#regional_aerosols
#aerosols_fast_response
#aerosols_slow_response
#precipitation
#temperature
#apparent_hydrologic_sensitivity
#energy_budget
#atmosphere_dynamics
#atmosphere_only_simulations
#atmosphere_ocean_coupled_simulations
#perturbed_simulations
#idealized_simulations

The PDRMIP uses idealized experiments that involve large increases in GHGs and aerosols to investigate the fast and slow responses of precipitation to aerosols.

Some standard facts in the aerosol response literature:

(1) Aerosols effects are complicated because (i) aerosols have short life span and therefore creates larger loadings when and where they are emitted, rather than being globally uniformly distributed like the greenhouse gases, (ii) aerosols create both direct and indirect effects, the latter of which is especially complicated and diverse, (iii) different species of aerosols do not have the same effects (e.g. sulfate cools, black carbon warms)

(2) Aerosols induce fast (i.e. within a few years) and slow (i.e. after several decades) responses

(2.1) Fast responses stem from atmospheric and land surface interactions

(2.2) Fast responses scale with global mean atmospheric absorption

(2.3) Fast responses can be investigated using atmosphere-only simulations

(2.4) Slow responses stem from atmosphere-ocean interactions

(2.5) Slow responses scale with global mean surface temperature

(2.6) Slow responses must be investigated with coupled simulations

Simulation experiments analyzed in this study

(1) SULASIA - 10 times present-day sulfate concentrations over Asia

(2) SULEUR - 10 times present-day sulfate concentrations over Europe

(3) BCASIA - 10 times present-day black carbon concentrations over Asia

(4) Control - all aerosol concentrations remain at present-day levels

(5) SO4x5 - 5 times global sulfate concentrations

(6) BCx10 - 10 times global black carbon concentrations

Seven GCMs participated. The perturbed concentrations were introduced as step changes and then kept constant over time. For each GCM, there was a fixed-SST (fSST; i.e. atmosphere-only experiment for assessing the fast response) run of 15 years, and a fully coupled run of 100 years. The last 10 years of the fSST and the last 50 years of the fully coupled runs were used, with the first few years being treated as spin-up. (side note: all models include direct effects of BC and sulfate, and semidirect effects of BC; but only some of the models include the full aerosol indirect effects on cloudes)

Metrics:

(1) Apparent hydrologic sensitivity (AHS): total precipitation change per unit global surface temperature change, in the fully coupled simulations

(2) Precipitation changes: $\Delta$P_{fast} is calculated from the fSST simulations, $\Delta$P_{total} from the coupled simulations, and $\Delta$P_{slow} = $\Delta$P_{total} - $\Delta$P_{fast}

(3) Forcing changes: effective radiative forcing at the top-of-atmosphere (RF_{TOA}) and the surface (RF_{surf}) was calculated from fSST; Net atmospheric absorption (AA) = RF_{TOA} - RF_{surf}

(4) L_c * $\Delta$P = $\Delta$Q + $\Delta$H, latent heat of concensation of water * precipitation = column-integrated diabatic cooling + column-integrated dry static energy flux divergence ($\Delta$Q = $\Delta$LW + $\Delta$SW - $\Delta$SH; H = L_c*P - Q)

Selected Results (the global energy budget and dynamics analysis not included):

(1) Total precipitation responses map: SULASIA results in very clear precipitation declines in the Asian monsoon region and southward displacement of the ITCZ; SULEUR induces precipitation declines in Mediterranean and Sahel-central Africa, and slight southward displacement of the ITCZ; BCASIA causes wetting of the Himalayas, north-wetting and south-drying of China, and northward shift of the ITCZ that result in drying of the tropical Indian and Southeast Asia waters. Strength-wise, the effect of SULEUR is weaker than SULFASIA due to the much lower sulfate loading in Europe than in Asia; the per-unit efficacy is in fact slightly stronger. The effect of BCASIA is much weaker than SULASIA, which is due to lower black carbon mass in the atmosphere than sulfate; the per-unit efficacy of black carbon is in fact stronger. The inter-model uncertainty in the response to black carbon is larger than to sulfate.

(2) Total temperature responses map: SULASIA and SULEUR generates temperature decreases that are especially strong in Asia and Europe, respectively; BCASIA generates temperature increases.

(3) In global total, the sensitivity of temperature to sulfate is negative, to black carbon is positive. The sensitivities of precipitation to sulfate and black carbon are both negative.

(4) The global $\Delta$P_{fast} scales linearly with global AA in the three regional perturbation experiments, and $\Delta$P_{slow} of the global changes scales with global $\Delta$T, which are consistent with past global experiment results. The scaling relationships between the regional $\Delta$P_{fast} and $\Delta$P_{slow} in Asia and Europe and the global forcings have more variability (i.e. scales less well).

(5) The global $\Delta$P_{total} scales linearly with regional RF_{TOA}. The regional responses v.s. regional forcings have slightly more uncertainty but follows a line better than (4).

https://journals.ametsoc.org/view/journals/clim/31/11/jcli-d-17-0439.1.xml