Overview
ClimateRiskCheck produces property-level physical climate risk reports for Canadian residential and commercial addresses. Each report assesses flood and wildfire exposure using government open data, supplements with regional climate projections and building characteristics, and assigns an overall risk rating.
All data is sourced exclusively from Canadian federal and provincial government datasets under open government licences. No proprietary third-party risk models are used. Every report carries a data provenance table identifying the exact dataset version used at generation time.
Reports are intended for use as an informational screening tool for portfolio due diligence, lending decisions, governance documentation, and audit support. They are not a substitute for a professional flood study, environmental assessment, or insurance underwriting opinion.
Geocoding & Location Accuracy
Addresses are geocoded against a national address database, restricted to Canada. Each result carries a precision level that is mapped to a confidence label shown on every report.
| Confidence Label | Meaning |
|---|---|
| ✓ Rooftop-level | Precise match to a specific building or street address. Highest accuracy. |
| ⚠ Neighbourhood centroid | Matched to the centre of a postal code or neighbourhood. May be 50–500 m from the actual property. |
| ✗ Approximate | Low-precision match. Reports at this confidence level are rejected before a credit is consumed. |
When confidence is Neighbourhood centroid, the report displays a prominent warning. In flood risk assessment, 50 metres can be the difference between a property inside and outside a flood zone. Lenders should verify the geocoded coordinates on the attached map before relying on flood risk scores for centroid-matched addresses.
City mismatch detection runs automatically. If the city parsed from the submitted address differs from the locality returned by the geocoder (excluding known amalgamation aliases such as Toronto, North York, Scarborough), a separate verification warning is added to the report.
Flood Risk Assessment
Ontario: Conservation Authority regulatory floodplain maps (TRCA, Peel, LSRCA, RVCA, CVC, MVCA, NPCA, GRCA, HCA, GSCA, NVCA, SVCA, MVC, SNC, KRCA, ORCA, CLOCA) — the maps municipalities use for development approvals. Contains information made available under Grand River Conservation Authority's Open Data Licence v2.0. Alberta: Flood Hazard Identification Program (Government of Alberta; Open Government Licence – Alberta). Saskatchewan: Water Security Agency flood hazard areas (Government of Saskatchewan Standard Unrestricted Use Data Licence v2.0). Newfoundland and Labrador: Water Resources Management Division flood risk mapping (Open Government Licence – Newfoundland and Labrador). British Columbia: Mapped Floodplains in BC (Historical) (Open Government Licence – British Columbia), City of Vancouver designated floodplain (Open Government Licence – Vancouver), City of Kamloops 200-year floodplain (Open Government Licence – Kamloops), City of Langley 200-year flood extent (Open Government License – City of Langley), City of Surrey 200-year floodplain (Open Government License – Surrey) and City of Abbotsford floodplains (Open Government Licence – City of Abbotsford). Quebec: Base de données des zones inondables (MELCCFP), Ville de Gatineau and Ville de Québec flood zones (all Creative Commons Attribution 4.0). NRCan's Canada Flood Map Inventory is used only as a coarse fallback. Reports carry the attribution each licence requires. Where each source applies is shown under Geographic Coverage below.
Scoring logic
The geocoded coordinate (lat/lon) is spatially joined against flood zone polygons using a two-pass approach: detailed CA portal data is queried first; if no detailed polygon intersects, a supplementary check against coarse federal CFM polygons is made before concluding Low.
Pass 1 — Detailed CA data (250 m search radius): If the point falls within a Conservation Authority polygon, the hazard zone label, estimated inundation depth, and return period are extracted and scored directly.
Pass 2 — Coarse CFM supplementary check: If no detailed polygon intersects but detailed CA data exists within 20 km (confirming the region is mapped), the scoring engine checks whether the property falls inside a federal Canada Flood Map Inventory (CFM) polygon with area < 100 km². CFM polygons are classified as coarse because they lack depth attributes and use broader delineation methods than CA regulatory maps. If a qualifying CFM polygon is found, the property is rated Moderate with a caveat noting the coarse data source and recommending verification with the local CA. The 100 km² threshold filters out broad regional CFM polygons that are not meaningful at the property level.
Jurisdiction guard: If no CA data exists within 20 km, or if the property falls within a known no-coverage jurisdiction (see Known Limitations), a Proxy Assessment or Coverage Insufficient result is returned regardless of CFM coverage.
British Columbia and Quebec: the openly available flood maps in these provinces cover only part of each province (in Quebec, Greater Montréal's maps are not open data), so nearby data is not treated as evidence about a property there. A property is rated only when it is within 250 m of a mapped floodplain; otherwise the flood result is Coverage Insufficient. This prevents, for example, Richmond or Burnaby addresses — which have floodplains but no loaded mapping — from reading Low simply because Vancouver's mapping is nearby.
| Rating | Criteria |
|---|---|
| Low | No flood zone polygon intersects the property (detailed CA data present within 20 km and no qualifying CFM polygon found), or estimated inundation depth < 0.3 m. |
| Moderate | Property within a mapped flood zone with estimated depth 0.3 m – 1.0 m, zone-attributed medium hazard (no depth attribute), or property inside a coarse CFM polygon (< 100 km²) where detailed CA data exists nearby but does not intersect. |
| High | Property within a mapped flood zone with estimated depth 1.0 m – 2.0 m, or zone-attributed high hazard (no depth attribute). |
| Extreme | Estimated inundation depth ≥ 2.0 m within a mapped flood zone. |
| Partial Assessment | No detailed flood zone data exists within 20 km of the property, or property is in a known no-coverage jurisdiction. Coverage reason is noted on the report (e.g. Proxy Assessment, No Regulatory Study Available). |
Return period
Where available, the flood zone polygon carries a return period (e.g. 1:100 year event). This represents the statistical annual probability of exceedance (a 1:100 year flood has a 1% chance of occurring in any given year, and a 26% chance of occurring over a 30-year mortgage).
Flood data confidence
Every report states a Flood Data Confidence level, derived from which data source produced the flood rating. It is an output field only — it never affects the flood rating or the overall rating.
| Confidence | Meaning |
|---|---|
| High | Rated from detailed Conservation Authority flood polygons at this location (including a Low rating where detailed coverage exists and the property falls outside all hazard polygons). |
| Moderate | Rated from a coarse federal Canada Flood Map Inventory polygon; not confirmed by detailed Conservation Authority mapping at this location. |
| Regional only | No detailed or coarse flood mapping at this location — the NRCan Flood Susceptibility Index regional screening signal is the only available flood signal. Regional context, not a property-level assessment. |
| None | No flood data available at any resolution for this location. |
Wildfire Exposure
Digitized perimeters of all Canadian wildfire events from 2010–2024. Used to calculate proximity of the subject property to the nearest historical burn area.
Gridded 90th-percentile Fire Weather Index values derived from 2010–2024 daily observations. FWI integrates temperature, relative humidity, wind speed, and precipitation into a composite measure of fire danger. The P90 value represents severe fire weather conditions.
Scoring logic
Burn distance and FWI P90 are each scored independently to Low / Moderate / High. The final wildfire rating is the higher of the two signals. Wildfire tops out at High — there is no Extreme category for wildfire (Extreme is flood-only, requiring depth ≥ 2.0 m).
| Signal | Low | Moderate | High |
|---|---|---|---|
| Nearest historical burn | > 5 km | 1 – 5 km | < 1 km |
| FWI P90 (2010–2024) | ≤ 20 | 21 – 35 | > 35 |
Final rating = higher of the two signals. Example: burn at 3 km (Moderate) + FWI P90 of 38 (High) → overall wildfire rating High.
2050 Climate Risk Outlook v0.20
Every report includes a forward-looking 2050 Climate Risk Outlook section providing directional hazard trajectory signals to mid-century, consistent with OSFI Guideline B-15's requirement to assess physical climate risk horizons to 2050 and beyond. This section is supplementary — it does not affect the current-conditions flood, wildfire, or overall risk ratings.
ECCC's Downscaled Climate Scenarios dataset, providing projected changes in mean temperature, precipitation, and extreme heat days to 2050 under SSP2-4.5 (moderate emissions) and SSP5-8.5 (high emissions) — the CMIP6 counterparts of RCP 4.5 and 8.5. Values are averaged over the census division that contains the property (all 293 census divisions in Canada).
Trajectory Classification
| Hazard | ↑ Increasing | → Broadly Stable | ↓ Decreasing |
|---|---|---|---|
| Flood | Precip change > +5% | Precip change −5% to +5% | Precip change < −5% |
| Wildfire | SSP5-8.5 temp rise ≥ 2.0°C or hot days ≥ 8 | Neither threshold met | — |
| Urban Heat | SSP5-8.5 hot days ≥ 5 | Fewer than 5 additional hot days | — |
Wildfire and heat trajectories, and the precipitation change behind the flood trajectory, use the SSP5-8.5 (high emissions) scenario as the planning-relevant horizon. An "Increasing" result for both is expected for virtually all addresses in Southern Ontario under current projections.
Supporting Statistics
- Temperature rise (°C): Projected mean annual temperature increase vs. 1981–2010 baseline, SSP2-4.5 and SSP5-8.5
- Precipitation change (%): Projected change in mean annual precipitation by 2050
- Additional hot days: Additional days per year exceeding 30°C by 2050, SSP2-4.5 and SSP5-8.5
Limitations
- Trajectories are regional-scale (census division), not property-level predictions.
- Values are census-division averages of a ~10 km grid; very large northern divisions average over wide climatic ranges.
- Additional hot days are interim figures from our previous regional dataset while the licence for the CanDCS-U6 hot-day index is confirmed with Environment and Climate Change Canada.
- NRCan FSI future-scenario rasters are not yet publicly available; the flood trajectory uses precipitation change only, not a direct susceptibility index projection.
Building Information
Statistics Canada's national building footprint dataset, containing year of construction, building type, number of floors, and footprint geometry, compiled from municipal and provincial open data. We load it for every province and territory except Nunavut (StatCan publishes no Nunavut file): 14.4 million footprints in total, including Ontario (5.7 million), Quebec (3.7 million), Alberta and British Columbia (1.3 million each). Coverage within a province depends on which municipalities contributed data — for example Alberta has 64 municipalities in the source, so many smaller Alberta towns are absent.
Matching logic
The nearest building footprint within 100 m of the geocoded coordinate is matched to the subject property. Where no building is found within that radius, the report notes the absence and recommends on-site verification. Building data is informational only and does not affect the flood or wildfire risk rating.
Attribute sourcing and floor estimation
Year of construction, building type, and floor count are sparsely populated in the source dataset — the majority of records contain no values for these fields. These are source-data gaps, not assessment errors, and are noted on the report where they occur.
Where floor count is absent but building height is available, the number of storeys is estimated at one storey per 3.0 metres of height, reflecting the Canadian residential standard. Estimated floor counts are labelled "estimated from building height" in the report to distinguish them from directly-sourced values.
Urban Sewer Flood Signal Toronto only
Toronto Water's Basement Flooding Study Areas dataset delineates the 67 sanitary subsewersheds where the City has conducted or is conducting a formal Basement Flooding Protection Study. Each polygon represents a sewer drainage catchment where documented basement flooding complaints were serious enough to trigger a structured engineering study and potential capital works program.
This dataset captures sewer backflow and combined-sewer overflow (CSO) risk — the dominant flood mechanism in older Toronto neighbourhoods — which is not captured by the federal Canada Flood Map Inventory or Conservation Authority portal data, both of which focus on riverine and surface flooding.
Signal logic
For properties within the City of Toronto administrative boundary, a spatial containment check is run against the 67 study area polygons. The result is a binary supplementary signal:
| Signal | Meaning |
|---|---|
| Elevated | Property falls within a designated study area. The subsewershed has a documented history of basement flooding complaints and is subject to a formal Toronto Water engineering study. Sewer backflow risk is elevated independently of river flood status. |
| Not Detected | Property is within Toronto but not inside a BFA polygon. Risk is reduced but not absent — BFA boundaries reflect subsewersheds formally studied, not all areas of potential risk. |
| Not Applicable | Property is outside the City of Toronto. This check does not apply. |
The urban sewer flood signal is supplementary only — it does not affect the overall risk rating, which is derived exclusively from flood zone (CFM/CA) and wildfire signals. It is presented as an additional data point for lenders and underwriters assessing Toronto properties.
Overall Risk Rating
The overall rating is derived from the flood and wildfire sub-ratings using a conservative aggregation: the overall rating equals the highest (worst) sub-rating, with a specific upgrade rule when both signals are elevated.
| Overall Rating | Derivation |
|---|---|
| Low | Both flood and wildfire are Low. |
| Moderate | The highest individual sub-rating is Moderate (neither flood nor wildfire reaches High). |
| High | The highest individual sub-rating is High (flood or wildfire, or both). |
| Extreme | Flood rating is Extreme (depth ≥ 2.0 m). Wildfire cannot produce Extreme — it is a flood-only outcome. |
| Partial Assessment | One or more dimensions returned Coverage Insufficient due to data gaps. Individual sections explain why. |
The overall rating is the higher of the flood and wildfire sub-ratings, using the order Low < Moderate < High < Extreme. If either sub-rating is Coverage Insufficient, the overall is forced to Partial Assessment.
Expected Annual Loss (EAL)
EAL translates a flood hazard rating into a dollar-denominated annual loss figure — the metric used by insurers and increasingly expected in OSFI B-15 physical risk disclosures. It is computed only when all three inputs are available: a Moderate, High, or Extreme flood rating; a flood depth estimate (metres); and a property value.
Formula
EAL = (1 / return period) × structural damage % × property value
Where return_period is the recurrence interval of the design flood event (typically 100 years), damage_pct is the fraction of structural value destroyed at that flood depth, and property_value is the assessed or estimated value of the structure.
Example: a $800,000 property at a 1-in-100-year event with 0.9 m depth → damage ≈ 21% → loss at event ≈ $168,000 → EAL ≈ $1,680/year.
Depth-damage curves
Structural damage percentages are derived from the Public Safety Canada / IBI Group residential depth-damage curves for single-family dwellings, which underpin the Canada Flood Risk Finder (Public Safety Canada, 2026). The curve maps inundation depth to percentage of structural replacement value lost:
| Depth (m) | Structural damage (%) |
|---|---|
| 0.00 | 0% |
| 0.15 | 4% |
| 0.30 | 8% |
| 0.60 | 15% |
| 0.90 | 21% |
| 1.20 | 27% |
| 1.50 | 33% |
| 1.80 | 40% |
| 2.10 | 47% |
| 2.40 | 53% |
| 3.00 | 63% |
| 4.00 | 75% |
Values between table entries are linearly interpolated. Depths above 4.0 m are capped at 75%. The curves represent single-detached residential construction typical of Canadian suburbs; they may underestimate damage for older pre-1960 construction or overestimate it for newer flood-resilient builds.
Representative depth for zone-only flood zones
Conservation Authority flood mapping is zone-only — a regulatory hazard classification with no measured depth. Where a flood zone carries no depth attribute, EAL uses a fixed representative depth by hazard tier, anchored to the Ontario Ministry of Natural Resources Technical Guide — River and Stream Systems: Flooding Hazard Limit (2002), which defines the floodway / flood-fringe boundary at a flood depth of 1 metre ("flow depth in excess of 1 m and/or flow velocities above 1 m/s can create significant hazards for developments"):
| Hazard tier | Regulatory zone | Representative depth | Basis |
|---|---|---|---|
| High | Floodway / One Zone / Special Policy Area | 1.0 m | Floodway is defined as depth > 1 m; 1.0 m is the conservative floor of that band. |
| Moderate | Flood fringe (Two Zone Fringe) | 0.5 m | Flood fringe is the 0–1 m band below the floodway threshold; 0.5 m is the band midpoint. |
| Extreme | (depth-driven only) | 2.0 m | Extreme band threshold — not produced by zone-only data. |
These depths are representative, not measured, and every report derived this way is labelled accordingly. The MNRF criterion fixes the band edges; the flood-fringe estimate (0.5 m) is the band midpoint and may be refined against a federal depth-frequency distribution if one is adopted. Commercial modelled-depth products are not used as the published basis.
Property value sourcing
A property value can be supplied at the time of the report request. When not provided, ClimateRiskCheck uses the Statistics Canada 2021 Census median dwelling value for the census division that contains the property — all 293 census divisions in every province and territory — as a regional proxy. The value used and its source (client-supplied or regional median) are clearly stated on every report.
Scope and limitations
EAL covers structural damage only. It does not include: contents loss, temporary accommodation costs, business interruption, remediation or soil remediation, or landscape/driveway damage. A full economic loss estimate would typically be 1.5–2.5× the structural EAL.
EAL is computed from a single design event (the 1-in-100-year flood) and does not integrate across the full exceedance probability curve. A multi-return-period integration would yield a more precise annualized figure; this approximation is appropriate for screening-level due diligence.
Wildfire EAL is not computed in the current version — wildfire damage functions for Canadian residential construction are not yet standardized in public data sources.
2050 Climate-Adjusted EAL v0.21
When both the current EAL and CanDCS-U6 temperature projections are available, the report includes a forward-looking flood loss estimate for 2050 under two IPCC emissions pathways (SSP2-4.5 — moderate emissions, and SSP5-8.5 — high emissions).
Method — Clausius-Clapeyron Scaling
Extreme precipitation intensity scales with atmospheric temperature at approximately +7% per °C of warming — the ECCC-endorsed midpoint for Canadian mid-latitude stations (Mailhot et al. 2007; Cheng et al. 2014). As the climate warms, a historical 1-in-T-year rainfall event becomes more frequent, compressing its effective return period.
Adjusted return period = T / 1.07ΔT
Adjusted EAL = (1 / adjusted return period) × structural damage % × property value
where ΔT is the CanDCS-U6 ensemble median temperature rise by 2050 relative to the 1981–2010 baseline, drawn from the nearest census division centroid.
IDF Data Source
Precipitation intensity baseline comes from the nearest Environment and Climate Change Canada (ECCC) Engineering Climate Datasets IDF station — all provinces and territories, 2,288 stations, Open Government Licence. Stations beyond 100 km are excluded.
The ECCC IDF values represent statistically homogenized historical records — the Clausius-Clapeyron scaling layer adds the forward-looking climate adjustment on top.
What Is Shown
The report displays two adjusted EAL figures alongside the current baseline EAL: one for SSP2-4.5 and one for SSP5-8.5, together with the effective return period for each scenario and the name of the ECCC IDF station used. Structural damage assumptions and property value are held constant — only the return period changes. A named EAL multiplier (e.g. 1.4× / 1.9×) is derived from the ratio of adjusted EAL to base EAL and displayed as a single headline metric — suitable for board reporting and OSFI Guideline B-15 mid-century physical risk disclosure. The multiplier is shown whenever adjusted EAL is available, regardless of the CanDCS-U6 flood trajectory direction.
Limitations
- A station must be within 100 km; very remote locations do not receive an adjusted EAL.
- The 7%/°C scaling factor is a regional midpoint; local topographic and storm-track effects may cause actual changes to deviate from this rate.
- CanDCS-U6 ΔT values are ensemble medians at the census division level, not property-level projections.
- Flood depth is not re-estimated under future climate; structural damage percentage is held at its current-conditions value.
- This is a single-design-event approximation — the same limitation that applies to the base EAL.
Geographic Coverage
What a report contains depends on where the property is. Find the province or area below and read across. Coverage follows what governments publish as open data — where official flood mapping is not available, the report says so (Coverage Insufficient) rather than guessing, and the overall result is a Partial Assessment.
✅ Available · ⚠️ Available in the places listed · — Not available
| Where | Flood rating | Flood susceptibility | Wildfire | Urban heat | Sewer backup | Building info |
|---|---|---|---|---|---|---|
| Ontario — Greater Toronto Toronto, York, Peel, Durham, Barrie | ✅ Conservation Authority mapping (TRCA, Peel, CVC, LSRCA, CLOCA) | ✅ | ✅ | ✅ | ✅ City of Toronto | ✅ |
| Ontario — Ottawa incl. Kanata, Orléans, Prescott-Russell | ✅ Conservation Authority mapping (RVCA, MVC, SNC) | ✅ | ✅ | ✅ | — | ✅ |
| Ontario — other mapped areas Hamilton, Burlington, Kitchener-Waterloo, Cambridge, Guelph, Brantford, Niagara, Peterborough, Kawartha Lakes, Grey–Bruce, Simcoe | ✅ Conservation Authority mapping | ✅ | ✅ | ⚠️ Hamilton, Kitchener-Waterloo, Cambridge, Guelph, Brantford, Niagara, Peterborough | — | ✅ |
| Ontario — Halton Oakville, Milton, Halton Hills | — Conservation Halton does not publish its mapping as open data | ✅ | ✅ | ✅ | — | ✅ |
| Ontario — not yet mapped London, Windsor, Elgin County, Northern Ontario | — Not available as open data yet | ✅ | ✅ | — | — | ✅ |
| Alberta | ⚠️ Provincial mapping along studied rivers: Calgary, Edmonton, Red Deer, Lethbridge, Medicine Hat, Fort McMurray, Canmore, Drumheller, High River | ✅ | ✅ | ⚠️ Calgary, Edmonton | — | ⚠️ 64 municipalities (many small towns absent) |
| Saskatchewan | ⚠️ 23 communities incl. Saskatoon, Regina, Prince Albert, Moose Jaw, Swift Current, Weyburn, Estevan (1970s–90s maps, river flooding only) | ✅ | ✅ | — | — | ✅ |
| British Columbia | ⚠️ Vancouver, Kamloops, Surrey and Abbotsford floodplains (incl. Sumas Prairie), the City of Langley's Nicomekl valley, and 1976–97 provincial floodplains (e.g. Squamish, Prince George, Grand Forks, Merritt). Not yet: Richmond, Burnaby, Township of Langley, Victoria, Kelowna | ✅ | ✅ | ⚠️ Metro Vancouver | — | ✅ |
| Newfoundland and Labrador | ⚠️ 32 study areas incl. St. John's, Corner Brook, Grand Falls-Windsor, Happy Valley–Goose Bay | ✅ | ✅ | — | — | ✅ |
| Quebec | ⚠️ Provincial flood-zone database plus Gatineau and Québec City mapping (e.g. Québec City, Gatineau, Sherbrooke, many river towns). Not covered: Montréal, Laval, Longueuil (mapping not open data) and places without a provincial study | ✅ | ✅ | — | — | ✅ |
| Manitoba, New Brunswick, Nova Scotia, Prince Edward Island | — Not yet | ✅ | ✅ | — | — | ✅ |
| Yukon, Northwest Territories, Nunavut | — Not yet | ✅ | ✅ | — | — | ⚠️ Yukon and NWT; no source data for Nunavut |
Flood rating comes only from official regulatory flood mapping, and a property must be within 250 m of a mapped flood area to be rated from it. In British Columbia, where the available maps are partial, a property away from a mapped floodplain is Coverage Insufficient, never Low. Flood susceptibility (NRCan, 300 m) is a regional screening signal available for every Canadian address; it never replaces the flood rating, and a low reading is not evidence of low flood risk. Wildfire uses national Canadian Wildland Fire Information System data. Building info is informational and depends on which municipalities contributed data to Statistics Canada. Regional 2050 climate projections are included for every report, from the nearest climate region.
Ontario flood mapping comes from these Conservation Authorities: TRCA, Credit Valley (CVC), Lake Simcoe (LSRCA), Central Lake Ontario (CLOCA), Rideau Valley (RVCA), Mississippi Valley (MVC), South Nation (SNC), Hamilton (HCA), Grand River (GRCA), Niagara Peninsula (NPCA), Otonabee (ORCA), Kawartha (KRCA), Grey Sauble (GSCA), Nottawasaga Valley (NVCA), Saugeen Valley (SVCA) and Maitland Valley (MVCA), plus Peel Region. Every source and its licence is described in its own section above, and each report's Data Provenance table records the exact version it used.
Known Limitations
- London (UTRCA) — UTRCA approved updated flood mapping for the City of London in May 2026, but publishes it through an online viewer only, not as open data. The Regulation Limit layer it does publish is not a floodline and carries a restrictive licence. We have asked UTRCA to release the floodplain layer under its existing open data licence.
- Elgin County (LPRCA, LTVCA, KCCA, CCCA) — data exists but portals require further investigation; no open floodline layer confirmed as of June 2026.
- Northern Ontario — no Conservation Authority open flood mapping coverage.
Addresses in these regions receive Coverage Insufficient with the NRCan Flood Susceptibility Index regional screening signal where available (Flood Data Confidence = "Regional only") or no flood signal at all ("None"). These gaps are re-checked periodically as Conservation Authorities expand their open data programmes.
Version History
| Version | Date | Changes |
|---|---|---|
| v0.40 | September 2026 | Added official floodplain maps for three Lower Mainland cities: Surrey's 200-year floodplain, Abbotsford's floodplains (including Sumas Prairie, flooded in 2021) and the City of Langley's 200-year flood extent along the Nicomekl River. Properties inside them rate Moderate, like other BC floodplains. Properties outside them in BC still show Coverage Insufficient rather than Low. The Township of Langley does not publish a floodplain map yet. No other results changed. |
| v0.39 | September 2026 | Reports for Greater Montréal (Montréal, Laval, Longueuil and the north and south shores) now explain why flood shows Coverage Insufficient: official flood maps exist there but are not yet published under a licence that lets us use them, and we have requested access. The note points to the Quebec government's public flood-zone map. You can also ask to be emailed when flood coverage reaches your city. No ratings changed. |
| v0.38 | September 2026 | Added official flood mapping for Quebec: the provincial flood-zone database, plus Gatineau's and Québec City's own flood zones. A property in a 0–20-year flood zone, or a Québec City ice-jam zone, rates High; in a 20–100-year zone, Moderate. This covers Québec City, Gatineau, Sherbrooke and many towns along mapped rivers. Greater Montréal is not covered: its regional flood mapping is not licensed for commercial use, and Quebec's new provincial maps are not yet published as data. As in British Columbia, a Quebec property away from mapped flood zones shows Coverage Insufficient rather than Low. No other province's results changed. |
| v0.37 | September 2026 | Reports without official flood mapping now state the distance to the nearest mapped watercourse or water body almost anywhere in southern Canada, using Natural Resources Canada's National Hydro Network. Previously this covered only part of the Toronto–Hamilton area. We also fixed three problems with the old calculation: large rivers and lakes were missing, most streams in western Canada were excluded by a data-coding issue, and distances were overstated east–west. For example, a Richmond, BC property previously read 21 km from water and now reads 710 m. This note is supplementary and does not affect any rating. |
| v0.36 | September 2026 | The 2050 Climate Risk Outlook now uses projections calculated for the census division each property is in — all 293 across Canada — directly from Environment and Climate Change Canada's CanDCS-U6 data. Previously it used the nearest of 48 regions, so, for example, a St. John's property received Halifax's projections. The previous figures also understated moderate-scenario warming in most places (Toronto: +2.1°C → +2.8°C), so 2050 temperature figures and the climate-adjusted loss multiplier are somewhat higher. We also corrected the scenario names: this dataset uses SSP2-4.5 and SSP5-8.5 (the modern equivalents of RCP 4.5 and 8.5), not RCP scenarios as previously stated. Additional-hot-day figures are unchanged for now while we confirm the licence for the new hot-day data. No risk rating changed. |
| v0.35 | September 2026 | Urban heat island coverage extended to Metro Vancouver (Vancouver, Burnaby, Richmond, Surrey, the North Shore, Coquitlam, Langley), Calgary and Edmonton. Each city layer averages 6–9 cloud-free summer 2025 Landsat scenes, with clouds removed before averaging, using the same thresholds as the Ontario cities. Parkland reads cooler than downtown in all three (for example Stanley Park 23.5°C against 30.4°C downtown Vancouver). The heat signal never affects a risk rating. |
| v0.34 | September 2026 | Expected Annual Loss now uses local inputs in every province. When no property value is supplied, the estimate uses the 2021 Census median dwelling value for the census division the property is in — all 293 census divisions across Canada. Previously it used the nearest of 48 regions, so a St. John's property, for example, was valued at Halifax's median. Values in areas that were already covered also move to the official census figures, so EAL amounts change slightly everywhere. The 2050 climate-adjusted EAL, which previously worked only in Ontario, now uses Environment and Climate Change Canada rainfall stations in every province (2,288 stations). No risk rating changed. |
| v0.33 | September 2026 | Added British Columbia flood mapping from openly licensed sources: the Province of BC's designated floodplains and alluvial fans (1976–97), the City of Vancouver's designated floodplain, and Kamloops' 200-year floodplain. None of these maps separates floodway from fringe, so a property inside one rates Moderate. BC's coverage is partial, so we also changed how BC properties are rated: a BC property is rated only when it is within 250 m of a mapped floodplain; otherwise the flood result is Coverage Insufficient. Without this, cities such as Richmond, Burnaby and Surrey — which have floodplains but no mapping we can use yet — would have shown Low just because Vancouver's mapping is nearby. The province's newer 2024–26 flood maps are not yet available for reuse. We checked that no other province's results changed. |
| v0.32 | September 2026 | Added official flood mapping for Saskatchewan and for Newfoundland and Labrador, and building information for every province and territory except Nunavut. Saskatchewan: Water Security Agency mapping for 23 communities, including Saskatoon, Regina, Prince Albert and Moose Jaw. Floodway rates High and flood fringe Moderate. Where the province mapped only the overall 1-in-500 flood area, with no floodway/fringe split, properties inside it rate Moderate. This mapping dates from the 1970s–1990s and covers river flooding only. Newfoundland and Labrador: 1-in-20 floodway (High) and 1-in-100 fringe (Moderate) for 32 study areas, including St. John's and Corner Brook; climate-change scenario areas are not used. Building footprints now cover 14.4 million buildings nationwide. We checked that none of this changed any Ontario or Alberta flood data. Reports using the new sources carry the attribution each licence requires. No rating thresholds or scoring rules changed. |
| v0.31 | September 2026 | Corrected the wording shown for Alberta locations that no flood study covers, such as Hanna. Since Alberta's full provincial flood mapping was added in v0.29, these messages wrongly said mapping for the area "exists but hasn't been loaded". They now say that Alberta's provincial flood mapping does not include a study covering the location. We also fixed the coverage message for western Alberta (west Calgary, Canmore, Banff, Grande Prairie), which was previously worded as if the location were in British Columbia. Wording only; no ratings, thresholds or scoring rules changed. |
| v0.30 | September 2026 | Added Alberta building footprints from Statistics Canada's Open Database of Buildings: 1.3 million footprints across 64 Alberta municipalities, including Calgary, Edmonton, Red Deer, Lethbridge, Medicine Hat, Wood Buffalo and Canmore. Alberta reports now show building information where the source dataset covers the location; many smaller towns are not in the source. Corrected this page's earlier statement that building coverage was strongest in "Ontario and British Columbia" — British Columbia buildings have never been loaded. When no building is found, reports now explain that this reflects dataset coverage rather than suggesting the address may be wrong. Building attributes are informational only; no rating thresholds or scoring rules changed. |
| v0.29 | September 2026 | Added Alberta's official flood mapping, from the provincial Flood Hazard Identification Program (Government of Alberta, Open Government Licence – Alberta). Properties along mapped Alberta rivers — including Calgary, Edmonton, Red Deer, Lethbridge, Medicine Hat, Fort McMurray, Canmore and High River — now receive a flood rating: High in a floodway, Moderate in a flood fringe, Low in a mapped area outside the floodplain. Alberta uses the same 1-in-100 standard as Ontario. The provincial dataset also includes larger, rarer-flood scenario areas; we deliberately do not use these, so every rating reflects the same regulatory standard. Rural Alberta locations with no study nearby continue to receive Coverage Insufficient. We checked that adding Alberta changed no Ontario data. Reports that use Alberta data now carry the attribution the Alberta licence requires, and reports now name eight Ontario Conservation Authorities that were previously listed only as "CA flood polygons". No rating thresholds or scoring rules changed. |
| v0.28 | September 2026 | Extended the NRCan Flood Susceptibility Index — a regional flood screening signal — from Southern Ontario to all of Canada, including the territories. Until now, addresses outside Southern Ontario received no flood information at all; they now receive this regional signal, while the detailed flood rating continues to rely on regulatory floodplain mapping where it is available. We also measured how reliable the signal is: at locations inside officially mapped floodplains in Alberta and Ontario, about one in four still score Low on this index. It is useful for screening, but a Low reading is not evidence of low flood risk, and every report now says so directly. The way the signal is aggregated was also made exact, which shifts about 9% of Ontario susceptibility readings by one band; no flood rating or overall risk rating changes as a result. We corrected an error in which reports generated since August were labelled with an out-of-date methodology version; that label is now checked automatically. We also corrected several descriptions: the report PDF now uses the same wording as this page for the flood susceptibility signal; its data sources table now lists every heat layer in use and the Central Lake Ontario Conservation Authority; reports with no flood mapping no longer name "CA flood polygons" as their source; and our internal methodology document now describes the two-step flood check (a 250 m search, then a 20 km coverage check) that this page already described. No rating changed. |
| v0.27 | September 2026 | Corrected a coverage gap in the Greater Toronto Area heat layer. The GTA spans two satellite imaging tracks, but scene selection was picking all three scenes from the same track — so roughly 40% of the area, including downtown Toronto, had no composite data and fell back to an older 2020 single-date reference layer that runs materially hotter. Scene selection now draws from every satellite track covering a city (the GTA layer now averages 8 scenes across 4 tracks). Effect on published results: a downtown Toronto property that previously read High (around 48°C) may now read Moderate (around 39°C). The new value is the better estimate — the land-cover gradient is now physically ordered, with forest canopy coolest and airport tarmac hottest, where previously a forest interior read hotter than an industrial area. Temperature thresholds are unchanged and no risk rating logic changed. Other cities were checked and serve their areas correctly. |
| v0.26 | August 2026 | Corrected cloud contamination in the Urban Heat Island data. Each city's land surface temperature layer averages three summer satellite scenes; previously, pixels obscured by cloud were included in that average. Cloud tops are genuinely cold, so those pixels pulled the average down and understated the heat reading — in the worst cases a property in a dense, paved area could be shown as Low Heat when it should have read Moderate or High. All scenes are now screened against the satellite's own cloud-quality band (cloud, cloud shadow, and cirrus removed) before averaging, and every city layer has been regenerated. Implausibly cold pixels fell from 11.0% to 1.9% in the Kitchener-Waterloo area, 10.6% to 1.6% in Hamilton, and 3.3% to 0.0% in the GTA. The error only ever understated heat — no report overstated it. Temperature thresholds are unchanged (Low below 30°C, Moderate 30–40°C, High above 40°C), and no risk rating or scoring logic changed. Also documented here for completeness: since June 2026 the heat layers have been three-scene composites rather than single satellite passes, and the GTA has been served by a current Landsat composite rather than the 2020 TRCA raster (now used only where the composite has no data). Urban Heat Island coverage now includes all of Halton — Oakville, Milton, and Halton Hills — served by the Hamilton layer. This affects the heat signal only; flood coverage in Conservation Halton is unchanged. |
| v0.25 | July 2026 | Refined the Conservation Halton jurisdiction check from a rectangular bounding box to the actual Conservation Halton administrative boundary (Ontario GeoHub / Land Information Ontario, Open Government Licence – Ontario). The previous box extended about 10 km east of Halton's true edge into Mississauga, so some Peel Region / Credit Valley Conservation addresses were shown as "Proxy Assessment" when they should have been scored from the neighbouring authority's data. Those addresses now receive their correct flood rating. Properties genuinely within Conservation Halton (Oakville, Milton, north Burlington) continue to receive a Proxy Assessment, because Conservation Halton does not publish its regulatory floodplain mapping as open data. |
| v0.24 | July 2026 | Recalibrated the representative flood depth used to estimate EAL on zone-only flood zones: flood-fringe (Moderate) depth raised from 0.3 m to 0.5 m, on the basis of the Ontario MNRF Technical Guide — River and Stream Systems: Flooding Hazard Limit (2002), which places the floodway / flood-fringe boundary at 1 m depth (flood fringe = 0–1 m band → 0.5 m midpoint). Floodway (High) representative depth unchanged at 1.0 m. Affects the EAL dollar figure on Moderate-rated zone-only properties only; no rating, coverage, or threshold changes. |
| v0.23 | June 2026 | Added CLOCA Regulated Area (Central Lake Ontario Conservation Authority) flood data covering Ajax, Whitby, Oshawa, Clarington, and Pickering (Durham Region east). Single dissolved multi-polygon (78 km²) including Lynde Creek Phase 2 (FHIMP-funded, completed 2025). Zone-only; properties within the regulated area are rated a maximum of High. CLOCA removed from Known Limitations. No scoring threshold or logic changes. |
| v0.22 | June 2026 | Added 2050 EAL Multiplier as a named output. Ratio of adjusted EAL to base EAL (e.g. 1.4× / 1.9× for RCP 4.5 / 8.5) displayed in the EAL section. Forward-looking note now shown whenever adjusted EAL is available, regardless of flood trajectory direction. No scoring logic changes. |
| v0.21 | June 2026 | Added 2050 Climate-Adjusted EAL. Clausius-Clapeyron precipitation scaling (+7%/°C, ECCC-endorsed midpoint) applied using the nearest ECCC IDF station (RASM2024, Ontario, 429 stations, Open Government Licence) and CanDCS-U6 temperature projections. Produces an adjusted return period and adjusted EAL range for RCP 4.5 and RCP 8.5. Supplementary — does not affect any current-conditions rating. |
| v0.20 | June 2026 | Added 2050 Climate Risk Outlook section. Directional hazard trajectory signals (↑ Increasing / → Broadly Stable / ↓ Decreasing) for Flood, Wildfire, and Urban Heat, derived from CanDCS-U6 ensemble median projections at the nearest census division. Flood trajectory from precipitation change vs. baseline; wildfire from RCP 8.5 temperature rise (≥2.0°C) or additional hot days (≥8); heat from RCP 8.5 additional days above 30°C (≥5). Supplementary only — does not affect any current-conditions rating. |
| v0.19 | June 2026 | Added four Conservation Ontario flood datasets: MVC Regulatory Floodplain (Mississippi Valley CA — west Ottawa, Kanata, Carp River corridor — 234 polygons), SNC Regulation Area (South Nation Conservation — east Ottawa, Prescott-Russell — 175 polygons), KRCA Regulated Areas (Kawartha Region CA — Lindsay, Kawartha Lakes, Fenelon Falls, Bobcaygeon — 1,837 polygons), and ORCA Floodplain (Otonabee Region CA — Peterborough, Lakefield, Selwyn — 79 polygons). All zone-only; all properties within a regulated area are rated a maximum of High. No scoring threshold or logic changes. |
| v0.18 | June 2026 | Added three Grey County Conservation Authority flood datasets: GSCA Regulations (Grey County / Owen Sound — 2,859 polygons), NVCA Floodplains (Nottawasaga Valley / Alliston / Wasaga Beach — 138 polygons), and SVCA Approximate Regulated Areas (Bruce–Grey County / Walkerton / Hanover / Port Elgin — 271 polygons). All zone-only; properties within a regulated area rated a maximum of High. Elgin County added to Known Limitations. No scoring threshold or logic changes. |
| v0.17 | June 2026 | Urban Heat Island coverage extended to Hamilton/Burlington (HCA area), Kitchener-Waterloo/Cambridge/Guelph/Brantford (GRCA area), and Niagara Peninsula (St. Catharines, Niagara Falls, Welland, Fort Erie, Grimsby). Data source: USGS Landsat C2L2 Surface Temperature (same calibration as Ottawa). No scoring logic or threshold changes — supplementary signal coverage expansion only. |
| v0.16 | June 2026 | Added Hamilton Conservation Authority (HCA) Regulated Areas as an integrated flood data source — 138 polygons covering Hamilton and Burlington. Zone-only data: all properties within a regulated area are rated a maximum of High. Hamilton removed from Known Limitations. No threshold or scoring logic changes. |
| v0.15 | June 2026 | Added Grand River Conservation Authority (GRCA) Regulatory Floodplain as an integrated flood data source — 4,057 polygons covering Kitchener, Waterloo, Cambridge, Guelph, Brantford, Fergus, and Dunnville (May 2026 data, GRCA Open Data Licence v2). Zone-only data: Floodway and One Zone polygons rated High; Two Zone Fringe polygons rated Moderate. Rating caps at High. Kitchener-Waterloo removed from Known Limitations. No threshold or scoring logic changes. |
| v0.14 | June 2026 | Added Flood Data Confidence indicator (High / Moderate / Regional only / None) to every report, reflecting the resolution of data behind the flood rating. No scoring logic changes. Added documented structural flood data gaps (Niagara, London, CLOCA, Northern Ontario) to Known Limitations. (CLOCA subsequently filled in v0.23; Niagara filled in v0.16.) |
| v0.13 | June 2026 | Added NRCan Flood Susceptibility Index (FSI) supplementary signal for Southern Ontario. Data: NRCan present-day composite 2000–2023 (Open Government Licence), resampled to 300 m. Thresholds calibrated against known addresses: Low < 35, Moderate 35–79, High ≥ 80. Most useful for Coverage Insufficient addresses. NRCan "regional screening only" caveat shown on all reports. Does not affect the overall flood rating or overall risk rating. |
| v0.12 | June 2026 | Urban Heat Island coverage extended to Ottawa / National Capital Region. Data source: USGS Landsat Collection 2 Level-2 Surface Temperature via Microsoft Planetary Computer. Same thresholds as GTA: Low < 30°C, Moderate 30–40°C, High > 40°C. No scoring logic changes — supplementary signal coverage expansion only. |
| v0.11 | May 2026 | Added Urban Heat Island signal. TRCA Average Surface Temperature 2020 raster (Landsat 8, 30 m) used to classify land surface temperature at the geocoded coordinate. Thresholds calibrated against the GTA distribution: Low < 30°C (parks, tree canopy, water-adjacent), Moderate 30–40°C (typical urban surface), High > 40°C (dense pavement, dark roofing). Coverage: TRCA jurisdiction — Toronto, York Region, southern Peel, Durham, and southern Simcoe. Properties outside coverage return Not Applicable. Signal is supplementary and does not affect the overall risk rating. |
| v0.10 | May 2026 | Added Toronto urban sewer flood signal. City of Toronto Basement Flooding Study Areas dataset (67 sanitary subsewershed polygons) used to flag combined-sewer overflow and sewer backflow risk — a flood mechanism not captured by regulatory floodplain data. For Toronto properties, produces a supplementary Elevated / Not Detected signal. Data refreshed daily from the City of Toronto Open Data portal. Does not affect the overall risk rating. |
| v0.9 | May 2026 | Flood scoring: added secondary check against federal Canada Flood Map Inventory (CFM) polygons. When detailed CA data exists within 20 km but does not intersect the subject property, the scoring engine checks whether the property falls inside a CFM polygon smaller than 100 km². If so, the property is rated Moderate with a coarse-data caveat. Polygons ≥ 100 km² are excluded as too broad for property-level assessment. Jurisdiction guard repositioned to apply after the coarse CFM check, correctly handling properties in jurisdictions adjacent to loaded CA sources. Fixes false Low ratings for Peel Region / Mississauga addresses inside CFM flood zones. |
| v0.7 | May 2026 | Added Expected Annual Loss (EAL) for flood-rated properties. Structural damage interpolated from Public Safety Canada / IBI Group residential depth-damage curves. Formula: EAL = (1 / return_period) × damage_pct × property_value. Property value defaults to StatCan 2021 Census median dwelling value by census division; appraiser- or lender-supplied value takes precedence when provided. Representative depth used for zone-only polygons: High → 1.0 m, Moderate → 0.3 m, Extreme → 2.0 m — clearly labelled on the report. Added NRCan National Hydrological Network watercourse proximity as a supplementary signal for Coverage Insufficient addresses. |
| v0.6 | May 2026 | Added jurisdiction guard to prevent cross-boundary false Low ratings. Halton Region (Conservation Halton jurisdiction) now correctly returns Coverage Insufficient rather than a Low rating derived from adjacent Peel Region polygons. Hamilton/Burlington boundary handling also corrected. |
| v0.5 | May 2026 | Building storey count estimated from structure height when not directly available in source data (estimated as height ÷ 3.0 m per floor), with estimates clearly labelled as such on the report. Documented attribute availability rates for the building dataset in Known Limitations. |
| v0.4 | May 2026 | Corrected and clarified scoring threshold documentation. Flood: Low < 0.3 m, Moderate 0.3–1.0 m, High 1.0–2.0 m, Extreme ≥ 2.0 m. Wildfire: independent burn-distance (5 km / 1 km) and FWI (20 / 35) thresholds, higher of the two signals governs. Overall: clarified Partial Assessment definition and that Moderate + Moderate resolves to Moderate overall. |
| v0.3 | May 2026 | Added geocode confidence badge (rooftop / centroid / approximate) to PDF report header. Added public methodology page. Flood data updated (8,353 polygons). |
| v0.2 | Apr 2026 | City mismatch detection added to geocoder. Static satellite map embedded in PDF. Interactive map added to web app result panel. All data sources loaded to production. |
| v0.1 | Apr 2026 | Initial release. Flood zone scoring, wildfire exposure, climate context, and building information. Credit-based billing with API key authentication. |
Data refresh schedule
CWFIS wildfire datasets (NBAC, FWI) are updated annually after each fire season. Flood zone data is refreshed as Conservation Authority portals publish updated polygons — typically quarterly. The ODB buildings dataset is updated when Statistics Canada releases new versions. All dataset versions are recorded in the pipeline audit log and printed on every report.