Woodland caribou · Ontario · 2015–2025

Ontario Suitable Caribou Habitat Analysis

Where woodland caribou core habitat survives across seven Ontario ranges, mapped every year at 30 m, and how much of each range harvest, fire and infrastructure have disturbed.

Ranges: 7Berens to Kesagami
Area analysed: 20.8 Mha
Pixel: 30 mLandsat scale
Annual maps: 112015 to 2025

Forest-dwelling caribou, Parc national des Grands-Jardins, Quebec · Photo: Alain Caron, Wikimedia Commons, CC BY-SA 4.0 (cropped)

Map of the seven Ontario caribou ranges, shaded by core habitat in 2025. Large dark patches dominate Kesagami and Pagwachuan in the east; Berens and Sydney in the west are lighter and more fragmented. — Dominant core patch size <100 ha 100–500 500–1,000 1,000–10,000 >10,000 ha Colour shows the patch-size class holding the most core habitat in each 750 m cell; strength shows how much of the cell is core. The small box marks the 27 km close-up used to explain the method.

Data sources

Dataset	Provider	Year	Resolution	Used for
Caribou range boundaries	Ontario MNR	2011	vector	Analysis unit
Boreal caribou ranges	ECCC, Recovery Strategy	2012	vector	Federal comparison
Forest land cover (VLCE2)	NRCan · Canadian Forest Service	2019	30 m	Potential habitat
Analysis Ready Inventory	Ontario	2025 release	stand polygons	Harvest, fire, blowdown
Forest harvest 1985–2022	NRCan · Canadian Forest Service	2022	30 m	Harvest gap-fill
Anthropogenic disturbance footprint	ECCC	2020	1:50,000	Infrastructure and 40-year fire

Core habitat, 20253.06 Mha−4.3% since 2015

Share of range disturbed43.6%up from 41.4% in 2015

Ranges above ECCC's 35% threshold6 of 7in 2025, on Ontario range boundaries

Range by range

What a range is, and what goes into each one

Ranges are Ontario's planning units for boreal caribou (MNRF 2014). Each is drawn to contain one local population from aerial surveys, telemetry, historical occurrence and calving and wintering areas, with edges set along rivers, lake chains, highways, forest management unit lines, the Far North boundary and provincial borders. They are not herd boundaries: ranges include unoccupied habitat, and Sydney, Berens and Kesagami share their populations with Manitoba or Quebec. ECCC expects at least 65% of each range to stay undisturbed, so this page uses 35% disturbed as the threshold.

ECCC ON3 · continuous range

Churchill

Forest management units in the range

Method

From land cover to core habitat

Five steps turn national land cover into core habitat. Steps 3 to 5 run once for each year: harvest accumulates from 1976, and fire counts for 40 years as in ECCC's method. The approach follows Mackey et al. (2024), extended from one snapshot to an annual series. Every step is in the processing notebook on GitHub.

Processing chain each input feeds the step above it

STEP 1AlignOne 30 m grid in EPSG:3978 for every layer

Caribou range boundariesOntario MNR

STEP 2Potential habitatConiferous forest + treed wetland inside the range

Forest land cover (VLCE2)NRCan · 2019 · 30 m

STEP 3DisturbanceHuman footprint buffered 500 m, plus fire from the last 40 years

Disturbance footprintECCC · 2020
Analysis Ready InventoryOntario · harvest, fire, blowdown
Forest harvest 1985–2022NRCan · 30 m

STEP 4Undisturbed habitatStep 2 minus step 3

STEP 5Core habitatMSPA: strip a 30 m edge, join patches, size them

Five patch-size classes100 · 500 · 1,000 · 10,000 ha

77 annual rasters · 7 ranges × 11 years Range statistics · area, % disturbed, risk class Map-ready layers · cloud-optimised GeoTIFFs

Close-up

Land cover, 2019

Follow one 27 × 27 km window in the Churchill range, inside Trout Lake Forest. Every 30 m pixel carries a class from NRCan's national forest land-cover map.

Coniferous 46.2% · Mixedwood 35.0% · Water 8.9%

Potential habitat

Keep the classes caribou use

Only coniferous forest and treed wetland stay in. Everything else is set aside before any disturbance is considered.

Potential habitat: 49.1% of the window

Disturbance, 2025

Union every disturbance layer

A pixel counts as disturbed if any source touches it: ECCC's 2020 footprint of roads, mines and settlements (buffered 500 m, merged with 40 years of fire); harvest since 1976 from Ontario's inventory or the national Landsat record, buffered 500 m; and inventory fire, blowdown and drought. Overlaps count once.

Disturbed: 45.2% of the window

Undisturbed habitat

Subtract disturbance from habitat

What remains is potential habitat that no disturbance reaches. In this window, just under a third of the potential habitat is lost.

Undisturbed habitat: 33.7% of the window

Core habitat

Strip the edges, size the patches

Morphological Spatial Pattern Analysis removes a one-pixel edge, joins touching pixels into patches and measures each one. Patches under 100 ha are treated as sub-functional; the rest are grouped at 500, 1,000 and 10,000 ha.

Core habitat: 21.0% of the window · 49.5% of it in patches over 1,000 ha

Land cover in the close-up window: mostly coniferous and mixedwood forest with lakes.

Potential habitat: coniferous forest and treed wetland in green.

Results

What changed, 2015–2025

Across the seven ranges, core habitat fell 4.3%, from 3.20 to 3.06 million ha, while the disturbed share rose from 41.4% to 43.6%. The rise comes in two steps: 2021, when the burned area counted in the western ranges jumps, and 2023, when it jumps in the central and eastern ones.

Share of each range disturbed, 2015–2025 Ontario range boundaries · hover for a year

Show the numbers

Berens range

Fire reshapes a range in a decade

Burned area counted in Berens nearly tripled between 2015 and 2025. Disturbance rose six points and core habitat fell 15%. Fire also works the other way: old burns leave the 40-year window, which is why Brightsand is the one range where disturbance fell.

Berens range in 2015, shaded by share of each cell disturbed. — Drag to compare
Undisturbed Share of cell disturbed, 0 → 100% Water

Berens range in 2025, with larger burned areas shaded plum. — Drag to compare
Undisturbed Share of cell disturbed, 0 → 100% Water

Fragmentation

Total core area hides how broken up it is

Sydney holds no core patch over 1,000 ha, and Berens none over 10,000 ha. 71% of Nipigon's core sits in patches under 100 ha. Two-thirds of Kesagami's is in patches over 10,000 ha.

Core habitat by patch size, 2025 % of each range's core area

<100 ha 100–500 500–1,000 1,000–10,000 >10,000 ha

Boundaries and forests

Boundaries, federal figures and forests

Three things shape how these numbers can be used: the range boundary they sit on, how they compare with ECCC's own disturbance figures, and the forest management units each range touches.

Range boundaries

Ontario's ranges drive the analysis; ECCC's federal ranges sit alongside.

For the seven shared ranges the two boundaries nearly coincide: 97–99% overlap, with areas within 1%. Ontario's boundaries are the ones its forest management plans work with. ECCC defines its 35% threshold on its own boundaries, so the federal comparison below uses those.

Map of northern Ontario showing Ontario's 15 caribou ranges and ECCC's 9 federal ranges, which follow almost the same lines for the seven analysed ranges. — Ontario range, analysed Ontario range, not analysed ECCC federal range (2012)

ECCC comparison

ECCC's own layers put the seven ranges at 29–55% disturbed; finer harvest and fire data add 2.3 to 12.7 points.

The ECCC value here is computed from ECCC's published map layers: its 2020 footprint (human disturbance buffered 500 m, merged with 40 years of fire) divided by its federal range area. It has not yet been matched against the percentages printed in ECCC's reports. Adding Ontario's stand-level harvest, the national Landsat harvest record and inventory fire moves Berens and Brightsand from Moderate to High, and Nipigon from Low to Moderate.

✓ DoneCompute % disturbed from ECCC's footprint and federal range layers.
✓ DonePut the same footprint on our 30 m grid and Ontario boundaries: it lands within 0.7 points, so grid and boundary handling add little error.
○ Not yetRebuild the footprint from ECCC's mapped roads, cutblocks and other features, and match it to the percentages ECCC publishes.
◐ PartlyAdd finer layers one at a time. Harvest and fire are in; an Ontario road layer and recent clearcuts mapped from satellite imagery are not yet.

ECCC layers and this analysis, 2020 % of range disturbed

ECCC layers (footprint ÷ federal range) This analysis (adds inventory and Landsat harvest, inventory fire)

Show the numbers

Dryden and Thunder Bay

Forests near Dryden overlap four analysed ranges; forests near Thunder Bay also reach an area not yet analysed.

The map shows forest management units within 140 km of each of the two forest-industry towns. Near Dryden, eight forests overlap Churchill, Brightsand, Sydney and Berens; Dryden Forest and Boundary Waters touch no range.

Near Thunder Bay, Black Spruce, Lake Nipigon and Kenogami overlap Nipigon, Brightsand and Pagwachuan, but also Ontario's Discontinuous Distribution area, up to 41% of a forest. That area is outside this analysis and ECCC does not map it as a range.

Report per range, not per forest. % disturbed and risk class describe a whole range; show each forest's share of the range instead.

Map of northwestern Ontario showing forest management unit outlines, caribou ranges and the towns of Dryden and Thunder Bay. — Range analysed Range not analysed Forest within 140 km Town

Caveats

Before quoting a number

Keep in mind

Percentages use Ontario's range boundaries unless marked ECCC.
Roads come only from ECCC's 1:50,000 mapping; no Ontario road network layer is loaded.
Harvest counts from 1976 with no recovery; fire drops out after 40 years, which is why some older disturbance disappears between 2015 and 2025.
Inventory depletions of unknown cause are treated like fire and also recover after 40 years.
Churchill, Kesagami and Pagwachuan show identical 2024 and 2025 values because their inputs end in 2023.
Blowdown and drought count as disturbance here; ECCC counts only fire.

Next steps

Rerun the ECCC comparison on federal boundaries.
Add an Ontario road layer and measure what it changes.
Analyse the Discontinuous Distribution area around Lake Nipigon and Thunder Bay.