Periodic Reporting for period 1 - SIMPHAC (The impact of Secondary Ice processes on Mixed-PHAse Clouds and Climate)
Berichtszeitraum: 2021-09-01 bis 2023-08-31
Clouds may never have had a more important meaning to society as today. They regulate the Earth's energy balance and are key drivers of how climate responds to changing greenhouse gas levels. Clouds generate precipitation, which has a direct impact on the supply of fresh water on Earth. Clouds however are the most elusive component of the climate system, and the largest source of predictive error in any atmospheric and climate model. Of all cloud types, mixed-phase (liquid water + ice) clouds are by far the most uncertain, and dominate the energy balance and precipitation in many regions of the globe. At the heart of this uncertainty is the inability to capture ice crystal formation and the explosive multiplication that can occur, which in turn fundamentally affect cloud radiative properties and lifetime. This phenomenon is known as Secondary Ice Production (SIP), but the exact mechanisms and their relative importance remain unknown; as a result a description of these processes remains incomplete in weather forecast and climate models.The main goal was to implement accurate mathematical descriptions of these processes in weather forecast and climate models and quantify their impacts on cloud. Our results indicated that from all SIP mechanisms, collisional break-up is the most effective and the inclusion of this mechanism can improve cloud representation at both weather forecasting and climate model scales
During the first year of the project we constrained and implemented two missing SIP mechanisms, collisional break-up and drop-shattering, in the Weather and Research Forecasting model (WRF). As most atmospheric models, WRF accounts only for a single SIP mechanism, the Hallett-Mossop process. The updated code was used to simulate a cold-air outbreak (CAO) observed near UK in 2013; CAOs are severe weather phenomena initiated by the transport of cold air masses from the Arctic to lower latitudes. The changes in the boundary-layer structure that usually take place along the air mass trajectory during CAOs result into changes in the cloud morphology, typically leading to a stratocumulus-to-cumulus transition (SCT). The correct prediction of this transition is critical for the accurate representation of precipitation and radiation patterns. Our results reveale that the default model, accounting only for Hallett-Mossop, fails to reproduce the observed cloud patterns ad the location of the SCT. Drop-shattering appears more effective in the examined cumulus conditions compared to Hallett-Mossop but when this mechanism operates alone, it cannot reproduce the observed range of particle number concentrations and accurately predict the observed SCT location. Collisional break-up is the only SIP mechanism that results in significant ice number enhancement in both stratiform and cumulus regimes. However, the efficiency of this process highly depends on the treatment of rimed fraction for cloud ice and snow. The simulation with highly rimed particles captures the observed SCT location and also reproduces the observed particle number concentration range. Yet, this experiment produced excessive break-up and overestimates the frequency of cloud-free conditions. Simulations with a low prescribed rimed fraction reduced this cloud-break overestimation, but produced somewhat lower particle concentrations and did not capture the observed SCT location. In summary, our study indicated that drop-shattering and collisional break-up are favored during CAOs and can potentially play a more critical role for the SCT than Hallett-Mossop. This highlights the need for more extensive experimental studies of these processes to improve existing parameterizations and better constrain uncertain parameters that affect their efficiency. The results of this study are published at Atmospheric Environment (Karalis et al. 2022) and have been presented to the atmospheric modelling community during the FoRCES annual meeting and during invited talks at educational institutes (e.g. the Karlsruhe Institute of Technology and the University of Oslo).
During the second year of the project we implemented the same processes (collisional break-up and drop-shattering) in the Norwegian Earth Sustem Model version 2 (NorESM2). The new implementations were evaluated against 2-year field observations from Ny-Alesund and satellite retrievals over the whole Arctic region (Sotiropoulou et al 2022). Our results indicate that the inclusion of the missing SIP mechanisms in NorESM2 improves the modeled ice properties, while improvements in liquid content highly depend on the treatment of primary ice production. Overall results are sensitive to the description of collisional break-up, which is the dominant SIP mechanism. Variations in the description of ice aggregation also modulates the efficiency of collisional break-up, with significant impacts on the modelled cloud ice levels. Nevertheless, enhancement in ice production though the addition of SIP mechanisms can result in enhanced cloud cover and decreased radiation biases at the top-of-the-atmosphere (TOA), compared to satellite measurements, especially during the cold months. These results indicate that inclusions of the missing SIP mechanisms in climate models may have significant impact on future projections of the Arctic climate. The results of the study are included in Sotiropoulou et al. (2022), in review for Journal of Climate, and have been presented at various conferences, such as EGU, the Arctic Science Summit Week (QuIESCENT workshop) and the European Aerosol Conference (all in 2022) and during invited talks at educational institutes (e.g. the Karlsruhe Institute of Technology and the University of Oslo) in 2022-2023. Results have also been communicated to the climate model developers of the three main European models (NorESM2, ECHAM and EC-Earth) who aim to update SIP representation in the next climate model generation.
During the second year of the project we implemented the same processes (collisional break-up and drop-shattering) in the Norwegian Earth Sustem Model version 2 (NorESM2). The new implementations were evaluated against 2-year field observations from Ny-Alesund and satellite retrievals over the whole Arctic region (Sotiropoulou et al 2022). Our results indicate that the inclusion of the missing SIP mechanisms in NorESM2 improves the modeled ice properties, while improvements in liquid content highly depend on the treatment of primary ice production. Overall results are sensitive to the description of collisional break-up, which is the dominant SIP mechanism. Variations in the description of ice aggregation also modulates the efficiency of collisional break-up, with significant impacts on the modelled cloud ice levels. Nevertheless, enhancement in ice production though the addition of SIP mechanisms can result in enhanced cloud cover and decreased radiation biases at the top-of-the-atmosphere (TOA), compared to satellite measurements, especially during the cold months. These results indicate that inclusions of the missing SIP mechanisms in climate models may have significant impact on future projections of the Arctic climate. The results of the study are included in Sotiropoulou et al. (2022), in review for Journal of Climate, and have been presented at various conferences, such as EGU, the Arctic Science Summit Week (QuIESCENT workshop) and the European Aerosol Conference (all in 2022) and during invited talks at educational institutes (e.g. the Karlsruhe Institute of Technology and the University of Oslo) in 2022-2023. Results have also been communicated to the climate model developers of the three main European models (NorESM2, ECHAM and EC-Earth) who aim to update SIP representation in the next climate model generation.
Our results have revealed the inclusion of the missing SIP mechanisms, and especially collisional break-up, in atmospheric models can substantially advance the representation of clouds. Including this mechanism in a weather forecasting model resulted in a more accurate prediction of the cloud fields during a CAO case, and particularly of the location of the transition from stratiform to cumulus conditions and the onset of heavy precipitation. Considering that CAOs are often being associated with severe weather events with harmful impacts on vegetation, building infrastructure and even human life, improving their prediction can contribute to improved emergency management. Finally, including these mechanisms in a climate model resulted in improved representation of the cloud properties over the Arctic and decreased TOA radiation biases. As the Arctic is the most climatically sensitive region of the planet, our study reveals that improving SIP descriptions in climate models can potentially improve the projections of the future climate.