Alaskan Data Visualization

Alaska operates on logistics, weather, and wildlife. In this environment, accurate data isn't a luxury—it's infrastructure. My approach to data visualization is rooted in a fundamental belief: Public data should be aggressively public, inherently accessible, and terrifyingly fast.

The Problem with Government APIs

We fund massive scientific operations (NOAA, ADF&G, USGS) that collect incredible, life-saving datasets. However, the interfaces provided to the public are often buried in 1990s-era FTP servers, complex XML schemas, or unresponsive web portals that fail miserably on a 3G mobile connection in Prince William Sound. The data exists, but the *accessibility* is broken.

The Intermediary Layer

My role as a systems designer is to build the "middleware" between raw government science and the Alaskan end-user. The architecture is straightforward:

• Ingestion: Cloudflare Workers routinely ping raw data endpoints (NDBC Buoys, Fish Weirs, Aurora Forecasts).
• Sanitization: The raw text/XML is parsed, anomalies are stripped, and it is converted into clean, minified JSON.
• Caching: The clean payload is stored globally on the Edge Network.
• Delivery: Frontend Progressive Web Apps (PWAs) request the clean Edge cache, resulting in blazing-fast loads even on satellite internet.

Practical Execution

This philosophy drives the primary tools I build:

• Alaska Fish Count App: Scraping fragmented daily fish weir counts across the state, unifying the schema, and presenting it in high-contrast charts for sport and commercial fishermen.
• Alaska Aurora App: Translating the NOAA Space Weather Prediction Center's complex Kp-index and hemispheric power models into a simple, geolocated "Go / No-Go" alert system for photographers.

Design Rules for Extreme Use

When a mariner is checking a buoy reading in 15-foot seas, they do not care about a spinning UI animation. High-level data viz here requires strict rules: Dark mode default (to preserve night vision on the bridge), high-contrast typography, zero layout shift (CLS), and functional offline fallback states via Service Workers.