Methods & data
A useful record explains where it came from, how it was reviewed, and what it cannot tell us.
From an observation to a finding
Real field records would remain separate from corrected results, reviewed interpretations, and public releases. A changed method would create a new version, preserving the original.
Different states mean different things
Reviewed
A stated review is complete. Review does not eliminate uncertainty.
Provisional
A record exists, but review is pending. Interpretation may change.
Stale
The last observation is older than the expected update window.
Offline
A known connection outage prevents a current display; history keeps its date.
Missing
No usable observation exists for the interval. It is not zero.
Invalid
A record failed a quality check and is withheld from the reported value.
Try all six states in the synthetic station →
Wildlife observations need effort
Camera events would be reported alongside valid camera-days, placement, season, and review status. Multiple images can depict the same animal. A few cameras on one property are not independent landscapes.
To evaluate a future change, an appropriate study would need observations before and after, suitable reference sites, sufficient duration, and an analysis of alternative explanations. A short feasibility pilot does not establish a migration route or causal harm.
Privacy and scientific independence
Precise sensitive wildlife locations and raw imagery of people would stay restricted. Public field media would undergo human review and metadata removal. Partners would agree credit and data terms; donors would not control findings. Null results and material corrections belong in the record.
The 3D model
The model on our homepage covers a square 30 km (about 19 miles) on a side, centered on the Bryce Amphitheater, in NAD83 / UTM zone 12N. Elevation comes from the U.S. Geological Survey’s 3D Elevation Program lidar. The surface image is USDA National Agriculture Imagery Program (NAIP) photography acquired in September 2021 at 0.6 m. Both are public domain. We used the copies already prepared for the Hoodoo app.
Detail is concentrated where you look first. The central 4,096 m square, from Bryce Point to past Sunrise Point, uses the 1 m lidar on a 4 m mesh (8 m on phones) and imagery at 1 m. Beyond it, the mesh spacing and image resolution grow steadily with distance, from about 4 m at the edge of that square to about 200 m of mesh spacing at the far edge; that outer land comes from a 3 m lidar product and imagery averaged to 4 m. The far terrain fades into haze, and the outer rim of the model always does, so no edge shows.
We added light from a fixed sun, with soft shadows and ambient occlusion computed from the lidar, and adjusted color so the exposed rock looks closer to how it appears on the ground. The Elevation view is tinted by height from about 1,650 to 2,850 m with 20 m contour lines. The vertical scale is true, not exaggerated. The opening view was chosen by Chase Kaczmarek: from above the plateau, looking east-northeast across the rim into the amphitheater.
Viewpoint markers use OpenStreetMap positions (© OpenStreetMap contributors). The plateau and Bryce Creek markers were placed on the lidar surface. Elevations in the tour are read from the 1 m lidar at those points; the value under your cursor comes from the model’s mesh and can differ by a meter or two. Both can differ from published viewpoint elevations. Bare-earth lidar leaves out trees, and resampling softens the thinnest hoodoos, so the model is a faithful picture of the ground’s shape at this scale, not a replica of every spire. It is not a live view.
Thor’s Hammer
The Thor’s Hammer model is built from the laser returns that a U.S. Geological Survey lidar flight recorded on the pillar in 2018, about 3,500 of them, read from the public copy of that point cloud. We rebuilt the pillar as a stack of horizontal outlines every 20 cm. An aircraft looking down records the tops of ledges well and almost nothing under an overhang, so the cap, the knob and the shoulders come from the lidar, while the neck under the cap and the waists below it follow the outline in a public-domain Park Service photograph taken from the west, in proportion to the cap. The ground around it is the same flight’s bare-earth returns; trees and smaller mounds are left out. Light comes from a fixed afternoon sun. Colors and the fine surface texture, within about 10 cm of the rebuilt surface, are drawn to look like the rock; they are not measured.
The beds in the cut-away view and the cross-section are read from the shape: where the outline juts out we show a harder bed, where it pinches in a softer one, following the Park Service’s explanation that layers with more calcium carbonate cement resist rain and stand out. No rock was sampled, and beyond the pillar the beds are assumed to lie flat. Heights are lidar elevations (NAVD88): the top of the cap is at 2,401 m, about 28 m above the slope on its uphill side and about 42 m above it on the downhill side. The section runs from the Sunset Point viewpoint through the cap, 97 m apart; the long profile uses the 1 m USGS elevation model.
Natural Bridge
The Natural Bridge model starts from the same 2018 U.S. Geological Survey lidar flight, read from the public copy of the point cloud: about 1.5 million laser returns in a 220 m square around the arch, some 14,000 of them on the span. Trees are left out by their echoes (a pulse through a pine crown splits into several dim returns; bare limestone returns one bright echo), and the highest remaining return every 25 cm gives the top of the rock. An aircraft looking down sees the top of the span but never its underside, so the opening needs other evidence. We use two kinds. First, every laser return marks the straight path back to the aircraft as open air; the scanner looked up to 17 degrees to either side, so some paths slipped under the span from both ends and outline the lower part of the opening. Second, a public-domain USGS photograph taken at the overlook in December 2014 shows the valley through the opening: we solved where that camera stood and how it was pointed by matching the photograph to the lidar, then treated every pixel that shows the valley as a sight line through open air. The two agree where they overlap.
The opening is drawn as one outline carried straight through the rock along the line of sight from the overlook, bounded below by the ground the laser found under the span. That is a simplification: the real opening may flare or narrow inside, and the photograph shows nothing below the snowy lip in front, so the lowest part comes from the lidar alone. Heights are lidar elevations (NAVD88): the top of the span is at 2,632 m; the opening is about 33 m tall from the ground beneath to the underside of the span and about 19 m wide halfway up, with about 16 m of rock above it. The Park Service gives the span as 85 feet (26 m). Colours, the thin layers and the shallow grooves are drawn to look like the rock, within about 15 cm of the rebuilt surface; they are not measured, and the harder and softer beds they suggest were not sampled.
The before-and-after view follows the Park Service’s account of how windows form and fall. The fin is the rock with the opening filled in; the window is today’s outline shrunk toward its middle; the fallen span is today’s outline swept upward until the rock above it is gone. These are drawings made from today’s model, not reconstructions of a dated past or a forecast of the arch’s future.
Queen’s Garden
The Queen’s Garden model covers a block about 430 m across, from the forested plateau at Sunrise Point down through the garden, using the same 2018 U.S. Geological Survey lidar flight: about 6.5 million laser returns, some 34 per square metre. Trees are left out by their echoes, as for Natural Bridge, and the highest remaining return every 40 cm is the top of the rock or ground. What is left of a tree after its crown is dropped, a narrow spike with crown echoes around it, is removed too; a few narrow rock spires near trees may have gone with them. The green marks show where the laser found tree crowns, except on standing rock: the same echo test also fires along the edges of hoodoos and fins, so it is not used there. Because the model is a height map seen from above, rock walls are drawn straight down and nothing under an overhang is shown.
The trail is the Queen’s Garden Trail as mapped by OpenStreetMap volunteers, from the Sunrise Point trailhead to the viewing spot at the end of the spur to Queen Victoria: 1.39 km each way. Its heights are the lidar ground under the mapped line, every metre. Mapped positions can be a few metres off, so where the line meets rock overhead the model shows the path passing beneath it; we do not claim a tunnel there. From the trailhead at 2,439 m the path drops to 2,326 m at Queen Victoria, 113 m lower, with a low point of 2,317 m on the way; adding every rise on the return gives about 140 m of climbing. The Park Service lists the trail as 1.8 miles (2.9 km) round trip with 450 feet (137 m) of elevation gain; its trailhead page gives 320 feet (98 m) of elevation change. The position of Queen Victoria is OpenStreetMap’s; we do not measure the hoodoo itself. Colours and fine texture are drawn, not measured.
How we prepare the field notes
The geology guide uses original explanations supported by NPS and USGS sources. Each article links its references and separates the source-check date from any scientific review. The first four notes were checked September 30, 2026; no independent geologist review is recorded.
The hoodoo process and Claron and Grand Staircase diagrams are original BCC teaching illustrations. They simplify geological ideas and are not measured maps, timelines or predictions. The observation prompts are editorial exercises, not scientific monitoring protocols. Hoodoo feature descriptions reflect its development edition, not a public release.
Sources & credits
3D model: elevation from the USGS 3D Elevation Program lidar (1 m in the center, 3 m beyond); imagery from the USDA NAIP 2021 acquisition; viewpoint positions © OpenStreetMap contributors. Tour notes cite their National Park Service pages individually.
Thor’s Hammer: lidar returns from the USGS 3D Elevation Program’s 2018 southern Utah collection (public copy on AWS, public domain) and the 1 m USGS elevation model; the neck traced from the NPS photograph credited below; positions of the pillar and Sunset Point © OpenStreetMap contributors.
Natural Bridge: lidar returns from the same 2018 USGS collection (public domain); the opening from the USGS photograph Bryce Canyon Natural Bridge, December 17, 2014, USGS Office of Communications (public domain), used as evidence and not reproduced on this site; positions of the arch and the overlook © OpenStreetMap contributors. Stage captions follow the NPS Natural Bridge wayside and Arches pages.
Queen’s Garden: lidar returns from the same 2018 USGS collection (public domain); the trail and the positions of Sunrise Point and Queen Victoria © OpenStreetMap contributors; trail figures from the NPS Queen’s Garden Trail and trailhead pages.
Hoodoo screens: captures of the app’s development build in the iOS Simulator, September 2026. They show USDA NAIP imagery, USGS elevation and OpenStreetMap trail data.
The landscape photographs depict Bryce Canyon National Park. They are archival photographs, not live views, acquisition targets, or land held by BCC. Each source record below identifies the image as public domain.
- Early Riser at the Amphitheater — Brian B. Roanhorse / National Park Service, October 30, 2015. Used in the scenic gallery.
- A tree grows at Sunrise Point — Heather Williamson, NPS Gallery, September 17, 2022. Used on the geology hub, in the geology-and-care article and in the scenic gallery.
- Thor’s Hammer at sunrise — NPS Gallery. Capture date has not been verified. Used in the scenic gallery, and its outline supplies the neck of the Thor’s Hammer model.
The hoodoos-and-swoosh logo is original artwork selected for BCC, created with an image-generation tool. The deer drawing is an original illustration, not a camera detection. No agency or partner marks are used, and photo credit does not imply endorsement.
Typography: Gelasio and Inter, locally served under the SIL Open Font License. Gelasio license · Inter license.
Scientific references
The example dataset is hand-authored and labeled synthetic at the record level. Its channel-mean transformation and quality screen are demonstrations, not an implementation of EPA’s correction or regulatory quality assurance.