I am a fourth-year PhD student in the physics department at Penn State. I obtained my B.S. in Physics at Yale.
At Yale I wrote my thesis with George Fleming on the representation theory of SO(3). Before that I worked for the National Oceanic and Atmospheric Administration (NOAA) Earth System Research Laboratory (ESRL) in Boulder, Colorado where I studied cloud and aerosol processes.
I am originally from Boulder and return as often as I can to get out into the mountains.
My Research
Analysis of albedo versus cloud fraction relationships in liquid water clouds using heuristic models and large eddy simulation
A paper written in collaboration with
Graham Feingold,
Franziska Glassmeier,
Takanobu Yamaguchi,
Jan Kazil, and
Allison McComiskey,
published in the
Journal of Geophysical Research: Atmospheres,
introducing a simple analytic model of cumulus to represent the albedo – cloud-fraction relationship. Albedo and cloud fraction are key macroscale properties in climate studies. While it is true that albedo increases monotonically with cloud-fraction the form of the relationship can vary from linear to super-linear. We tie a super-linear response to cloud fields in which clouds widen as they deepen, as in cumulus.
Stratocumulus cloud top radiative cooling and cloud base updraft speeds
A poster presented at the December 2017 AGU meeting in New Orleans.
(Fall 2017) AGU Boundary Layer Clouds and Climate Change[PDF]
Atmospheric Modeling
Our goal at ESRL is to gain a comprehensive understanding of the Earth system. This means that we study processes as small as solar photons driving stratospheric chemistry and as large the global circulation of the jet streams. Clouds, which cover nearly two-thirds of the world's surface at any given time, are an essential part of this system.
Clouds may affect our lives on very short timescales, but their climatic patterns affect our lives much more permanently. In 1974 Twomey showed how clouds couple the climate to pollution. The mechanism depends on microphysical interactions between water vapor and pollution particles called aerosols. When water vapor is lifted in the atmosphere to a temperature where it is cool enough to reach saturation, clouds usually cannot form because the tiny droplets of water break apart too easily. It is only through the action of aerosols acting as cloud droplet nuclei that cloud droplets grow large enough to sustain a cloud. So a small amount of aerosol in the atmosphere is necessary. But with more aerosol, more cloud droplets would form from the same amount of condensed water leading to clouds with different optical properties. Quantifying these changes requires a sophisticated theory of atmospheric chemistry and microphysical processes. In addition, to have any hope of applying our theory we have to locate and quantify precisely the sources of aersol, both natural and anthropogenic. How sensitive are clouds to perturbations in aerosol forcing? How sensitive is the climate to these perturbations? By extension NOAA is interested in concrete practical questions like, How would rain patterns change? And how would we respond? This problem bridges the gap between science and policy-making as we seek to determine in what ways our behavior alters global atmospheric dynamics.
Although there is no substitue for measurements on real clouds, the open atmosphere is difficult to measure and it is impossible to perform replicable experiments. In response to nature's unyielding resolve to never repeat herself, we have created numerical models which simulate natures of our own. With these models we can run large-scale, replicable experiments on realistic clouds which we may then analyze at our leisure and in explicit detail. The influence of numerical models on atmospheric science has been profound.
Processes relevent to atmospheric science encompass a staggering range of scales. Aerosol activation and cloud droplets occupy nano and micro scales while the global circulation occupies thousands of kilometers. To fully resolve all of these processes in a single model will be beyond computational ability for the foreseeable future. But even if we could resolve all of these scales at the same time, we may not want to or even need to. Depending on the application it may be unimportant to track the location of every droplet in a cloud. Thus there exists a hierarchy of models to cater to the diverse needs of atmospheric research. The model my group uses is of intermediate variety, resolving domains from a few kilometers wide to perhaps 100km with a resolution ranging from 10m to 100m. Any process occuring in a gridbox smaller than 10-100m is not resolved explicitly by the model. The model my group uses is called the System for Atmospheric Modeling (SAM).
Stratocumulus
The following is a simulation of stratocumulus based on a case measured by the GEWEX Cloud System Study (GCSS). The domain is 38.4 km on both sides and 1600m high. The video displays a full 24 hour cycle with the background color giving an indication to the time of day. The simulation begins at sundown and goes through a 10 hour night before sunrise.
During the night the cloud layer is thick. During the day shortwave heating from solar radiation causes the cloud layer to thin. This video illustrates a decoupling between the upper, cooling layer from the lower heating layer.
Cumulus
This is a closeup of a small corner of a standard simulation using SAM. The case simulated here is BOMEX.
