KinNoKi Labs

GIS case study · 2013–2026

From field maps to
mapping products.

More than a decade turning field observations, forestry records, imagery, public data, and rural infrastructure questions into maps people can actually use.

The work is not making a pretty map. It is turning imperfect real-world information into a useful, understandable decision tool.
Generalized map reconstruction with invented land shapes coloured yellow, blue, pink, green, and grey for different field and paperwork states.
Public-safe reconstruction Invented geometry; no basemap, names, areas, or client location.

ArcMap 10.1 · Forestry operations

Field status, made legible

Problem

A forestry job held areas in several paperwork states. The map needed to distinguish confirmed, new, and unresolved information without pretending uncertain records were settled.

Inputs

Working GIS layers, property-reference information, previous mail-outs, and field notes that still needed checking.

Method

Revised colour categories in ArcMap and carried uncertainty into the legend instead of smoothing it away.

Output

A field and office reference map with a deliberately visible “needs review” state.

It supported: reconciling the map and paperwork before follow-up.

It did not establish: survey boundaries, title, Crown status, or present-day conditions.

Cropped winter aerial mosaic showing a winding stream through snow-covered forest, with visible seams between the creator's source photographs.
Creator-captured imagery Tight crop from the original mosaic; structures and location context removed.

Field capture · Raster stitching

A first aerial mosaic

Problem

See a whole flown area rather than a folder of disconnected frames.

Inputs

Camera images captured by Dan during a flight over the work area.

Method

Stitched overlapping frames into one continuous, zoomable PDF while keeping the imperfect first-pass seams visible.

Output

A continuous winter mosaic that made terrain and image coverage easier to review.

It supported: reviewing coverage, stitch quality, and broad terrain relationships.

It did not establish: survey-grade position, orthophoto accuracy, or current conditions.

Generalized dark regional shape containing a regular point grid classified green, yellow, orange, and red for relative modelled visibility.
Public-safe reconstruction Invented geography and classes; utilities, addresses, basemap, and exact results removed.

QGIS 3.30 · Scripted spatial analysis

A regional question, screened spatially

Problem

Compare where a hypothetical 200-metre turbine might be visible from fewer or more buildings across a large candidate area.

Inputs

A regular candidate-point grid, land-screening geometry, and modelled building-visibility counts.

Method

QGIS plus an adjustable scripted analysis developed iteratively from the plain-language question.

Output

A classified regional screening map that made relative patterns visible.

It supported: comparing parameters and focusing further research.

It did not establish: wind resource, siting approval, grid access, land rights, environmental suitability, or project viability.

Generalized planting review reconstruction with invented green treatment blocks, white field access line, and green or red observation points.
Public-safe reconstruction Invented treatment blocks and sample points; landowner, counts, outcome, imagery, and location removed.

QGIS 3.32 · Field verification

Field observations, returned as a decision map

Problem

Turn a planting inspection and its tally into a clear account of what was observed and where follow-up might be needed.

Inputs

Field plot observations, tree-condition counts, photographs, and a treatment boundary.

Method

Reviewed how the tally should be interpreted, connected observations back to the map, and kept the field evidence alongside the summary.

Output

A plot map and companion tally for review and follow-up.

It supported: reviewing a historical, sample-based field outcome and locating areas to revisit.

It did not establish: a whole-site inventory, a current condition, or a result independent of the agreed tally method.

NS Marks The Spot browser interface showing a public OpenStreetMap background, a locally imported USGS GeoPDF listed in the map layers panel, and its extent outlined over the map.
Product acceptance capture Local import UI over OpenStreetMap; imported GeoPDF content is not displayed.

Native + browser mapping · Product engineering

The map became a product

Problem

Historical scans, public records, and field position arrive in different formats — often with weak connectivity, privacy constraints, and source-specific licence terms.

Inputs

Official and public sources, user-owned images and GeoPDFs, and the local context a researcher brings to the map.

Method

Offline-first native tiles; browser georeferencing with embedded GeoPDF frames or manual control points; local-file handling, source receipts, and licence gates.

Output

NS Marks The Spot as an iPhone companion and an open browser map.

It supports: source-linked research, field orientation, and better questions about historical geography.

It does not establish: survey position, title, access, present road or trail condition, tax-sale status, or legal advice.

Capture from the exact public NS Marks The Spot source. The visible background retains OpenStreetMap contributor attribution; the listed USGS GeoPDF is represented only by metadata and an extent outline.

The practice behind the pictures

A map earns trust at its edges.

01

Start with the decision

Choose data, scale, and output from what someone needs to understand or do next.

02

Carry uncertainty

Unknown, unresolved, and source-limited information stays visible instead of being polished into certainty.

03

Protect the source

Private artifacts stay private; public layers keep their attribution and licence conditions.

04

State the boundary

Every useful map also says what it cannot prove and which official source must be checked.

Have a map-shaped problem?

Bring the messy source material.

Start with the field notes, old PDF, public dataset, or repeated question. The first job is to work out what the evidence can honestly support.