White Mountains historical atlas
White Mountains
Through Time
Watch federal forestland assemble, forests return, working railroads recede, roads reshape valleys, and mountain communities grow—without pretending every year has equal evidence.
Research workbench
Compare the evidence, not just the years
Opening historical release…
Selected evidence
Forest return · 1913 to 2025
Nnorth-up locked
Preparing the historical workbench
The map will open on a north-up comparison of the first dated acquisition evidence and the current endpoint.
Interactive map unavailable
This browser could not start WebGL. The timeline, story cards and source record remain usable below.
Open the source chronology
Evidence gap
No mapped sample is available at one of these dates.
Earlier1913
Later2025
earlier evidence
later evidence
Draft reconstruction
coverage edge
confirmed absence
no evidence
Source-aware timeline
Snap to real evidence dates
Gaps are honest: an empty tick means the source is known, but this prototype has no mapped sample there yet.
renderable verified / authoritative
cataloged source; mapping pending
Draft reconstruction / illustrative fixture
explicit no verified local coverage
annual evidence era begins in 1985
Loading the release manifest and sample evidence…
Coverage will appear when a date is selected.
Curated entrances
Three ways into a changing landscape
The workbench opens with a question and a place, not a wall of layer switches.
Loading the infrastructure story…
Loading the community story…
Map of data-source time
The timeline changes character as evidence improves
Dates describe the evidence itself—map edition, image exposure, transaction or annual composite—not a fictional uniform historical layer.
Mapped context
State maps, post-road maps, railroad surveys, county maps, Hurd’s atlas and Sanborn town-center sheets.
- Generalized outside detailed insets
- Edition date is not always observation date
- Absence on a map is not proof of absence
Federal forest assembles
Parcel-by-parcel White Mountain National Forest acquisitions connect ownership policy to visible forest recovery.
Map convention: current recorded parcel geometry colored by documented action date. It is not proof of the polygon’s exact 1913 boundary.
- Transaction date and former owner
- Recorded acreage
- Continuous chronology, irregular geography
Photographic evidence
Patchy archival aerial frames begin early; systematic statewide orthophotos become the detailed comparison backbone after 1992/98.
- Coverage footprints must always be visible
- Exposure date stays separate from scan date
- UNH/SPNHF terms require item-level review
Repeatable measurement
Landsat begins the region-wide satellite record; annual canopy and land-cover products begin in 1985.
- Early MSS evidence is coarse
- Annual does not mean error-free
- Product version remains part of provenance
Early maps · county atlases · Sanborn · USGS editions
WMNF parcel acquisitions
Archival aerials · statewide orthophotos
Landsat · annual canopy and land cover
Editorial and data method
Every shape must carry its evidence
A visually exact line can still be historically uncertain. The production release keeps source, date, coverage and interpretation separate so the interface can show that uncertainty honestly.
Observed is not published
Image exposure, map edition, acquisition transaction and dataset release dates are stored independently.
Coverage is a first-class layer
A Sanborn sheet, aerial frame or quadrangle never silently implies statewide coverage.
Fuzzy stays visible
Draft reconstructions retain a visible “Draft reconstruction” badge and use hatching or dashed boundaries; verified-only is an optional filter, not the default.
Immutable releases
The viewer opens a versioned manifest. Evidence artifacts are content-addressed so a published comparison remains reproducible.