Spectral Overlap of Greenhouse Gases with Clouds
Top row: one-year average of variables associated with the longwave cloud radiative effect (CRE) in cloudy conditions during a yearlong run of the global storm resolving model X-SHiELD: a) the difference between cloud top and surface temperature and b) the mass-weighted column relative humidity. Panel c): the spatial distribution of the longwave CRE. Subplots d)-g): spectral all-sky (grey) and clear-sky (blue) brightness temperature at points sampled to show a range of temperature, relative humidity, and CRE values. The Planck function of each curve yields the OLR, while the total CRE is defined as the integrated difference between the clear-sky (blue) and all-sky (grey) OLRs, or the area between the two curves.
The spectral distribution of outgoing longwave radiation (OLR) is determined by which atmospheric constituent first becomes optically thick when viewed from above (CO2, water vapor, or cloud). Using pencil-and-paper models for the spectral variation of OLR, we disentangle the effects of surface temperature, cloud top temperature, and relative humidity on Earth’s longwave energy balance. Our expressions yield understanding of physical phenomena, illustrating that the cloud radiative effect (the difference in OLR in an atmospheric column with and without a cloud) is linear in the difference between surface and cloud top temperature because clouds and gases share the longwave spectrum. Additionally, changes in water vapor and CO2 emission temperature can provide a stabilizing response to warming surface temperatures (cloudy-sky feedback) even in cloudy columns with unchanging clouds.
Pre-print: Paulina Czarnecki and Robert Pincus. ESSOAR.
Science Team at the WCRP km-scale Hackathon.
Understanding Radiative Forcing by Optically Thin Gases
Comparison of spectrally-integrated pre-industrial to present-day CFC-12 forcing in CFC-12 only atmospheres. Atmospheric conditions (temperature profiles) are taken from the March 1981 mean in ERA5. In panel a), the forcing given by our analytical theory is shown. Panel b) shows the corresponding line-by-line forcing. Finally, panel c) shows the zonal means of the model (red) and LBL (black) for easy direct comparison.
Optically thin greenhouse gases such as the ozone-depleting substances don't cool to space within the atmosphere, yet they still exert a sizeable radiative forcing. In this work, we extend an existing paper-and-pencil approach to CO2 radiative forcing in order to study radiative forcing by optically thin gases, such as CFC-11 and -12, and gases that are not as optically thick as CO2, such as N2O and methane.
Paper: Paulina Czarnecki, Lorenzo M. Polvani, and Robert Pincus, 2025. doi:10.1175/JCLI-D-25-0233.1.
Talk: at the Equilibrium Climate Sensitivity (ECS) Symposium.
Data-Driven Quadrature for Broadband Spectral Integration
Below an example of a spectrum of outgoing longwave radiation, major absorbers are noted. Dots mark the members of three independent, optimal wavenumber subsets. The colors correspond to these independent optimizations. The size of the dots is proportional to the importance of that spectral point to the broadband flux integration.
Integrating radiative flux across the electromagnetic spectrum, though integral to weather and climate modelling, is computationally expensive; existing parameterizations are fast and accurate but are difficult to understand and implement. As an alternative, we propose Data-Driven Quadrature: we optimize a small set of spectral points and associated weights so that the weighted sum of the monochromatic flux at these points is equal to the total integrated flux. Our method can accurately reproduce flux and heating rate profiles in present-day conditions and radiative forcing by CO2 in the longwave. We are extending the method to the shortwave as well as variability by a wide range of greenhouse gases.
Paper: Paulina Czarnecki, Lorenzo M. Polvani, and Robert Pincus, 2023. doi:10.1029/2023MS003819.
Python package: code to perform your own quadrature optimization, wrapped into a package by Neal Ma. github | pypi
In konrad: implemented as an optional shortwave and longwave scheme in single-column radiative convective equilibrium model konrad.
github | pypi | konrad + DDQ documentation
In ARTS: released as a fast scheme alongside line-by-line radiative transfer code ARTS. github | website | paper | Python module
Pre-Doctoral Work
During the American Institute of Mathematics' workshop on dynamics and COVID-19, I worked with several other graduate students to investigate the statistical relationship between air pollutants and COVID-19 transmission.
Paper: Laura Albrecht, Paulina Czarnecki, and Bennet Sakelaris, 2021. doi:10.6339/21-JDS1010
As an undergraduate student, I worked with Michal Zochowski and his research group to study dynamics in a neuronal network.
Paper: Paulina Czarnecki, Jack Lin, Sara Aton, and Michal Zochowski, 2021. doi:10.3389/fnetp.2021.759131