American Toxics

This is an interactive map of Superfund contaminated toxic waste sites across the U.S.

americantoxics.com or read the background

Development

bun i
bun dev

Supabase database setup

1. Install the Supabase CLI (if you haven’t already):

   bun i -g supabase

2. Initialize Supabase in this project:

   supabase init

3. Configure your database credentials in .env:

   NEXT_PUBLIC_SUPABASE_URL=<your-supabase-url>
   NEXT_PUBLIC_SUPABASE_ANON_KEY=<your-supabase-anon-key>
   SUPABASE_SERVICE_ROLE_KEY=<your-service-role-key>  # for server-side operations

4. Run the migration to create the sites table:

   supabase db push
   # or, manually:
   # psql $SUPABASE_DB_URL -f supabase/migrations/0001_create_sites_table.sql

5. Import the site data JSON into your database:

   bun lib/tasks/import-sites.ts

6. Run the migration to create the scores table:

   supabase db push
   # or, manually:
   # psql $SUPABASE_DB_URL -f supabase/migrations/0002_create_scores_table.sql

Now, import and use the client from lib/supabaseClient.ts:

import { supabase } from "@/lib/supabaseClient";

// Fetch all sites
const { data: sites, error } = await supabase.from("sites").select("*");

Updating sites list

• Download core CSV from EPA to lib/data/sites-raw.csv

  curl -L "https://hub.arcgis.com/api/v3/datasets/c2b7cdff579c41bbba4898400aa38815_0/downloads/data?format=csv&spatialRefId=3857&where=1%3D1" \
    -o lib/data/sites-raw.csv

• Download GeoJSON from EPA to lib/data/sites.geojson (gitignored because huge)
• bun run lib/tasks/csv-to-json.ts to convert to JSON
• bun x @mrodrig/json-2-csv-cli lib/data/sites.json -o lib/data/sites-processed.csv if you want to convert back to CSV

Environment variables

Variable	Description
OpenAI for chat completions	OPENAI_API_KEY
Supabase database: key	NEXT_PUBLIC_SUPABASE_ANON_KEY
Supabase database: URL	NEXT_PUBLIC_SUPABASE_URL
Supabase: server-side operations	SUPABASE_SERVICE_ROLE_KEY
Mapbox for maps	NEXT_PUBLIC_MAPBOX_ACCESS_TOKEN

MIT License