Andrea Dittus (Univ. of Reading): Regional impacts of climate stabilisation at different global warming levels
Shang-Ping Xie (Scripps Institute of Oceanography): Joint low cloud/WES feedback as a conduit to tropical SST patterns
Casey Wall (Univ. of Oslo): Assessing effective radiative forcing from aerosol–cloud interactions over the global ocean
09:18:22 From Xu, Yangyang : You made a quick comment on reduced OHC uptake in the long-run. Is that related to your runs or ZECmip? Thanks!!
09:22:11 From Paulo Ceppi : We have a related paper (led by G Zappa) finding that SST patterns and associated circulation changes are indeed important – though we focused on subtropical climates. https://www.pnas.org/doi/abs/10.1073/pnas.1911015117
09:23:01 From Andrea Dittus : Thanks for the question! My understanding is that it is the case for both, but in ZECMIP this effect is roughly compensated by the decline in atmospheric CO2 concentrations
09:23:34 From Geeta Persad : Question for Andrea: Given the desire for the GWL framing to be policy-responsive, it does seem problematic that it ignores the ways in which different trajectories of non-CO2 forcers create different regional impacts at the same nominal GWL. Could you use the analysis of the SSP 3-7.0 vs. SSP 2-4.5 runs that you did to provide guidance on the variables or regions for which the GWL approach should be treated with particular caution?
09:23:36 From Phil Rasch : @Andrea, I missed the first couple minutes of the presentation, so you may have presented it, but were the simulations done with ensembles? Is natural internal variability an issue? Apologies if you covered this.
09:26:58 From Mark Zelinka : @Geeta: I have a student working on this exact question. Would love to chat with you about it some time!
09:27:14 From Andrea Dittus : @Paulo It’s a cool paper - normally I try to mention it if I have time, sorry I didn’t manage this time!
09:27:54 From Paulo Ceppi : No worries Andrea – just mentioning it because it was relevant to Karsten’s question!
09:29:13 From Andrea Dittus : @Geeta: Yes possibly! Would need to be careful to make sure that the regional differences are the same for transient and stabilisation runs
09:31:56 From Andrea Dittus : @Phil: We have 6x 500 year runs, with each of the 6 runs branched-off from a different point in the transient runs. I think we capture a lot of internal variability in 500 years, but there could definitely still be an effect of internal variability, particularly on shorter time scales
09:34:11 From Robert Wood : @Casey: Does the inclusion of SO4 in the regression change the conclusions about the role of the other (meteorological) predictors?
09:36:26 From Robert Wood : @Casey: or in other words, how much additional skill does the addition of SO4 add to the meteorological CCFs?
09:36:33 From Larry Di Girolamo : @Casey: So the MODIS/CERES cloud retrievals carry substantial bias, which on a monthly basis, varies with region and season. See https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2019JD031063, which has recently been validated with field data in https://acp.copernicus.org/articles/22/8259/2022/. Keep in mind that CERES and MODIS cloud property retrievals use essential the same method and both applied to MODIS, so CERES and MODIS cloud properties are not independent datasets. These biases would certainly impact that values of ERF_ACI.
09:40:41 From Paquita Zuidema (she/her) : Hi Casey - how do you know the cloud response is to SO4 and not another aerosol that the SO4 is co-varying with? I am primarily thinking of smoke in the southeast Atlantic and dust in the north equatorial Atlantic. Smoke emissions will also generate sulfate. I joined late so feel free to ignore Q if already addressed. One approach is to ignore months w substantial other aerosol emissions but I am not sure how well that works with dust.
09:40:48 From Jonah Bloch-Johnson : @Casey it looks like in the extratropical Pacific, the LWP has the opposite relation with log_10 s than in the tropics/subtropics, at least in CERES - is there a known reason for this?
09:40:51 From Lazaros Oreopoulos : @Casey: Is it possible that the MERRA-2 910 hPa sulfate concentration has inherent biases that depend on cloud fraction (starting point is assimilation of cloudless MODIS pixel radiances).
09:41:07 From Clare Singer (she/her) : @Casey: It seems like your ERFaci estimate is more certain (narrower), but your ECS estimate has not narrowed. Why is that?
09:41:07 From Geeta Persad : Thanks for the interesting talk, Casey! How much does the spread in the gamma values from CMIP6 models contribute to the size of the confidence interval in your ERF_aci estimate relative to other sources of uncertainty?
09:41:46 From Larry Di Girolamo : @Casey: Also consider the following in mind for the low cloud fraction estimate from MODIS/CERES. Cloud detection of low clouds over ocean are largely trigger by the visible channel test, which uses a fixed threshold over ocean for a given sun-view geometry. If aerosols increase and the thresholds remain fixed, more clouds would be detected.
09:42:46 From Geeta Persad : @Mark — just saw your response to my question to Andrea. Yes, we should definitely discuss! This is something we’re very interested in as part of the Regional Aerosol Model Intercomparison Project. Would be exciting to see if those runs could be useful to your student when they become available (we’ve got runs available from some models already, but not yet CMORized, etc.): https://gmd.copernicus.org/preprints/gmd-2022-249/
09:45:26 From virendra ghate : @Casey, thanks for an interesting talk. You showed in stratocumulus regions, reduction in re, reduction in LWP and an increase in cloud fraction due to an increase in aerosols. The reduction in LWP and increase in cloud fraction seem to be at odds. Given entrainment adjustments, we should have decrease in LWP with smaller re, but that should also decrease cloud fraction? Sorry if I am missing something basic.
09:46:24 From Mark Zelinka : Thanks for the info Geeta, will reach out to you….
10:10:40 From Geeta Persad : Thanks, Casey! I’m not sure whether to be comforted by that or not given how much spread there is in the CMIP6 ERF_aci! Maybe the ratio of global to ocean ERF_aci has less spread among the CMIP6 models than the total ERF_aci?