Projects: School

Public

Thesis

This was a study that examined the effect of the Appalachian Mountains on the precipitation distribution of tropical cyclones that impinge upon them. With the help of my advisor, Anantha Aiyyer, I began by classifying the 28 storms that passed through our study area between 1979 and 2006 into four general track types. I analyzed the track composites resulting, examined individual case studies, and ran numerical simulations with topography (control) and without (experimental) to isolate the effects of the Appalachians.

Climate Modeling

Using the R programming language with a gridded soil moisture dataset spanning the years 1950-2004, I attempted to correlate historical periods of flooding against a field-tested empirical orthogonal function (EOF: a statistical decomposition method relating the spatial distribution of a variable with its time series) showing zonal (east-west) hurricane track shifts through time as determined by Xie et al. (2005). The goal was to find how much of this flooding was due to hurricane activity.

Numerical Model

In a class taught by Dr. Matthew Parker, I was required to code a numerical model and later use it for a mesoscale (ie: greater in extent than storm-scale) scientific experiment. I first hardwired flux form advection, leapfrog time differencing, and lateral sponge boundaries into the code. I later added Asselin filtering, artificial diffusion, Rayleigh dampening, and Kessler microphysics.

Boundary Conditions

Unless your model domain is of global scale, you must choose some artificial boundary where your model-grid ends. Boundaries are areas of discontinuity that can negatively affect numerical solutions. Fortunately, there are methods available that serve to limit the detrimental effects these discontinuities can have on model results.

Here is a brief discussion of some of the boundary conditions commonly employed in atmospheric models. These are periodic, rigid, dampened, and sponge boundaries.

Finite Differencing

Finite differencing is an explicit numerical means of solving spatial and temporal derivatives. I’ll show you the math that allows us to predict the future!

Come with me on a fantastic journey through time and space!

Radar Design

A team project where money was no object, our task was to design a custom Doppler radar system for the greater Boston area. We determined the bounds of our system by studying the dominant weather patterns of the region and a selection of severe weather case studies. For both the extreme events, and the more typical conditions, the challenge was to strike a balance between output sounding resolution and update speed, between creating research-grade products and maintaining everyday functionality.

Charlotte Storm Water

Since its inception, stream restoration has been guided by the Rosgen stream classification system. Typical parameters include cross-sectional dimension, channel and valley shape, grade, sinuosity, stream features, channel material distribution, and geographic setting. While integral, this system is limited to (quasi)static streams. With the great majority of our work at Habitat Assessment & Restoration Professionals was done in urban(izing) settings, whose watersheds contain large areas of impervious cover (and are thus highly susceptible to flash flooding and frequent channel modification), we find this system to be severely lacking in its present state. With this study, Dr. Forsythe seeks to improve upon this (otherwise beneficial) status quo.

Green Roof

The Biology Department turned to Dr. Forsythe for help in determining how best to study the local environmental effects of green roofs. I employed the use of Erdas Imagine on aerial photography to come up with an unsupervised land-use classification map that delineated a number of different land-surface types. This would allow them to choose an appropriate local diversity of locations across the city. We also helped set up their control study, measuring in situ surface properties such as temperature, thermal inertia, and humidity.

Charlotte Mecklenburg Utilities ArcGIS

Given a list of addresses where grease blockages had been found and cleared from sewers over the four-year time-span, I created a color-coded map of grease spills for CMU using ArcGIS.

CH2MHill

CMU brought CH2MHill in to investigate the cause of persistent complaints of odor logged at certain locations in Charlotte. They proposed that the odor was due to sewage that was not moving quickly enough through the system. This was allowing bacteria to build up at these points of stagnation. I was contracted through the UNC-Charlotte Environmental Assistance Office to collect construction as-built plans of nearby sewer lines. I was responsible for mapping the grade of these individual pipes and flagging areas of shallow slope.