Evidence PacksFlood SusceptibilityNRCanAg LendingRural AppraisalProduct Launch

By Township Canada

Flood Susceptibility in Evidence Packs: What Ag Lenders and Rural Appraisers See

Farmland, ag lending, rural appraisal, and environmental screening packs now report NRCan flood-prone acres and the 2000 to 2023 trend for each parcel.

A loan file or an appraisal report needs a sentence about flood risk that the writer can defend. "Low-lying in places" is not that sentence, and neither is a screenshot of a coloured map. The evidence packs for farmland portfolios, ag lending, rural appraisal, and environmental screening now carry one: for each parcel, the acres of land that score high on Natural Resources Canada's flood susceptibility maps today, and what the 2000 to 2023 record says about the direction.

Take the published sample on the farmland investment pack page: eight quarter sections in Kneehill County, Alberta, 1,279 acres of land in total. Across the eight, 29 acres score 50 or higher today. None of the parcels has half its land at that level, so no parcel carries a flood watch flag. Three of the eight show a rising trend over 2000 to 2023, and the pack says so in one line. That is a paragraph a lender can lift into a credit memo, with the source and data date attached.

What an evidence pack is

An evidence pack turns the data behind the Township Canada parcel report into a file shaped for one kind of professional work. You send a list of legal land descriptions, quarter sections or LSDs in Alberta, Saskatchewan, or Manitoba. We run the pack and send back JSON, CSV, and GeoJSON for your own tools, plus a printable report for the file. The farmland and ag lending packs take up to 25 parcels in one run. Packs are run as pilots today, so there is no self-serve checkout: each pack page shows a real sample and a link to request one.

Every pack says what it is not. The farmland pack, for example, states that it covers public land characteristics for comparison and is not a price, a rent, or an investment recommendation. Flood figures follow the same rule.

The flood fields in a pack

Each parcel gets these figures, from NRCan's Flood Susceptibility Mapping, summarised for its quarter section:

  • Acres at 50 or higher today, with the share of the land. NRCan scores every 30 m of land from 0 to 100 for how likely it was to flood in a given year, and its present-day estimate follows recent years closely.
  • Today's average score on the quarter. The farmland pack's flood table adds the 90th percentile, which shows how high the wettest tenth of the quarter reaches.
  • The trend and highest years across NRCan's 24 annual maps, written as "Rising (+3.3 a decade); highest 2017, 2021, 2022" or "No clear trend; highest 2021, 2015, 2017". A direction is called only when a standard statistical test finds one and the change is at least 2 points a decade. Year-to-year swings are large, so "No clear trend" is the most common answer.
  • Open water, reported as its own share. Lakes and rivers read high on NRCan's maps, so open water is left out of every acre count and average. Wetlands stay in, because a flood-prone slough edge on a farmed quarter is part of the answer.

The ag lending and farmland packs add two kinds of flags on top of those figures. A watch flag marks a parcel where half or more of the land scores 50 or higher today. An info flag notes a rising trend. On the farmland portfolio pack, both roll up into a count, such as "3 of 8 parcels show a rising flood susceptibility trend over 2000-2023", and the farmland pack adds the total acres at 50 or higher across the whole holding.

NRCan also publishes 2050 to 2100 climate scenarios. Packs list them under "Not included" with the reason: across the Prairies, those maps sit above each location's own 2000 to 2023 range and barely differ between scenarios, which reads as a model baseline shift rather than a climate signal.

Reading the Kneehill County portfolio

The flood table in the farmland sample is short, because most of these quarters score low across the board. Six of the eight have no land at 50 or higher, and one has a single acre. One parcel stands out:

SW-3-32-24-W4: 28 acres (18% of the land) at 50 or higher today, an average of 30, a 90th percentile of 52, and a rising trend of about 3.3 points a decade, with 2017, 2021, and 2022 the highest years.

That parcel does not cross the watch threshold, but it is the one to read twice. Most of its land scores below 50, the wettest part of it reaches past 50, and it has been trending up. For a loan renewal, that justifies a question to the borrower about drainage and wet-year yields on that quarter. For a portfolio buyer, it is the quarter to visit after a wet spring.

The other two rising parcels read differently. SE-36-30-24-W4 rises by about 3.2 points a decade but averages 8 today, and SW-21-29-22-W4 rises by about 2.0 points a decade and averages 10. Neither has any land at 50 or higher. A rising trend on low ground is worth a line in the file, which is why it is an info flag and not a watch flag. Treating every rising trend as a risk would bury the one parcel that matters.

The empty watch column is evidence too. A lender who can write "no parcel has half its land at flood susceptibility 50 or higher (NRCan, present day)" has a cleaner file than one who writes nothing.

Ag lending, appraisal, and environmental screening

The same figures appear in three other packs, shaped for their work:

  • Ag lending: one profile per parcel, with flood susceptibility and flood history beside the energy encumbrances and water. The published sample on the ag lending pack page is SE-36-30-24-W4, and its flags include "Flood susceptibility has been rising since 2000", with the trend and highest years in parentheses.
  • Rural appraisal: a subject quarter and its comparables, side by side. In the published sample on the rural appraisal pack page, the subject NE-18-32-23-W4 and all three comparables have no land at 50 or higher and no clear trend. Flood does not separate the subject from the comps here, and the appraiser can say so.
  • Environmental screening: a checklist row, "Flood-prone land (NRCan 50 or higher, today)", marked found or not found, with the acres, average, and trend in the detail.

The corridor packs for pipeline integrity and utility rights-of-way do not carry flood figures.

The parcel report PDF, for one quarter at a time

For a single quarter section, you do not need a pack. With the Agriculture Bundle, the Ag Report PDF now includes a flood card with the same figures: acres at 50 or higher, today's average, the four bands, a year-by-year chart from 2000 to 2023 with a dashed line at 50, the highest years, and open water left out. Use the interactive Agriculture tab while you work through a list, and the PDF when the quarter goes into a file. The Agriculture add-on guide explains how to build the PDF and how the monthly report allowance counts.

What the figures can and cannot carry

NRCan intends its flood susceptibility maps for regional and national screening, not for predicting flooding at individual properties or for engineering decisions. In a pack, that makes the flood figures what the rest of the pack is: evidence that supports a professional judgement, with the source and data date on the page. For a building site or a drainage decision, the provincial flood hazard map and a site assessment still come first. The NRCan flood susceptibility data guide covers how NRCan built the maps and how we summarise them.

To run a pack on your own parcels, open the pack page for your work and request a pilot with a list of legal land descriptions. A handful is enough for a first pack.