/m2 | As in PI |
| Vegetation | Pre-industrial fixed | |
| Ice sheets | Same as PI | |
| Topography and coastlines | Same as PI |
Please use the discussion panel to comment this table!
115ka (CORE)
| PMIP3 | Alternative solution | |
|---|---|---|
| Orbital parameters | [ ecc = 0.041421 ] - [ obl = 22.405° ] - [ peri-180° = 110.88° ] | |
| Date of vernal equinox | March 21 at noon | |
| Trace gases | [ CO2 = 273 ppm ] - [ CH4 = 472 ppb ] - [ N2O = 251 ppb ] - [ CFC = 0 ] - [ O3 = Same as PI ] | |
| Aerosols | Same as PI | |
| Solar constant | 1365 W/m2 | As in PI |
| Vegetation | Pre-industrial fixed | |
| Ice sheets | Same as PI | |
| Topography and coastlines | Same as PI |
Please use the discussion panel to comment this table!
Transient 130ka-125ka (EXTRA)
| PMIP3 | Alternative solution | |
|---|---|---|
| Orbital parameters | varying, model interpolates values given on a 1k resolution. See values here: http://www.paleo.bris.ac.uk/~ggdjl/pmip/orbit_pmip.out Values are given for 132-115ka. | |
| Date of vernal equinox | March 21 at noon | |
| Trace gases | varying, model interpolates values given on a 0.1k resolution. See values here: http://www.paleo.bris.ac.uk/~ggdjl/pmip/pmip_hol_lig_gases.txt Values are given for 132-115ka. | |
| Aerosols | Same as PI | |
| Solar constant | 1365 W/m2 | As in PI |
| Vegetation | Pre-industrial fixed | |
| Ice sheets | Same as PI | |
| Topography and coastlines | Same as PI |
Please use the discussion panel to comment this table!
Transient 8ka-2ka (EXTRA)
| PMIP3 | Alternative solution | |
|---|---|---|
| Orbital parameters | varying, model interpolates values given on a 1k resolution. See values here: http://www.paleo.bris.ac.uk/~ggdjl/pmip/orbit_pmip.out Values are given for 10-0ka. | |
| Date of vernal equinox | March 21 at noon | |
| Trace gases | varying, model interpolates values given on a 0.1k resolution. See values here: http://www.paleo.bris.ac.uk/~ggdjl/pmip/pmip_hol_lig_gases.txt Values are given for 10-0ka. | |
| Aerosols | Same as PI | |
| Solar constant | 1365 W/m2 | As in PI |
| Vegetation | Pre-industrial fixed | |
| Ice sheets | Same as PI | |
| Topography and coastlines | Same as PI |
Please use the discussion panel to comment this table!
Notes
- Trace gas information was obtained from ice core data, all on the EDC3 chronology. For more information (including figures) and details (including IDL code) of the interpolation, please see the webpage below (including a readme file)
http://www.paleo.bris.ac.uk/~ggdjl/pmip
- Orbital values from Berger (1978). Calculated using B78 fortran code downloadable from http://www.astr.ucl.ac.be/index.php?page=AstronomicalInsolationForcing
- Some models may calculate their own orbital parameters internally, which is fine.
- For accelerated simulations, the trace gases and orbits are identical to the non-accelerated simulations, but have a smaller timestep.
Initial conditions
For snapshots:
| PMIP3 | Alternative solution | |
|---|---|---|
| Initial conditions | Branch off PI after adjustment | Same as in PI |
| Model spin up | Same as in PI |
For transients:
| PMIP3 | Alternative solution | |
|---|---|---|
| Initial conditions | Branch off PI after adjustment | Same as in PI |
| Model spin up | 2kyr or less before transient starts | Same as in PI |
For initial conditions, and especially spinup, the key thing is to document what you do!
For the transient simulations, orbits and trace gas values are given for up to 2ky before the start date of the simulation.
Participating groups
The following groups intend to perform experiments for the Eemian interglacial sub-project
BRIDGE, Bristol, UK
People involved: Joy Singarayer, Paul Valdes, Dan Lunt
We plan to use HadCM3 and HadGEM2 (the version being used for IPCC simulations) to perform the core interglacial OA experiments. HadCM3 will be used for the optional further snap-shot simulations, and FAMOUS for transient Eemian simulations
[ PMIP3 Wiki Home ] - [ Help! ] - [ Wiki syntax ] - [ Top ]
Discussion
Climate “optimums” of MIS 5e & 11 correspond to intervals with a significantly reduced (5e) or nearly vanished (11) Greenland ice-sheet, thus with likely a more zonal North Atlantic circulation. In addition, MIS 11 has been long enough to lead to near-equilibrium state of major elements of the Earth surface system (notaby relative sea-levels). Should'nt we focus some work on “the Earth's climate without Greenland ice sheet” ?
Claude Hillaire-Marcel
Yes, agreed. this is one of the proposed 'extra' experiments. If enough of the PMIP community are interested in doing a coordinated experiment without Greenland then this can be done.
To Joy/Daniel/Paul
I compared the values of the orbital parameters, i.e. those that you propose and those that I computed … And they are different.
Could you check what happened? Marie-France
This has now been corrected - thanks for spotting this!
Our orbital parameters for the NCAR CCSM simulations agree with those of Marie-France for 130 ka and 125ka.
I like Claude's idea of looking at the Greenland Ice Sheet sensitivity to the 5e forcing. The magnitude of the Greenland retreat during this interval has still to be precisely constrained (i.e., ice-sheet models show a large spread in predictions, interpretations of the ice-core data are debated, and the Kopp et al. 2009 Nature paper has more sea-level rise coming from Antarctica than Greenland during 5e). Could various ice-sheet extents be tested under say 125 ka forcing (a PI, a slightly reduced and a greatly reduced) to see which would be most “accurate” as part of the runs?
We made a transient Holocene (9.5ka-0ka) experiement using coupled ocean-atmosphere model with forcing of varing orbital parameters by 10 times orbital accelaration design. Are there any difference between the run with orbital accelaration design and the run with model interpolated values given on a 1k resolution? Will it be better using the orbital accelaration design for Holocene transient run?
This is a good point. For the accelerated runs, the only difference is that the user will use the same boundary conditions but accelerated. I will put a note on the experiment description about this.
For the transient experiments, the option could/should be given of allowing the models to calculate the orbital parameters themselves (subject to some check that they are able to do so accurately). Many models are able to do this.
For the Holocene transient, more realistic boundary conditions could be employed. In particular, reconstructions of solar irradiance exist that span the entire period of the experiment e.g. Steinhilber et al (2009), which is one of the standard reconstructions specified as an option for the Last Millennium experiment. Over the period 8-2ka, changes in trace gases and solar irradiance contribute radiative forcings of similar magnitudes, so it's arguably inconsistent to specify changes in one and not the other.
Yes, agreed - the user can have the option of using their own code, but ideally they should check that their orbital parameters are not far removed from the B78 solution.
My question for the solar forcing is how accurate are the reconstructions (presumably from 10Be). And can we also do this for the last interglacial? I would probably suggest that the inclusion of solar forcing be an optional additional experiment, unless the forcings are known very accurately.
Agree. Given the uncertainties in reconstructions of past solar irradiance, an optional additional experiment for 8-2 ka that includes solar forcing is arguably the best solution.
The values of orbital parameters proposed here (1) are different from two other resources (2) and (3):
(1). PMIP3 proposed:
(2). from ftp://ftp.ncdc.noaa.gov/pub/data/paleo/insolation/orbit91
YEAR ECC OMEGA OBL
-115 0.043983 109.54 22.438 -125 0.042308 304.76 23.818 -128 0.041094 257.12 24.142 -130 0.040129 225.73 24.247
(3) from http://aom.giss.nasa.gov/srorbpar.html
Year Ecc Obliquity Long. of Perihel. (A.D.) (degrees) (degrees) —— ——– ——— ——– -130000 .037319 24.2512 17.948 -128000 .038231 24.2405 49.101 -125000 .039395 24.0350 96.198 -115000 .041373 22.5869 258.350
Does anybody can help to explain this difference? Thanks!
Liya Jin
(1) and (3) are strictly the same, only the dates are different. For example 130kaBP corresponds to -128050 AD. The orbital parameters are computed according to Berger (1978) for (1) and (3) and according to Berger and Loutre (1991) for (2). Other astronomical solutions are available, e.g. Laskar et al (2004). I hope this will help you. Marie-France
Thanks for your explanation.
We also had a request to put up orbital and trace gas values back to 10k in the holocene and 132k in the eemain, so that people could spin up the model with something fairly realistic (although not that realistic as the ice sheets begin to change significantly in these windows)
This has been done.