Three flights from the Ice in Clouds Experiment–Tropical (ICE-T) field campaign examined the onset of ice near the ascending cloud tops of tropical maritime cumuli as they cooled from 0° to −14°C. Careful quantitative analysis of ice number concentrations included manual scrutiny of particle images and corrections for possible particle-shattering artifacts. The novel use of the Wyoming Cloud Radar documented the stage of cloud development and tops relative to the aircraft sampling, complemented the manual estimates of graupel concentrations, and provided new clear evidence of graupel movement through the rime-splintering zone. Measurements of ice-nucleating particles (INPs) provided an estimate of primary initiated ice. The data portray a dynamically complex picture of hydrometeor transport contributing to, and likely resulting from, the rime-splintering process. Hundreds per liter of supercooled raindrops ascended within the updrafts as the cloud tops reached 0°C and contributed in part to the 0.1 L−1 graupel detected soon after the cloud tops cooled to −5°C. Rime splintering could thus be initiated upon first ascent of the cloud top through that zone and arguably contributed to the 1 L−1 or more graupel observed above it. Graupel ascending/descending into, or balanced within, the rime-splintering zone were found. In wider, less isolated clouds with dying updrafts and tops near −14°C, ice particle concentrations sometimes reached 100 L−1. Future 3D numerical modeling will be required to evaluate if rime splintering alone can explain the difference of three to four orders of magnitude in the observed INPs and the graupel observed at −5°C and colder.

Abstract. Cirrus cloud thinning (CCT) is a relatively new radiation management proposal to counteract anthropogenic climate warming by targeting Earth's terrestrial radiation balance. The efficacy of this method was presented in several general circulation model (GCM) studies that showed widely varied radiative responses, originating in part from the differences in the representation of cirrus ice microphysics between the different GCMs. The recent implementation of a new, more physically based ice microphysics scheme (Predicted Particle Properties, P3) that abandons ice hydrometeor size class separation into the ECHAM-HAM GCM, coupled to a new approach for calculating cloud fractions that increases the relative humidity (RH) thresholds for cirrus cloud formation, motivated a reassessment of CCT efficacy. In this study, we first compared CCT sensitivity between the new cloud fraction approach and the original ECHAM-HAM cloud fraction approach. Consistent with previous approaches using ECHAM-HAM, with the P3 scheme and the higher RH thresholds for cirrus cloud formation, we do not find a significant cooling response in any of our simulations. The most notable response from our extreme case is the reduction in the maximum global-mean net top-of-atmosphere (TOA) radiative anomalies from overseeding by about 50 %, from 9.9 W m−2 with the original cloud fraction approach down to 4.9 W m−2 using the new cloud fraction RH thresholds that allow partial grid-box coverage of cirrus clouds above ice saturation, unlike the original approach. Even with this reduction with the updated cloud fraction approach, the TOA anomalies from overseeding far exceed those reported in previous studies. We attribute the large positive TOA anomalies to seeding particles overtaking both homogeneous nucleation and heterogeneous nucleation on mineral dust particles within cirrus clouds to produce more numerous and smaller ice crystals. This effect is amplified by longer ice residence times in clouds due to the slower removal of ice via sedimentation in the P3 scheme. In an effort to avoid this overtaking effect of seeding particles, we increased the default critical ice saturation ratio (Si,seed) for ice nucleation on seeding particles from the default value of 1.05 to 1.35 in a second sensitivity test. With the higher Si,seed we drastically reduce overseeding, which suggests that Si,seed is a key factor to consider for future CCT studies. However, the global-mean TOA anomalies contain high uncertainty. In response, we examined the TOA anomalies regionally and found that specific regions only show a small potential for targeted CCT, which is partially enhanced by using the larger Si,seed. Finally, in a seasonal analysis of TOA responses to CCT, we find that our results do not confirm the previous finding that high-latitude wintertime seeding is a feasible strategy to enhance CCT efficacy, as seeding in our model enhances the already positive cirrus longwave cloud radiative effect for most of our simulations. Our results also show feedbacks on lower-lying mixed-phase and liquid clouds through the reduction in ice crystal sedimentation that reduces cloud droplet depletion and results in stronger cloud albedo effects. However, this is outweighed by stronger longwave trapping from cirrus clouds with more numerous and smaller ice crystals. Therefore, we conclude that CCT is unlikely to act as a feasible climate intervention strategy on a global scale.

Abstract. Cirrus cloud thinning (CCT) is a relatively new radiation management proposal to counteract anthropogenic climate warming by targeting Earth’s terrestrial radiation balance. The efficacy of this method was presented in several general circulation model (GCM) studies that showed widely varied radiative responses, originating in part from the differences in the representation of cirrus ice microphysics between the different GCMs. The recent implementation of a new, more physically based ice microphysics scheme (Predicted Particle Properties, P3) that abandons ice hydrometeor size class separation into the ECHAM-HAM GCM, coupled to a new approach for calculating cloud fractions that increases the relative humidity (RH) thresholds for cirrus cloud formation, motivated a reassessment of CCT efficacy. In this study, we first compared CCT sensitivity between the new cloud fraction approach and the original ECHAM-HAM cloud fraction approach. With the P3 scheme and the higher RH thresholds for cirrus cloud formation, we find a significant cooling response of −0.36 Wm−2 only for our simulation with a seeding particle concentration of 1 L−1, due mostly to rapid cloud adjustments. The most notable response is the reduction of the maximum global-mean net top-of-atmosphere (TOA) radiative anomalies from overseeding by more than 50 %, from 9.0 Wm−2 with the original cloud fraction approach, down to 4.3 Wm−2 using the new cloud fraction RH thresholds by avoiding artificial ice-cloud expansion upon ice nucleation. We attribute the large positive TOA anomalies to seeding particles overtaking both homogeneous nucleation and heterogeneous nucleation on mineral dust particles within cirrus clouds to produce more numerous and smaller ice crystals. This effect is amplified by longer ice residence times in clouds due to the more realistic, slower removal of ice via sedimentation in the P3 scheme. In an effort to avoid this overtaking effect of seeding particles, we increased the default critical ice saturation ratio (Si,seed) for ice nucleation on seeding particles from the default value of 1.05 to 1.35 in a second sensitivity test. With the higher Si,seed we eliminate overseeding and are able to produce cooling responses over a broader range of seeding particle concentrations, with the largest cooling of −0.32 Wm−2 for a seeding particle concentration of 10 L−1, which suggests that Si,seed is a key factor to consider for future CCT studies. However, the global-mean TOA anomalies contain high uncertainty. In response, we examined the TOA responses regionally and found that specific regions only show a small potential for targeted CCT, which is partially enhanced by using the larger Si,seed. Finally, in a seasonal analysis of TOA responses to CCT, we find that our results do not support the previous finding that high-latitude wintertime seeding is a feasible strategy to enhance CCT efficacy, as seeding in our model enhances the already positive cirrus longwave cloud radiative effect. Instead, our results show that summertime cooling occurs due to adjustments of lower-lying mixed-phase and liquid clouds. Therefore, we conclude that CCT is unlikely to act as a feasible climate intervention strategy on a global scale, and should be investigated further with higher-resolution studies in potential target regions and with studies dedicated to assessing potentially realistic seeding particle materials.

The Hamburg Aerosol Module version 2.3 (HAM2.3) from the ECHAM6.3‐HAM2.3 global atmosphere‐aerosol model is coupled to the recently developed icosahedral nonhydrostatic ICON‐A (icon‐aes‐1.3.00) global atmosphere model to yield the new ICON‐A‐HAM2.3 atmosphere‐aerosol model. The ICON‐A and ECHAM6.3 host models use different dynamical cores, parameterizations of vertical mixing due to sub‐grid scale turbulence, and parameter settings for radiation balance tuning. Here, we study the role of the different host models for simulated aerosol optical thickness (AOT) and evaluate impacts of using HAM2.3 and the ECHAM6‐HAM2.3 two‐moment cloud microphysics scheme on several meteorological variables. Sensitivity runs show that a positive AOT bias over the subtropical oceans is remedied in ICON‐A‐HAM2.3 because of a different default setting of a parameter in the moist convection parameterization of the host models. The global mean AOT is biased low compared to MODIS satellite instrument retrievals in ICON‐A‐HAM2.3 and ECHAM6.3‐HAM2.3, but the bias is larger in ICON‐A‐HAM2.3 because negative AOT biases over the Amazon, the African rain forest, and the northern Indian Ocean are no longer compensated by high biases over the sub‐tropical oceans. ICON‐A‐HAM2.3 shows a moderate improvement with respect to AOT observations at AERONET sites. A multivariable bias score combining biases of several meteorological variables into a single number is larger in ICON‐A‐HAM2.3 compared to standard ICON‐A and standard ECHAM6.3. In the tropics, this multivariable bias is of similar magnitude in ICON‐A‐HAM2.3 and in ECHAM6.3‐HAM2.3. In the extra‐tropics, a smaller multivariable bias is found for ICON‐A‐HAM2.3 than for ECHAM6.3‐HAM2.3.

Abstract. To date the climate intervention (CI) proposal of cirrus cloud thinning (CCT) was only assessed in general circulation models (GCMs) using a globally uniform distribution of artificial ice nucleating particles (INPs). In this study, we made the first attempt using the ECHAM-HAM GCM to simulate CCT using a fully prognostic cirrus seeding aerosol species. Seeding particles were assumed to be made of bismuth-triiodide and were emitted into the atmosphere following aircraft emissions of black carbon (soot). This new approach drastically reduced the number concentration of seeding particles available as INPs in our cirrus ice nucleation sub-model compared to the globally uniform approach. As a result, we found that in order to achieve a significant signal we needed to reduce the assumed radius of emitted seeding particles by an order of magnitude to 0.01 μm and scale the mass emissions of seeding particles by at least a factor of 100 or 1000. This latter scaling factor led to a large net TOA warming effect of 5.9 Wm-2. This warming effect was a clear response to overseeding with a large concentration of seeding particles (> 105 L-1 in the northern hemisphere) that was most evident in the tropics. Due to this undesired effect, in a second series of simulations we avoided seeding the tropics by restricting emissions to only the northern hemisphere (NH) during winter. We also found a small and insignificant effect, or overseeding, which for the extreme case was reduced compared to the global aircraft emission scenario (2.2 Wm-2). Ice crystal radius anomalies were not what we expected, with the largest reduction in size found for the case with a mass scaling factor of 10 instead of the extreme, x1000, scenario. We attributed this peculiar behavior to the differences in the competition between different seeding particle concentrations and background particles. Finally, we also found that seeding with such large concentrations increased the albedo effect of mixed-phase clouds in the NH due to less efficient cloud droplet consumption, consistent with previous findings from our model. Overall, however, based on this study it is recommended to pause further modeling efforts of CCT unless more observational-based evidence of aerosol-ice-cloud interactions indicates favourable conditions for producing the desired outcome of this CI proposal.