§ I · What we do, in a sentence

Parcelscope reads thirteen public records against a single Davidson County address and renders what the records say in plain English. The underlying data is federal, state, and local. The reading is ours. Every finding shows its source.

That is the whole product. Everything below is the explanation.

§ II · The spatial join

Every report begins with one operation: locating your parcel on a map and asking each public dataset what it has within the relevant radius.

The radius depends on the dataset. For underground storage tanks and leaking tank cases, we look one mile in every direction — wider than the typical Phase I ESA radius for USTs, which under ASTM E1527 is usually a quarter to a half mile. We chose one mile because, for residential buyers, even a closed UST a mile away is a piece of historical context worth knowing about. For state remediation sites and dry cleaner records, we look half a mile, because contamination from those sources tends to be more localized. For landfills and solid waste transfer stations, we look three miles. For radon, we look at the EPA's county-level zone designation, because the underlying geology operates at that scale.

The radii are not arbitrary. They reflect how each kind of contamination actually travels — through groundwater, through soil, through air, or not at all. We chose them by reading the methodology notes that accompany each public dataset and by following the radii the EPA, ASTM, and peer-reviewed environmental health literature use for the same purposes.

§ III · How we read distance

Distance from a parcel to a hazard is measured in feet, not in zip codes. A landfill 0.4 miles upwind is not the same as a landfill three miles downwind, and the report never rounds the distinction away.

Each finding includes:

The bearing matters. A dry cleaner half a mile northeast of your home, on prevailing southwest winds, is downwind of your home. Solvent vapors and historical groundwater plumes from that cleaner are unlikely to reach you. The same dry cleaner half a mile southwest, on the same prevailing winds, is upwind — its history is part of your air. We do not draw conclusions from this alone, but we surface it because the reader deserves the full geometry.

§ IV · How we read prevailing wind

For Davidson County, prevailing wind comes from BNA — Nashville International Airport — the nearest NOAA weather station with continuous hourly data. We use a ten-year window of hourly observations, from 2014 through 2023, and calculate the directional mode: the compass direction the wind comes from most often.

For Davidson County, that direction is south-southwest, roughly 205 degrees. A meaningful share of hours over a multi-year window have wind coming from somewhere within the southwest cone. Wind from the northeast — the opposite direction — is meaningfully rarer.

Every finding in your report is read against this prevailing direction. We are not modeling specific air dispersion or particle transport. We are noting the geometry. A reader who wants to think about exposure in detail has the bearing and the prevailing direction; the reading is theirs.

A note on what this method does not do: it does not capture micro-climate variation within Davidson County, which is real and which we do not yet model. The Cumberland River corridor, the Highland Rim ridges, and the dense canopy of certain neighborhoods produce local airflow patterns that diverge from the BNA station record. For most addresses, the BNA prevailing direction is a reasonable proxy. For addresses near the river or on the rim, it is an approximation. The report says so where it matters.

§ V · How we deduplicate

Several of the underlying datasets describe the same physical facility. The TDEC underground storage tank database lists active tanks. The EPA's UST Finder includes the same tanks plus historical ones. The TDEC LUST database lists facilities with documented releases. A single gas station may appear in all three under slightly different names — "Forest Hills Shell," "Forest Hills Shell #1234," "FH Shell Station" — and at slightly different addresses, because the agencies geocode independently.

If we reported each appearance as a separate finding, the reader would see one facility three times. We do not.

Our cross-source reconciliation matches facilities by name similarity, address, and coordinates, then collapses them into a single canonical finding with all the regulatory history attached. A facility with one active tank and two historical leaking tank cases shows up once, with the tank counts and case statuses listed clearly. The reader sees the truth — one facility, three regulatory events — instead of three findings to reconcile manually.

This is the most algorithmically interesting part of the engine. It is also the part that took the longest to get right.

§ VI · How we score

Each report includes a score from 0 to 100, with a band: Excellent, Good, Moderate, Elevated, or High Risk. The score weights the thirteen categories by the magnitude and proximity of findings within each.

A few principles that shape the score:

Distance matters more than count. A single active leaking tank case at 0.1 miles weighs more than five closed cases at 0.9 miles. Concentrated proximity is the variable, not raw enumeration.

Closed cases weigh less than active ones. A regulatory case that the state has formally closed has been remediated to state standards. We do not pretend it never existed — the report names it — but we weight it less than an open case. The score reflects the practical present.

Wind direction modifies airborne categories. Findings upwind of the parcel weigh more in air-related categories (PM2.5 from highways, solvent vapors from cleaners) than the same findings downwind. Groundwater categories are not modified by wind.

Some categories are informational, not scored. Tree canopy, aircraft noise contour band, prevailing wind itself — these are surfaced because the reader should know them, but they are not included in the numeric score because the literature does not support a clear scoring relationship. We tell you what we know without inflating the certainty.

The exact weights, multipliers, deduplication thresholds, and proximity decay curves are kept confidential as proprietary methodology. We publish the categories we read, the radii we search, the principles that govern scoring, and the sources we cite. We do not publish the precise calibrations — those are the product of substantial engineering work and represent the core technical asset of Parcelscope.

§ VII · What we cite

Every finding shows its source. Every source shows its retrieval date. If a TDEC LUST case appears in your report, the report names the case identifier, the regulatory status as of the retrieval date, and the agency that maintains the record. If you want to verify a number, you can.

The peer-reviewed studies we lean on for methodology are named in the report itself. The primary one is Krzyzanowski et al., Proximity to Golf Courses and Risk of Parkinson Disease, JAMA Network Open, May 2025 — a case-control study from Barrow Neurological Institute and the Mayo Clinic that found living within one mile of a golf course was associated with a 126% increased risk of Parkinson disease compared to living more than six miles away. The Parcelscope golf-course finding is built on that work, and we say so explicitly. We do not paraphrase the study's conclusions; we link to the study and let the reader read it.

Other peer-reviewed sources inform specific categories: HEI's traffic-related air pollution literature for the major-road proximity score, EPA's radon zone methodology documentation, FAA's DNL contour standards. These are listed in the Sources & Citations page.

We do not invent sources. We do not paraphrase what we have not read carefully. The bibliography is real.

§ VIII · What Parcelscope is not

It is worth being explicit about what we do not do.

Not a Phase I Environmental Site Assessment. Phase I ESAs follow the ASTM E1527-21 standard, require interviews with current and former owners, include site reconnaissance by a licensed environmental professional, and produce a document that satisfies the federal All Appropriate Inquiries standard for CERCLA liability protection. Parcelscope is none of that. Phase I work is appropriate for commercial real estate transactions and certain residential transactions; if you need one, hire a licensed environmental consultant. We will not pretend our $49 report is a substitute.

Not a soil test. We do not sample soil, water, or air at your parcel. The findings come from public records of what other people have observed and reported. If you want laboratory-grade measurements of what is on your specific lot, hire an environmental testing firm.

Not a property condition assessment. We do not evaluate roof age, structural integrity, plumbing, HVAC, or any other condition of the home itself. Hire a home inspector for that work.

Not a substitute for a real estate attorney. Tennessee disclosure law, deed history, easements, and other legal questions about your transaction are properly handled by counsel.

Not predictive. Parcelscope reads what is in the public record. It does not predict future contamination, future regulation, or future health outcomes. The reader should treat the findings as historical and current — a baseline for decisions, not a forecast.

§ IX · A note on the reader

The product is built on an assumption: that the buyer of a Davidson County home is capable of reading thirteen environmental records and making their own decisions about what they mean. We do not soften findings to reduce anxiety. We do not amplify findings to drive urgency. We name what is in the record, in plain English, with the geometry intact.

The reader is the expert on their own life. We are the expert on the records. The handoff between those two roles is the report.

Parcelscope is currently focused on Davidson County, Tennessee — Volume I. The methodology described here is what the engine does as of this writing. We will update this page when the methodology changes.