Archives

Four aerosol classifiers sit within 1.2 points of each other on lab accuracy. On a real background day, their false-alarm rates differ by 3×, and the alarm rule matters more than the model.

A Python reproduction of a 1999 cooking model, and the choices hidden inside its doneness score.

After my AMT paper brought Mean Teacher to mass spectra, I rebuilt the classifier with a control that shares everything except the consistency loss. Across three test folds, that loss added nothing measurable to overall macro-F1.

The same small neural network looked useless on one storm season and skillful on two held-out winters. What changed was the test, not the network.

CDC WONDER's privacy protections systematically hide data from the populations we most want to study—small, rural, and minority communities. This article quantifies the bias and offers practical strategies for researchers.

Temporal Fourier projection extracts wave wavelengths from noisy spatio-temporal data. In these synthetic tests, it gives finer wavelength resolution than a full-plane 2D FFT.

Tracing the evolution of scientific inquiry from its philosophical roots to today's complex institutional landscape.

Explore how to characterize wave-like data using mathematical tools and visualization techniques, including the Hilbert transform, to extract key properties such as angular frequency and wavelength.

Exploring the invisible yet influential world of gravity waves in Earth's atmosphere, their mathematical underpinnings, and observational insights.

Explore the history and impact of the Howard University Program in Atmospheric Sciences (HUPAS), a pioneering initiative that transformed diversity in atmospheric sciences and paved the way for underrepresented students.

A deep dive into calculating momentum transfer and pressure tensors in DSMC simulations, with applications to atmospheric wave analysis

Exploring the connection between molecular pressure tensors and Reynolds stress tensors, and how they reveal the fundamental principles of momentum transport in fluids.

A quick and accessible exploration of Reynolds stress tensors in spherical coordinates, explaining their derivation and similarity to Cartesian coordinates for scientifically literate readers.