A statewide weed dashboard that lives in one file

The problem

The inventory data already existed and was already authoritative, published as ArcGIS REST services. What was missing was a way to look at it that did not require a licensed dashboard product, a per-seat viewer, or a copy of the data that would start drifting the day it was made.

What we built

One HTML file. It queries the published services directly at load, holds the records in the browser, and renders a clustered map alongside charts that all read from the same filtered set. There is no server component, no build step, no database, and nothing to license. It runs from any host that can serve a static file.

How it works

Live query, no copy

Records come from the state's own services on every load, paged to handle up to several thousand features. What the dashboard shows is what the program's authoritative layer says right now.

Clustered map

Up to 8,000 points cluster at zoomed-out scales and separate as you move in, so a statewide view stays readable and a county view stays precise.

Seven filters, chained

Species, location, and status filters narrow together, and every chart and the map redraw against the same filtered set. There is no state where the chart and the map disagree.

CSV export of what you see

Whatever the filters currently show exports to CSV. That single feature is what turns the dashboard into the first step of the annual report instead of a thing to look at and then go rebuild in a spreadsheet.

What it changed

The program got the views it wanted, owns the code outright, pays nothing per seat, and can hand the file to anyone. It also got a template: the same approach scales up to a self-hosted build on PostGIS with MapLibre GL when a program outgrows the single-file version.