← All work · Self-initiated demonstration on public data
Case study · Property & flood risk
Jacksonville Property & Flood Risk
Takes messy county property records, matches each home to its flood zone and sales history, and flags anything that doesn't add up, so you can trust the numbers before you buy, insure or lend.
The problem
Answering simple questions about a property (what is it worth, is it in a flood zone, what sold nearby) means combining four public sources: county parcel maps, the property appraiser's tax roll, FEMA flood maps and state sales records. They come in different formats, with broken map shapes, duplicate addresses and inconsistent ZIP codes, and it's easy to lose records along the way without noticing.
What it does
- Loads the county, FEMA and state records and fixes broken property shapes without ever editing the originals.
- Removes duplicate addresses and tidies up ZIP codes, keeping a record of every change.
- Matches each property to the FEMA flood zones it touches, including whether it's in a high-risk area and how much of the lot is covered.
- Matches state sales records to properties and runs automatic checks every time the data is rebuilt, so nothing goes missing quietly.
What you get
- One clean, checked property database, with one row per parcel.
- An interactive map you can share.
- A count of what was loaded, repaired and removed, so you can see the data is complete.
- No owner names published: the public map shows only address and values.
By the numbers
Built with: PostgreSQL/PostGIS, Python, GeoPandas, dbt, Leaflet (EPSG:2236)
Contact
Tell me about your data.
Send a short note about your area, your data and what you need. I reply by email, usually within one business day, or you can book a call directly.