Community characteristics

Three layers count buildings, housing units, and people on a 30 m grid. They come from the USDA Forest Service's Wildfire Risk to Communities program (Populated Areas publication, RDS-2020-0060-2), derived from building footprints overlaid with U.S. Census records. Buildings are counted from footprints above a 40 m² (430 ft²) minimum on habitable land — no water or permanent ice/snow. AOI sums are the meaningful number; a single 30 m pixel value is an estimate, not a head-count.

How WyldFire processes every layer

Each WRC raster is clipped to the USGS Wyoming state boundary plus a 24-km buffer for edge tolerance on AOIs that straddle the state line, scaled to compact integer storage using a documented scaleFactor where appropriate, and re-encoded as a Cloud Optimized GeoTIFF on S3, addressable via /vsicurl/ HTTP range reads. The viewer multiplies by scaleFactor when rendering and reporting; direct COG readers must apply it themselves.

Layers at a glance

Layer What it measures Native res scaleFactor Years
bc Building count 30 m 1.0 2020
hu Housing unit count 30 m 0.1 2020
pc Population count 30 m 0.01 2020

Building count bc

What it is

"The approximate number of buildings within a 30-meter pixel. No spatial smoothing is applied to this layer."

Whole-number counts. Only buildings with footprints larger than the WRC minimum of 40 m² (≈430 ft²) are included, and only those falling on habitable ground (open water and permanent ice/snow excluded).

How upstream computes it

WRC starts from a CONUS building-footprint dataset and assigns each footprint centroid to its containing 30 m pixel. The pixel value is a count of the number of qualifying footprint centroids that fall inside it.

How WyldFire processed it

Clipped to Wyoming + 24-km buffer. No scaling (scaleFactor = 1.0) because buildings are genuinely countable objects — fractional values would have no physical meaning. Re-encoded as a COG on S3.

Storage vs display

Native integer counts. Viewer renders with a class-breaks color ramp (the bins are a viewer convention for legibility, not part of the raster).

30 m semantics

A pixel of 3 means three qualifying footprints had their centroid in that 30 m × 30 m square (≈0.22 acres). Aggregate across an AOI for the headline number; individual pixels are useful for finding clusters but should not be read as building inventories.

Caveats

"The datasets presented here are the product of modeling, and as such carry an inherent degree of error and uncertainty. These datasets are intended to provide nationally-consistent information for the purpose of comparing relative wildfire risk among communities nationally or within a state or county. Data included here are not intended to replace locally-calibrated state, regional, or local risk assessments where they exist."

Outbuildings under 40 m² are excluded. Footprint completeness varies by region. Not for parcel-level analysis.

Year availability

2020.

Provider & citation

USDA FS WRC, RDS-2020-0060-2. Jaffe, Melissa R.; Scott, Joe H.; et al. Wildfire Risk to Communities: Spatial datasets of wildfire risk for populated areas in the United States. Fort Collins, CO: Forest Service Research Data Archive. doi.org/10.2737/RDS-2020-0060-2

Housing unit count hu

What it is

"Housing is defined as buildings that are likely to have occupants, according to building footprints and 2021 U.S. Census Bureau datasets. … The approximate number of housing units in each 30-meter pixel."

A subset of bc restricted to the buildings most likely to contain occupants, then Census housing-unit counts are allocated across those qualifying footprints.

How upstream computes it

Census-block housing-unit counts (2021 American Community Survey vintage) are distributed onto the WRC building footprints judged habitable, then aggregated to 30 m pixels. The allocation step is why per-pixel values are not whole numbers — Census counts are block-level, and dividing them across multiple footprints in the same block produces fractional pixel values.

How WyldFire processed it

Clipped to Wyoming + 24-km buffer. Scaled by scaleFactor = 0.1 so the Census-allocation values (typically 0 to ~10 per pixel, often fractional) compress to small integers in storage. Re-encoded as a COG on S3.

Storage vs display

Multiply raw integer pixel values by 0.1 to recover housing-unit counts. Viewer renders with a class-breaks color ramp.

30 m semantics

A pixel of 0.7 hu does not mean "0.7 households live here." It reflects the Census allocation step — a fractional share of a block's housing-unit total has been attributed to this pixel. Sum across the AOI for the headline number; never read a single pixel as a household count.

Caveats

Subject to the same WRC use-limitations quoted in the bc section. Allocation uncertainty is highest where blocks are large relative to the building footprint density. Census vintage is 2021; anything built more recently is missing.

Year availability

2020.

Provider & citation

USDA FS WRC, RDS-2020-0060-2. doi.org/10.2737/RDS-2020-0060-2

Population count pc

What it is

"The approximate number of people living in each 30-meter pixel. No spatial smoothing is applied to this layer."

How upstream computes it

Census-block population counts are allocated onto the same habitable-footprint set used for hu, then aggregated to the 30 m grid. Same general allocation approach as hu, with the population numerator substituted for the housing-unit numerator.

How WyldFire processed it

Clipped to Wyoming + 24-km buffer; scaled by scaleFactor = 0.01 (population values per 30 m pixel run higher than housing-unit values, so a finer scale is used); re-encoded as a COG on S3.

Storage vs display

Multiply raw integer pixel values by 0.01 to recover person counts. Viewer renders with a class-breaks color ramp.

30 m semantics

A pixel of 2.4 pc means an estimated 2.4 people are attributed to this 30 m square through the Census-allocation step. Aggregate across the AOI for population sums; per-pixel values are estimates, not headcounts.

Caveats

Same WRC use-limitations quoted in the bc section. Allocation uncertainty applies; transient population (workplaces, tourism, seasonal residences) is not separated from residential population.

Year availability

2020.

Provider & citation

USDA FS WRC, RDS-2020-0060-2. doi.org/10.2737/RDS-2020-0060-2

Reference XMLs: D:\wkRepos\WyldFire\importantFiles\layerSourceMetadata\