––° ––mph ingest HOLD

Field · warped mesh

Home Education

Academic Path: Present & Future

One discipline.
Many fields.

I aim to develop expterise in computational modeling of spatial-temporal systems through combining computer science, applied mathematics, and geospatial computation with domain-specific knowledge in physical and biological systems. I've designed the path to follow to achieve this goal.

3

Core Master Programs

compute · math · space

6

Physical Domains

what the field is

4

Methodologies

how it is modeled

x, y, t

The object

a field, not a point

The Reasoning

The geospatial piece is not decoration. Most of these systems are not simulations in abstract coordinates.

They are fields evolving through space and time. A storm is a state on a grid. An outbreak is a state on a contact surface. A catchment is a state on a terrain. If the model cannot hold that field — store it, step it, join it to the landscape — it is a picture, not a computation.

The degrees are therefore not a pile of credentials. They are one discipline, built in three layers, then pointed at physical systems and sharpened with methods. The hub is computational modeling of spatial-temporal systems. Everything else is a spoke: what you are modeling, or how.

The map

Hub · one discipline What you model How you model it Currently underway

Educational pathway as a hub and spokes Computational modeling of spatial-temporal systems at the center, built from computing, mathematics, and spatial computation, with physical-domain programs on the left and methodological programs on the right. Hub Computational modeling spatial · temporal · fields Weather risk Illinois Biological physics Johns Hopkins Epidemiology Michigan Hydrology Michigan Tech Meteorology Mississippi State Oceanography Colorado Computing Georgia Tech Mathematics Washington Spatial computation Illinois Remote sensing UConn Statistics Penn State Operations research Columbia AI & machine learning Washington
One hub. Two kinds of spoke: the physical system you are modeling, and the method you use to model it.

A hub of three programs — computing, mathematics, spatial computation — with physical domains on one side and methods on the other. The same structure is written out below.

The hub

Core sequence

One discipline, assembled in order: a machine that can hold a distributed state, the mathematics that can step it, and a spatial representation that keeps the field attached to the world. These three are underway.

In progress

Computing

M.S. Computer Science

A field model is a distributed system. Computing systems is the background that lets a state live across machines, time steps, and failure domains — not just a notebook on a laptop.

View program In progress

Mathematics

M.S. Applied & Computational Mathematics

Discretization, operators, and stability. The difference between a picture of a storm and a model that can take a time step without falling apart.

View program In progress

Spatial computation

M.S. CyberGIS & Geospatial Data Science

Coordinates are not a plotting convenience. GIS at this level is how a field is stored, queried, and joined to the landscape it claims to represent.

View program

What you model

Physical domains

Optional spokes. Each is a physical system whose state is a field. Weather and climate risk sits over the rest: whatever the field is, someone eventually has to price the exposure.

Weather risk

M.S. Weather, Climate Risk & Data Analytics

The domain that sits over the others. Every field eventually becomes an exposure: wind, flood, heat, crop, casualty. Risk is where the model has to price a decision.

View program

Biological physics

Graduate Certificate in Applied Biomedical Engineering

Tissue, transport, and device-scale physics. Same grammar as the atmosphere: a continuum, a mesh, a time step.

View program

Epidemiology

M.S. Computational Epidemiology & Systems Modeling

A population as a field. Space, contact, and time — an outbreak is a distributed system, not a single compartment on a whiteboard.

View program

Hydrology

Graduate Certificate in Water Resources Modeling

Catchments and channels. The hydrograph is a time series of a spatial process, not a curve in isolation.

View program

Meteorology

M.S. Geosciences, Applied Meteorology

The atmosphere as the canonical spatial-temporal field. Observation, analysis, and the 24-hour problem.

View program

Oceanography

Graduate Certificate in Oceanography

A second fluid. Coupled to the atmosphere, slower, and just as spatial — the other half of the air–sea field.

View program

How you model it

Methods

The other kind of spoke. Observation, inference, allocation, and approximation — the operators you apply to a field once you can hold it.

Remote sensing

Graduate Certificate, Remote Sensing & Geospatial Data Analytics

How the field is observed from orbit and aircraft. The ingest before the model — pixels that have to become a state.

View program

Statistics

Graduate Certificate, Applied Statistics

Uncertainty, inference, and whether a signal is real. The 24-hour delta is an inequality; deciding it happened is a statistical object.

View program

Operations research

Graduate Certificate, Operations Research

Allocation under constraint. Once the field is a forecast, someone still has to decide what to pre-position, and where.

View program

AI & machine learning

Graduate Certificate in AI & Machine Learning for Engineering

Function approximation on the field — nowcasting, downscaling, emulation — without pretending the physics went away.

View program

Read from the top

A domain supplies the physics. The object of the work is still the same: a field in space and time. Methods observe it, reduce it, allocate against it, or emulate it. Underneath, three programs make the computation possible.

That is the whole pathway. Not a list of schools. A stack that ends at risk.

From physical domain through methods down to the computational hub What · physical domains weather risk · epidemiology · hydrology · meteorology · oceanography · biomedical The object a field evolving through space and time · x, y, t → state How · methods remote sensing · statistics · operations research · AI / ML Computing Georgia Tech Mathematics Washington Spatial Illinois Hub · computational modeling of spatial-temporal systems

Adjacent

Further interests

Not a second pathway, and not credentials. The other surfaces that keep showing up once a field has to be drawn, indexed, distributed, or proven.

Computer graphics

Rendering · meshing · volumetric

The actual math and systems behind visualizing fields evolving over space and time.

Computational geometry

Algorithms for spatial data

Meshes, predicates, and the geometry a field has to sit on before anyone draws or queries it.

Databases

Query · indexes · storage engines

The design of the databases themselves. Spatiotemporal stores — Uber H3 and the rest.

Distributed systems

How a state lives across machines without pretending the network is reliable.

Formal verification

Proofs about programs, protocols, and discretizations. Whether the system does what the write-up claims.