How we measure this data

Why we measure over developed land, not over the whole area

This is the decision that most affects the numbers, so it comes first.

An area's boundary contains water, marsh, parkland, dunes and farmland as well as houses. If you calculate "what percentage of this place is in a flood zone" over the whole boundary, you are largely measuring how much undeveloped marsh it happens to contain — which tells a homeowner nothing about their own street.

So every areal figure here uses developed land inside the Census place boundary as its denominator — the land people actually live on, identified from the National Land Cover Database. Flood-zone coverage, sewer district coverage, and the sampling points for elevation and water table all follow this rule. Where a figure is measured at a single point rather than over an area, that point is the centroid of the developed land, not the geographic center of the boundary.

Water table depth and elevation

Depth to groundwater is one of the most useful numbers on this site for a homeowner, because it drives basement seepage, sump-pump duty cycles and how well a cesspool disperses.

For each area we take 15 sample points, chosen deterministically — seeded so the same area always yields the same points, and so any run can be reproduced — from within the developed land inside its Census place boundary. Elevation at each point comes from lidar; depth to water comes from the relevant regional groundwater survey. The published figure is the average across those points.

What it does not tell you: this is an average across the developed part of a area, not a reading at your address. Groundwater varies street to street and moves seasonally — it is generally shallowest in late winter and spring. Treat it as the local baseline, not a measurement of your lot.

FEMA flood zones

Two separate figures come from the FEMA National Flood Hazard Layer, and the distinction between them matters.

The zone we publish is the flood zone found on developed land, not one that touches only marsh or open water inside the boundary. That zone can still be a thin strip: a VE zone along the shore may cover well under 1% of an area's developed land. So the zone is never shown without the share below it, and where that share is under 1% the page says how small it is.

The percentage is the share of developed land inside a Special Flood Hazard Area, using the developed-land denominator described above.

What it does not tell you: FEMA zones describe surface flooding. They say nothing about groundwater seeping into a basement, which can happen well outside any mapped zone and is not covered by a standard NFIP policy. Zones are also periodically remapped — check the current FEMA Map Service Center for your own address before making a decision.

Sewer coverage, and cesspool reliance

Sewer coverage is the share of an area's developed land that falls inside a sewer district that is active. Districts that are proposed, under design or retired do not count — a district that does not yet exist does not take anyone's wastewater.

Where a district layer shows no coverage, we confirm the result independently against the EPA NPDES register of permitted wastewater treatment works. An area recorded as unsewered because no permitted treatment works of any size lies within its boundary is a positive finding, not an absence of data.

Sewer status decides which services we show. Where an area is sewered we do not offer cesspool service, and where it is unsewered we do not offer municipal sewage-backup service, because those are the wrong trades for the property.

What it does not tell you: coverage is areal, not per-property. In a partially sewered area, whether your house is connected depends on your street, and the only authority on that is the district itself.

How streets are counted

The street lists and the "look up your street" box count Suffolk County's address points: one primary point per address, placed by the county. Place markers are left out. A street appears in a list only when it has at least five addresses, because below that the count is noise; the lookup box includes streets with three or more so that short streets can still be found.

• Flood zone. An address counts when its point falls inside a FEMA special flood hazard area, the same map the flood-zone figure comes from. That says the address is mapped inside the zone. It does not say the building floods, and an address outside the zone is not safe from water.
• Sewer main. An address counts when an active county sewer main runs within 45 m (about 150 ft) of its point. That means a main runs past it, not that the house is connected. Village-run systems are missing from the county's layer, so a street that is not listed is never described as unsewered. Proposed mains are the county's plans on file, not a schedule.
• Groundwater. Depth to groundwater is read at each address point from the USGS map of Long Island's water table, drawn from wells measured in April and May 2013. A street is listed by the median depth under its addresses. It is one spring's map, not a live reading, and the differences between nearby streets mostly follow the ground rising and falling.

What it does not tell you: whether water reaches a particular house. A street count is a starting point for a question, not an answer about your property.

How the address check works

The address check reads the same county address points as the street lookup and adds what the public maps show at each one. Every figure is read at the address point, which the county places on the building for most addresses.

• Flood zone. Inside, close to (within 30 m of the edge) or not mapped inside a FEMA special flood hazard area. Mapped, not flooded.
• Ground height. USGS 3DEP elevation, in cells about 10 m across, in feet above sea level. It describes the ground around the house, not the floor. Checked against the USGS point service at 20 random addresses: the median difference was 0.3 ft and the largest 3.6 ft, on steep ground.
• Groundwater. The same USGS 2013 map as the street lookup, read at the address. "Within reach of a basement floor" means mapped at 14 ft or less, and of a cesspool ring 16 ft or less.
• Storm surge and evacuation zones. Both are the county's layers. The surge zones come from the National Hurricane Center's SLOSH model and an Army Corps inundation layer and are worst-case planning maps, not forecasts. The evacuation zones are a different map: some addresses are in an evacuation zone and in no surge zone, so the two are never treated as one finding.
• Sewer. A main is stated only where the county's active mains run within 45 m, as on the street lookup. An address with no mapped main is never described as unsewered. A sewer district is the county's boundary, not a connection.
• House type, year built and living area. From the New York State assessment-roll tax parcel points for 2025, matched to an address by house number, street and town. House type is on most parcels. Year built and living area are filled in for only some towns, so a missing value means the roll does not record it, not that the house is new or small.
• Comparisons. The share of addresses in the same area, and in the county, that have lower ground or deeper groundwater than this one. An area with fewer than 100 addresses with a reading is compared with the county only. A comparison never decides whether a line appears.

What it does not tell you: whether water has reached or will reach a particular house. It is a quick screen, not an elevation certificate or a flood determination. The street and house number you pick stay in your browser. A copied link to a report carries them after the #; the page takes them out of the address bar before anything else on it runs, so they are not sent to us or to the site's analytics.

Soil drainage

Drainage class comes from the USDA SSURGO soil survey, taken as the most common class across a 5-point sample within the area, and reported as one of five standard classes.

What it does not tell you: SSURGO maps native soil. It does not know about the backfill around your foundation, which is frequently different and is what actually governs how fast water leaves the side of a house. Where the survey returns no class for an area we publish none rather than guessing one.

Tree canopy

Tree canopy is the share of an area’s land under tree cover in the NLCD Tree Canopy Cover product for 2025, measured over the same boundary as our other areal figures with open water excluded from the denominator. Where an area has no boundary of its own we publish no figure rather than borrowing a neighbor’s.

We use it for one thing: storm power outages. Trees are the leading cause of storm outages in the Northeast — research across the region finds vegetation implicated in the large majority of storm trouble spots, and in up to 90% of outages during extreme weather. A sump pump fails when its power fails, so how much vegetation stands over the local distribution network bears directly on how long a basement goes unprotected.

What it does not tell you: it is not a prediction that trees will damage your house. We looked for a published source connecting canopy cover to storm damage to structures and did not find one — the available evidence points the other way, with tree species and condition mattering far more than how much canopy a town has, and one recent wind-engineering study finding trees reduce wind loading on low-rise walls depending on approach angle. So we do not make that claim anywhere. The figure is also canopy across the whole area, not canopy over houses specifically: woodland and preserve count toward it. And it is a 2025 measurement of a modeled annual product — NLCD derives each year from the Landsat record by harmonic prediction rather than surveying it afresh, so treat it as a good estimate of that year rather than an observation of it.

Housing age and seasonal occupancy

On building codes: New York adopted the model codes carrying today’s wind-load provisions effective 2003, and the statewide Uniform Code before that dates to 1984. We report housing age because it is measured; we do not convert it into a storm-damage score. The share of an area’s homes built before 1970 — the cutpoint our data uses — corresponds to no New York code event, so treating it as a wind-vulnerability measure would attach a real mechanism to an arbitrary date. See the storm damage page for what we do say instead.

Average house age, the share of housing built before 1970, and the seasonal share come from the US Census American Community Survey 5-year estimates, joined on Census place GEOID. Seasonal occupancy is table B25004 item 006 — vacant, for seasonal, recreational or occasional use — over B25001, total housing units. Population is from the Decennial Census.

What it does not tell you: ACS figures are survey estimates with margins of error, and they are averages over a whole place. A 1958 median tells you what the local building stock is like; it does not tell you when your house was built or what its plumbing has been through since.

Flood insurance claims

Claim counts and total paid come from OpenFEMA's redacted NFIP claims dataset, filtered to the county and aggregated by reported ZIP code, against a frozen snapshot so figures are reproducible.

These figures are ZIP-scoped and we never attribute them to a single area. That is a rule, not a caveat. Many areas share a ZIP code with one or more neighbors, so a ZIP total covers several areas at once. Where a page shows a claims figure it names the ZIP, because saying "X has had N flood claims" when the number actually covers four neighboring places would be false.

What it does not tell you: NFIP claims only exist where someone held a federal flood policy and filed. Uninsured damage does not appear, and neither does groundwater seepage or a failed sump pump, which NFIP does not cover.

Hurricane evacuation zones and shelters

Both come from the county emergency-management authority's published map services, and both are assigned over the developed part of each area rather than its geographic center — a boundary's centroid can easily sit in marsh, in a different zone from where the houses are.

Suffolk County's surge zones are lettered A to E by storm-surge risk. Zone A has the highest surge risk and is evacuated first; Zone E has the lowest and is evacuated last. An area outside every zone reads "Not mapped".

"Not mapped" is published as exactly that, and never as "no risk." Shelters frequently serve several neighboring areas, so the same shelter appears on more than one page; that is the real answer, not a duplication artifact.

What it does not tell you: evacuation orders are issued by the county for specific zones and specific storms, and shelter openings vary by event. Before a named storm, confirm with your county's emergency-management office directly. Nothing on this site is an emergency instruction.

Named water features

Bays, rivers, creeks and ponds named on our pages are cross-checked against the USGS Geographic Names Information System, matched on features whose coordinates fall inside the Census place boundary.

Where GNIS lists no named feature inside a boundary, that is recorded as a sourced finding, not missing data, and the page simply does not mention a water body. We do not reach for a feature in a neighboring area to fill the gap.

The 1–5 service scores, and what they are not

Internally each area carries a 1–5 score for five services, and a page for that service is built where the score reaches 3. The scores are not measurements, and we never publish them as a risk level. Whether a problem can exist in a place is decided by a fixed test on a measured fact, never by how the place ranks against the rest of the county: a ranking says "worse than the county median", which is not the same as "this happens here".

• Cesspool overflow — built wherever an area is not mostly on public sewer, because that is where cesspools are; the score only orders those areas.
• Sewage backup — considered only where an area is mostly on public sewer, because elsewhere the same failure is a cesspool.
• Basement flooding — the strongest of three routes: groundwater depth against fixed cutpoints (10, 20, 40 and 80 ft); older housing and soil drainage; and storm surge with the share of developed land in a flood zone. Barrier-beach places built on pilings, with no below-grade level, score 1. Elevation is left out because it correlates ~0.94 with water table depth and would count the same signal twice.
• Sump pump — the same groundwater cutpoints, or the sewage ejector pump under a basement bathroom, where an area is not mostly sewered and its housing is older.
• Frozen pipe — a within-county ranking of seasonal-home share and average house age, equally weighted, because the classic burst is an empty house with the heat off.

The quantities behind the scores are measured and we publish those — the water table depth, the housing age, the seasonal share. The score itself stays internal, because a score dressed up as a measurement is misleading.

Comparisons against neighboring areas

A single measurement says how deep the water table is somewhere. It does not say whether that is unusual for that area — and no area's own data can answer it. So each area is compared against the areas nearest it, and the one or two measurements where it genuinely differs are reported.

"Genuinely differs" is set from the measured spread of those differences across the county rather than picked: the bar for each metric is the 75th percentile of the observed area-versus-neighbors gap. Below that bar we say the area is in line with its neighbors, which is a true and useful answer rather than a failure to find something to say. The number of neighbors is always stated so you can weigh it. A neighbor here is a nearby area, not always one that shares a border: every area within a mile counts, nearest first, up to six, and a place with none that close, such as a Fire Island village, is compared with its nearest two.

No comparison is published where an area has fewer than two neighbors in our data, or where its own measurements are inherited from a boundary it shares with a neighbor — comparing those would compare a number with itself.

What a figure means for a home, and where that comes from

Beside some figures an area page says in one sentence why the figure matters to someone with water in the house. Those sentences change with the figure, and each rests on a public source, not on our own experience of a place:

• Flood zone and who pays. Most homeowner’s policies do not cover flood, and a home in a mapped flood zone with a federally backed mortgage must carry flood insurance (FEMA, National Flood Insurance Program). A sudden, accidental burst pipe is usually covered; a slow leak from poor upkeep usually is not (National Association of Insurance Commissioners). Your own policy decides, not this page.
• House age. Homes 30 years old were three times as likely to have a plumbing supply or drainage problem, in a study of water damage claims by the Insurance Institute for Business & Home Safety, published by the Connecticut Insurance Department.
• Heating fuel. A fuel oil tank can float or tip over in a flood, and the oil should be removed before the water is pumped out, never pumped into the yard (New York State Department of Environmental Conservation). That oil is delivered and gas is piped is how the two fuels arrive; we publish no rate of homes losing heat.
• Cesspools and flooded ground. A system that drains into the soil cannot drain while that soil is saturated, and pumping it then is a temporary fix at best (US EPA, septic systems after a flood). EPA writes about septic systems; a cesspool drains into the ground the same way.

What we will not do

• We do not fill blanks. Where a source has no value for an area, the page says nothing rather than carrying an estimate. A blank is a finding.
• We do not publish rankings as measurements. See above.
• We do not reword pages to make them look different from each other. Where two pages differ, it is because what they say about the place differs.
• We do not claim a presence we do not have. This is a connection service with no premises in the areas it covers, the footer of every page says the specialists are independent, and the structured data says we have no premises.

Corrections

If a figure is wrong about your area we want to know, particularly if you have a source. Contact details are on your region's page, linked above. Corrections are made against the source data and republished, not patched on a single page.