Statistics Canada Fertilizer Shipments 32-10-0039-01

The official answer to a question the input trade asks continuously: how much fertilizer actually reached Canadian farms this campaign. Fertilizer shipments to Canadian agriculture markets is Statistics Canada's annual cube 32-10-0039-01, collected through the Fertilizer Shipments Survey (SDDS 5148) and the successor to the retired CANSIM table 001-0069. Every observation belongs to a July-to-June fertilizer year - the campaign calendar the input trade plans on - and is cumulated across four running windows: July to September, July to December, July to March and the full year.

The cube cuts each campaign four ways by nutrient content - nitrogen, phosphate, potash and sulphur - and down a geography ladder that runs from the Canada total through Eastern and Prairie provinces aggregates to all ten provinces and three territories individually. Values publish in thousands of metric tonnes with the unit and scalar declared on every row. The series reaches from 2006/2007 through 2025/2026, which makes it the current official shipment record for a country that mines potash at global scale and exports nitrogen. It scores 9/10 in our catalog against a 7.81 mean - see how the fertilizers & agricultural chemicals slice ranks, or get a sample of this dataset cut to your nutrients.

Flat, typed and immediately readable - one cumulative observation per row, geography, nutrient and window attached. Values verbatim from the published cube:

# one row per geography x nutrient x cumulative window
REF_DATE: 2006/2007   GEO: Canada           nutrient: Nitrogen  period: July to September  VALUE: 290
REF_DATE: 2006/2007   GEO: Canada           nutrient: Nitrogen  period: July to December   VALUE: 789

REF_DATE: 2025/2026   GEO: British Columbia nutrient: Potash    period: July to December   VALUE: 2
REF_DATE: 2025/2026   GEO: British Columbia nutrient: Sulphur   period: July to September  VALUE: 1

Read together, the rows demonstrate the cube's design. Canada's nitrogen cumulative climbing from 290 to 789 thousand tonnes between the September and December windows is the season building - autumn application clearing before winter. The British Columbia rows show the other axis: a smaller market resolving individually (2 thousand tonnes of potash by December, 1 thousand tonnes of sulphur by September) rather than being absorbed into a regional aggregate.

Every row also carries its plumbing: the declared unit and scalar factor, so 789 always reads as 789,000 tonnes and never as 789; vector and coordinate identifiers that make any cell citable; and status flags that keep preliminary prints distinguishable from frozen estimates. Request a sample and rows come back shaped to your nutrients and provinces.

Sixteen documented columns define every observation, verified against the delivered cube rather than reconstructed from documentation prose. The honest headline is that the analytical payload is one column - VALUE, cumulative shipments in thousands of metric tonnes - and everything else exists to locate, type and trust that number.

Three columns locate the row: the fertilizer-year reference, the geography with its standard identifier code, and the nutrient. A fourth, the cumulative period, positions the observation inside its campaign. Three declare the measurement: unit, scalar factor and the scaled value itself. Two make every cell addressable - the vector identifier and the cube coordinate - so a figure you cite reproduces exactly. And the rest is status discipline: preliminary-versus-frozen flags, suppression symbols, terminated-series markers and declared decimals, which together mean absence and revision stay explicit instead of silently corrupting a sum.

There is no hidden tail beyond the sixteen - nothing sits behind an additional-fields conversation for the core cube. Reshapes and companion tables fold out under additional fields on request.

Geography - the full Canadian ladder on every nutrient: the Canada total, the Eastern and Prairie provinces aggregates the trade reports against, and all ten provinces and three territories individually, each keyed with a standard geographic identifier so joins onto other Canadian geographies hold.

Temporal - fertilizer years 2006/2007 through 2025/2026, roughly twenty campaigns on the July-to-June calendar the industry plans around. Releases land annually - the current cycle was released in March 2026 and subsequently revised through August - and revisions are part of the record: a published correction trail documents earlier estimate revisions, and the newest campaign arrives as partial-year cumulatives that will firm up as windows close.

Granularity - one row per geography x nutrient x cumulative window, in thousands of metric tonnes with scalars declared. Two deliberate absences shape expectations: there is no crop split (shipments feed agriculture markets as a whole, not corn or wheat specifically) and there is no price dimension anywhere in the cube - it is pure tonnage. Rank it yourself across the best fertilizers & agricultural chemicals datasets.

API, files, or your warehouse. Daily, weekly, or hourly.

Pick the channel your team already works in and the cadence your models want; the field dictionary above travels unchanged through all three. Rows arrive flattened out of the cube with geography, nutrient and window already normalized and scalars resolved, so joins against crop, weather or price series need no unit conversion archaeology first.

Every delivery ships the full dictionary, the sample rows and the coverage profile mapped to the nutrients and provinces you named - and a sample comes first regardless of channel.

• Input-demand sizing - cumulative tonnage by nutrient converts directly into product demand for manufacturers and distributors, measured at the farm gate rather than inferred; see our market researchers use cases page.
• Producer and rail-side monitoring - provincial potash and nitrogen flows give supply-chain watchers a domestic-movement series on an official calendar.
• Seasonality and pacing models - the four cumulative windows turn each campaign into a build-up curve worth forecasting against.
• Cross-border framing - Canadian shipments beside US consumption and application records give a continental view on matched mid-year-ending calendars; see our data scientists use cases page.
• Policy and environmental baselines - nutrient arriving on farm by province is the upstream input for loading and emissions conversations.
• Citation-grade sourcing - every cell carries vector and coordinate identifiers, so any figure in a deck or paper reproduces on demand; see our investors quants use cases page.

Market researchers and consultants get citable tonnage behind Canadian fertilizer market sizing, split by nutrient and province instead of one blended national ratio. Data scientists and ML engineers get a tidy long-format panel that pivots on geography x nutrient x window keys with scalars already declared - features assemble without a cleaning pass. Investors and quant researchers read shipment pace as demand-side signal for potash and nitrogen producers, ahead of the revenue proxies everyone else uses. Journalists, academics and students anchor input-cost and soil stories to the national statistical office's own numbers. Competitive intelligence teams benchmark private estimates against the official record before trusting them. Sales and growth teams rank territories by where nutrients actually move.

Because fit is proven with rows, not adjectives. A sample comes back cut to the slice you name - Saskatchewan potash cumulatives for the last five campaigns, nitrogen by province for one season's build-up, the sulphur picture nationally - with the dictionary, scalars and status-flag behaviour visible so verification happens before anything is wired in. If your question is American application behaviour rather than Canadian shipments, see the head-to-head with USDA ERS Fertilizer Use and Price; if the vocabulary is new, start with plant nutrient, explained. Otherwise, get a sample of this dataset scoped to your nutrients.