Five demographic futures. One country's roads, homes and water systems. This is what population growth and decline actually do to the places Canadians live — told through the census subdivisions where it happens.
The VISION model places every one of Canada's 4,928 tracked communities on a path to 2050. Some paths lead sharply up. Others lead down. Almost none of them are flat.
Under —, Canada's population rises from — in 2025 to — by 2050 — a change of —. But a single national number hides almost everything that matters about where that happens.
Every shade of green on the map is a community projected to grow past 2025 levels; every shade of red is one projected to shrink. Under this scenario, of all the population growth in Canada lands in — alone.
Nationally, — of communities are growing, — are shrinking, and the rest sit roughly flat.
Take a real example rather than an average.
A mid-size city adding tens of thousands of residents means new schools, new water connections, and roads built for a town of one size now carrying a city of another.
Not every community's story is growth. This one's population is moving the other way.
A shrinking community still has to maintain every road and pipe it built for a larger population — the fixed costs don't shrink with the people. Chapter 6 comes back to exactly this problem.
This is where scenario matters most. Try the picker at the top of the page. Switching from a low-growth to a high-growth future doesn't just scale the numbers up — it changes which communities are growing and which are shrinking.
Population 65 and over changes by — to 2050 under this scenario, against — for the population overall. Age structure shifts even in places where the total heads count is barely moving — a fact that matters for the kind of housing and services a community will need, not just how much of it.
Every resident the population model adds needs somewhere to live. The housing model translates the population story into units, by type, by place.
Housing unit demand nationally rises — between 2025 and 2050 under —. Every one of those units has to be built, serviced, and connected to water and roads somewhere specific.
Housing demand grows fastest in — (— by 2050), and slowest in — (—). The same regional pattern chapter 1 showed for people repeats almost exactly for homes — because it's the same people needing them.
It isn't only how many homes get built — it's what kind.
Nationally, the share of housing demand met by high-rise units moves from — in 2025 to — by 2050 — density is rising fastest exactly where land is scarcest.
Population change doesn't stay a demographic fact for long. Within a few pages of the model it becomes a specific number of specific homes, of a specific type, in a specific place — which is exactly the resolution a planner needs and a national average can't give.
The transport model rebuilds Canada's 2021 commuting patterns — the actual, measured flow of people from home to work — and re-fits them to each scenario's population and housing geography.
That's the change in daily one-way commuting trips nationally under —, a — shift from 2025. It is built from the real 2021 census commuting matrix, not assumed.
These lines are Canada's busiest commuting corridors. The single busiest, —, alone carries — trips a day. Corridors like it cluster inside a handful of metro regions — commuting in Canada is fundamentally a metro-region phenomenon, not a coast-to-coast one.
The share of trips that cross a municipal boundary moves from — in 2025 to — by 2050 under this scenario — travel is becoming — as growth spreads across metro regions rather than staying inside a single city's limits.
This view switches to road capacity: the share of each community's road network expected to be at or beyond its practical limit at peak hours by 2050.
Growth doesn't add traffic uniformly — it adds it to the corridors and communities already closest to their limits. Chapter 6 shows what happens where that overlaps with a road network already in poor physical condition.
Every community needs treated water for as many people as it plans to house. Not every community has told Statistics Canada how much treatment capacity it actually has — and a few that have are misleading if taken at face value.
Grey communities on the map either didn't file a usable capacity return, or reported a number too small to be their real supply — almost always because they buy treated water wholesale from a larger neighbour rather than running their own plant. Comparing their tiny reported number against full demand would invent a shortfall that doesn't exist, so this dashboard doesn't. That leaves — communities where the comparison is real.
Among those, — are projected to need more treatment capacity than they report having by 2050 under — — up from — today. Together they are home to roughly — people.
Scale varies enormously between constrained communities.
That gap is — of projected 2050 demand.
Adding up every constrained, comparable community, closing the 2050 gap under this scenario means roughly — of additional treatment capacity, serving the people who live in those communities today.
Underneath population, housing, transport and water sits a physical climate that is itself shifting. This chapter uses the same CanDCS-M6 climate projections that feed the water stress model in Chapter 4 — shown here on their own terms.
Every population scenario pairs with its own climate pathway, following the same CMIP6 SSP-RCP convention climate science uses. Under —'s pathway, Canada's population-weighted average temperature rises from — (the 2021 climate normal) to — in the 2071–2100 period — about — of warming. Every scenario shows warming; they differ only in how much.
The hottest day of the year gets hotter too — from — to —. Heat extremes are what strain water systems and infrastructure hardest, more than a shift in the yearly average implies on its own.
Annual precipitation moves from — to — — flat to slightly higher, nationally. The longest dry spell of the year barely moves in the national average, but that average hides real regional differences: the Prairies and the Territories already run the longest dry stretches in the country, wet or dry season aside.
More heat and evaporative demand, without a matching rise in reliable water availability, is exactly the signal the water stress index folds into the demand-versus-capacity picture in Chapter 4. Climate change here doesn't invent a new problem — it sharpens the ones already visible in growing, water-constrained communities.
Population growth, housing demand, congestion, water constraints and aging infrastructure don't always coincide. When they do, the effects compound rather than add.
Every community on this map is scored from 0 to 5 on how many growth-side pressures coincide: rapid population growth, rapid housing growth, a water constraint, poor road condition, and road congestion. The brightest communities are carrying several at once.
Under —, — communities — home to roughly — people — face rapid population growth, rapid housing growth, and an infrastructure problem, all at the same time.
A shrinking population still owns every kilometre of road it built at its peak. As fewer residents share that fixed cost, the burden per person rises — regardless of whether the community ever intends to spend the money.
Most communities carry no compounded pressure at all. The story isn't that every place is at risk — it's that risk is sharply concentrated, and concentrated in different places depending on which future arrives.
Five scenarios, one dataset, the same conclusion each time: infrastructure planning is a geography problem before it is a budget problem.
Growth and decline are both real, often in neighbouring communities, and where each happens depends heavily on which of the five futures actually unfolds.
Demographic and infrastructure change don't move independently. The communities under the most pressure are usually facing several kinds of pressure at once, not one.
Capacity questions are local, not national. A country-wide number on water, roads or housing hides more than it reveals; the answer is different in every community and every scenario.
Nothing here is a new calculation. Every figure is read from the same pipeline behind the full analytical dashboard, and can be found there at the community level.
| Section | Source |
|---|---|
| Population, housing | VISION model, CSD-level, SSP1–SSP5, 2020–2050 |
| Commuting & roads | Statistics Canada 98-10-0459-01 (2021 census commuting matrix), re-fit to each scenario; VISION road network volume/capacity |
| Water demand & capacity | VISION water consumption model; Canada's Core Public Infrastructure Survey 2022 (treatment capacity) |
| Climate | ClimateData.ca / PCIC CanDCS-M6, bias-adjusted CMIP6 ensemble median, national CSD coverage |
| Infrastructure condition & cost | Canada's Core Public Infrastructure Survey 2022; Statistics Canada 38-10-0271-01 (water losses) |
Community selections ("spotlights") are chosen by the largest or most extreme value in each category, with a minimum population or network size applied so a single short road segment in a small community can't stand in for a real congestion story. Full methodology, known data-quality issues and every threshold used are documented in the full dashboard's Data & Methods tab and in docs/NIA2_REPORT2_PIPELINE.md.