Research
Research Interests
Global climate change
Cloud feedbacks
Anthropogenic aerosol forcing
Aerosol–cloud–precipitation interactions
Hydrological cycle
Improvement of climate models using observational constraints
My Personal Projects
Boundary layer entrainment
Entrainment of warm, dry air into the stratocumulus-topped boundary layer is an important control on cloud thickness and extent. One of the major contributors to uncertainty in climate projections is the uncertain sensitivity of the cloud-top entrainment to changes in thermodynamics, dynamics, radiative fluxes, and microphysics. As the characteristic scales of cloud-top turbulence will preclude process modeling of these entrainment sensitivities over large areas for the foreseeable future, model-based projections are marred by numerical artifacts such as "numerical mixing". How large are these artifacts; what are the contributions due to the poorly resolved inversion and due to parameterized cloud processes, convection, and turbulence; and is there spatial resolution coarse enough that regional or larger simulations are feasible but the entrainment is not dominated by numerical artifacts?
If we can understand the source of these artifacts well enough to reduce them, it may become feasible to parameterize the enhanced entrainment in polluted clouds well enough to get a general circulation model (GCM) estimate of the positive effective radiative forcing through aerosol–cloud interactions (ERFaci) adjustments as well as the negative ones. Reducing the artificial entrainment may also lead to a better representation of the stratocumulus base state in GCMs.
Constraining parameterized convection in GCMs
Lateral entrainment into cumulus clouds is a popular tuning knob for parameterized convection that strongly affects the vertical distribution of water vapor and cloud condensate and with it the cloud, water-vapor, and lapse-rate feedbacks. Observational constraints on entrainment can be obtained by comparing cloud water path and geometric thickness, from which the departure of the cloud's vertical development from adiabatic condensation can be calculated. To perform such an analysis globally requires satellite retrievals of cloud water path, cloud top height, and cloud base height. While the first two are readily available for the same cloud field from the A-Train, the cloud base has historically been mostly unknown.
We have developed an algorithm for CALIOP that provides cloud base heights with an uncertainty estimate for most clouds along the CloudSat/CALIPSO track (Mülmenstädt et al, 2018). The cloud-base data set will yield a global climatology of cloud subadiabaticity, which, together with process-scale cloud field modeling, can be used to constrain GCM-scale entrainment.
Reducing uncertainty on aerosol–cloud interactions and cloud feedbacks
GCMs overestimate precipitation frequency, which — given the energetic constraint on total rainfall — leads them to produce an unrealistically drizzly intensity spectrum. In many models, these problems are particularly severe in warm (liquid cloud) rain processes.
Does too frequent occurrence of warm rain cause models to overestimate the aerosol cloud lifetime effect? Even if the parameterized autoconversion in the model correctly captured the precipitation formation sensitivity to cloud droplet number and thence anthropogenic aerosol, the global estimate provided by a model would still be too high if too many clouds were precipitating globally. I am investigating this hypothesis by modifying the parameterized precipitation in the ECHAM-HAM GCM and by diagnosing warm rain statistics in multi-model and perturbed-physics ensembles.
In Mülmenstädt et al. (2015), we derived a global climatology of warm rain occurrence fraction from CALIPSO–CloudSat satellite observations. This climatology provides an observational constraint on the relative importance of warm and cold rain processes in GCMs.
In some CMIP5 and AeroCom models, we can diagnose the warm-rain fraction and compare it to our satellite climatology. We find that the modeled land–sea contrast is smaller and the warm rain fraction is larger by an order of magnitude in comparison to the satellite data. In this project, I am investigating whether the light precipitation bias is linked to the warm-rain bias.
C&GC Research Group Projects
CAMS/MACC
Copernicus is the European Union's program for the establishment of a European capacity for earth observation. One of the products provided by the Copernicus Atmospheric Monitoring Service (CAMS) is radiative forcing by anthropogenic aerosol. My work within this project is to develop estimates of radiative forcing through aerosol–cloud interactions (RFaci). Learn more about CAMS...
The precursors to CAMS were the ECMWF's Monitoring Atmospheric Composition and Climate (MACC-II and MACC-III) programs, in which I also participated. Learn more about MACC...
Former Projects
For information on projects I've worked on in the past, see my former projects page.