How Jackson County Uses GIS When Nobody Is Watching
Most people think geographic information systems are just maps on screens. They're not. In Jackson County, Indiana, GIS is how we decide where to put a water main extension without triggering a three-month engineering review, or how we catch a zoning variance that should never have been approved. The tool itself doesn't matter as much as the data layering strategy behind it. I spent two years setting up our parcel-level floodplain overlay here. The county sits in a transition zone between the Ohio River basin and the northern glaciated hills, which means standard FEMA flood maps are wrong in about 18 percent of the mapped area. You can see it immediately if you compare the published FIRM sheets against our LiDAR-derived terrain model. The real work starts when you realize the FIRM boundary misses entire sub-divisions in the southern part of the county near Brownstown. The setup itself is straightforward if you already have an ArcGIS or QGIS environment running. We pulled parcel shapefiles from the county auditor, merged them with township road centerlines from INDOT, and then clipped everything to the 100-year floodplain layer from FEMA's Map Service Center. The trick most people miss is the coordinate transformation. Jackson County uses NAD 83 (2011) State Plane West Indiana FIPS 1602, but a lot of the state-level datasets default to NAD 83 (G1762). If you don't reproject before you clip, your parcel edges shift by roughly 1.2 meters. That matters when you're determining whether a septic tank sits inside or outside a riparian buffer zone.
We run a simple model builder script that flags any parcel with a 50-foot buffer overlap onto a designated high-water stream corridor. It used to take our planning staff about 40 hours per quarter to do this manually. Now the script runs overnight and spits out a flagged parcel list by morning. The catch is that the script fails when parcel boundaries haven't been updated since the 1990s. We found three subdivisions where the digitized lines didn't match the recorded plat. The workaround was pulling the original survey plats from the recorder's office and re-digitizing the boundary in a separate editing geodatabase, then publishing a corrected feature class back to the main GIS server. Another thing nobody warns you about: tax parcel ownership data is not the same as legal ownership. Our GIS ties a parcel to the assessor's record, which lags by 6 to 18 months after a sale. During that gap, the mapped owner on our system is wrong. We solve this by running a monthly merge against the auditor's recorded deed dataset. The mismatch rate is around 4 percent every cycle. You either accept that or build an automated matching job, which takes extra development time and still produces false positives when two buyers share a last name. For infrastructure planning, we use a network analyst approach rather than the old distance-decision method. Instead of drawing a circle around a fire station and saying anything inside gets served, we run travel-time polygons based on the actual road hierarchy. A house 0.4 miles from the station on an unimproved gravel road might take longer to reach than one 1.1 miles away on County Road 600 East. This changed our emergency response zone boundaries in the rural northwest township. The polygon showed a coverage gap that the Euclidean buffer completely missed.
There are real limitations to what GIS can do for Jackson County planning. The data quality is only as good as the last surveyor who updated the parcel. We've got no utility records baked into our system. If someone wants to map stormwater culvert capacity, they're either entering data by hand or using a field tablet with a ruggedized GPS that costs more than our entire annual software license. And the county's current enterprise license covers about 12 named users. Once you go past that, the per-seat cost makes it impractical for every planner to have their own GIS workstation. We use a web-based viewer for the field staff instead, which strips out most of the analysis capabilities but is enough for basic parcel lookups and site plan reviews. If you're trying to replicate this in another rural Indiana county, start with your floodplain data and verify it against LiDAR. Don't trust the FEMA map layer without ground truthing at least a few sample points. Then fix your parcel ownership sync before you build anything on top of it. Everything else is just layers on top of those two foundations.
Get the Full Details
