Quantifying sensitivities of ice crystal number and sources of ice crystal number variability in CAM 5.1 using the adjoint of a physically based cirrus formation parameterization

• Published in: Journal of geophysical research. Atmospheres 2015-04, Vol.120 (7), p.2834-2854
• Main Authors: Sheyko, B. A., Sullivan, S. C., Morales, R., Capps, S. L., Barahona, D., Shi, X., Liu, X., Nenes, A.
• Format: Article
• Language: English
• Subjects: adjoint; Adjoints; Aerosols; Atmospheric research; cirrus; climate model; Coarsening; Crystals; Dust; Geophysics; heterogeneous nucleation; homogeneous nucleation; Ice; Ice crystals; Nucleation; Parametrization; Sulfates; Tropical environments; Tropics; Variability; Velocity
• ISSN: 2169-897X; 2169-8996
• DOI: 10.1002/2014JD022457

We present the adjoint of a cirrus formation parameterization that computes the sensitivity of ice crystal number concentration to updraft velocity, aerosol, and ice deposition coefficient. The adjoint is driven by simulations from the National Center for Atmospheric Research Community Atmosphere Model version 5.1 CAM 5.1 to understand the sensitivity of formed ice crystal number concentration to 13 variables and quantify which contribute to its variability. Sensitivities of formed ice crystal number concentration to updraft velocity, sulfate number, and is sufficient but sulfate number concentration is low, indicating a sulfate‐limited regime. Outside of the tropics, competition between homogeneous and heterogeneous nucleation may shift annually averaged sensitivities to higher magnitudes, when infrequent strong updrafts shift crystal production away from purely heterogeneous nucleation. Outside the tropics, updraft velocity is responsible for approximately 52.70% of the ice crystal number variability. In the tropics, sulfate number concentration and updraft jointly control variability in formed crystal number concentration. Insoluble aerosol species play a secondary, but still important, role in influencing the variability in crystal concentrations, with coarse‐mode dust being the largest contributor at nearly 50% in certain regions. On a global scale, more than 95% of the temporal variability in crystal number concentration can be described by temperature, updraft velocity, sulfate number, and coarse‐mode dust number concentration. Key Points A cirrus parameterization adjoint is developed The adjoint is used for ice crystal variability attribution in the CAM 5.1 Ninety‐seven percent of variability is from updrafts, sulfate, and dust number and temperature