TdM > MSC GeoMet > OGC API
OGC API
The Open Geospatial Consortium (OGC) API standards are a set of specifications developed by the OGC to facilitate the interoperable exchange of geospatial data and services over the web. These standards are designed to enable seamless integration and sharing of geospatial information across different systems, applications, and organizations. The following documentation will go over the basics of the OGC API - Features and OGC API - Coverages standards and provide examples of how to query GeoMet-OGC-API using these standards.
Main endpoints
GET / - Landing page
The GeoMet-OGC-API landing page with links to view the data collections, processes, and conformance classes implemented by the API.
GET /collections - All data collections
https://api.weather.gc.ca/collections
Retrieves information describing the set of supported collections and provides a list of collections with data type and a brief description.
GET collections/{collectionID} - Collection-specific information
https://api.weather.gc.ca/collections/hydrometric-daily-mean
Retrieves descriptive information about a specific collection, either of type feature or type coverage. For example, the URL above returns the root page for the Hydrometric Monitoring Stations feature collection.
GET /openapi - OpenAPI documentation
https://api.weather.gc.ca/openapi
The API documentation links provide a Swagger page of the API as a tool for developers to provide example request/response/query capabilities. A JSON representation is also provided.
GET /conformance - API conformance definition
https://api.weather.gc.ca/conformance
The conformance page provides a list of URLs corresponding to the OGC API conformance classes supported by GeoMet-OGC-API.
Common Query Parameters
The following section provides information regarding common query parameters often used when making requests to both feature and coverage collections. Query parameters are passed with the request URL using the ? character to separate the query parameter from the URL and the & character to separate multiple query parameters.
f - Format query parameter
https://api.weather.gc.ca/collections/?f=json
The format query paremeter is used to specify the format of the server response. Depending on the collection type, the value of the format query parameter could be one of of the following: html, json. The example above returns the list of available collections from GeoMet-OGC-API as a JSON object.
For feature collections, results can also be returned as a CSV file via ?f=csv.
For coverage collections, the output format of the coverage can be specified via this parameter. Depending on the collection and its supported output formats, the following values are supported: GRIB, GTiff, and NetCDF in GeoMet-OGC-API but their support is on a per coverage collection basis.
bbox - Bounding box query parameter
https://api.weather.gc.ca/collections/aqhi-observations-realtime/items?bbox=-71,46,-65,49
Retrieves information according to a specific region or bounding box. It is a rectangular area defined by two sets of coordinates that represent the minimum and maximum extents of the box. bbox coordinate values must be fall within the bounds of EPSG:4326. The coordinates specified must be in the following order: minX, minY, maxX, maxY.
datetime - Datetime query parameter
https://api.weather.gc.ca/collections/hydrometric-daily-mean/items?datetime=1972-10-30
Retrieves information according to a specific date, year or time interval. It is highly recommend to use RFC 3339 compliant datetime strings (i.e. YYYY-MM-DD or YYYY-MM-DDTHH:MM:SSZ). The datetime parameter can be used in conjunction with the .. keyword to specify a time extent. For example, datetime=2012-05-20/.. returns all items from 2015-05-20 onwards. datetime=../2015-11-11 returns all items older than 2015-11-11.
OGC API - Features
The OGC API - Features (OAFeat) standard provides a list of standardized specifications for querying vector-based geospatial data (points, lines and polygons) on the web. Listed below are example OGC-API feature collection endpoints, examples for spatial, temporal and property filtering, and further details regarding pagination.
GET collections/{collectionID}/items - Feature collection items
https://api.weather.gc.ca/collections/hydrometric-daily-mean/items
Returns items from a specified collection.
The above URL returns items from the hydrometric-daily-mean collection:
GET collections/{collectionID}/items/{itemID} - Single feature overview
https://api.weather.gc.ca/collections/hydrometric-daily-mean/items/01AA002
Returns the item with the corresponding ID from the requested collection and provides its full set of properties.
The above URL returns the item with ID 01AA002 from the hydrometric-stations collection:
GET collections/{collectionID}/items?{query parameters} - Query a feature collection
Feature collection queries allow for spatial, temporal, and property filtering. Filter parameters can be combined to formulate a more complex query. The first parameter of a query must be specified with ? and subsequent parameters with &.
Below are a few examples of various types of queries, ranging from simple bbox and datetime filtering queries to more complex queries comprised of multiple query parameters. All queries are made to the hydometric-daily-mean feature collection.
Spatial
Query by bounding box (minx, miny,maxx, maxy). https://api.weather.gc.ca/collections/hydrometric-stations/items?bbox=-71,46,-65,49
Returns items from the feature collection that intersect the provided bounding box:
Temporal
Query for an instantaneous value with a single date: https://api.weather.gc.ca/collections/hydrometric-daily-mean/items?datetime=1972-10-30
Returns feature collection items that match the provided datetime value:
Query for a time interval: https://api.weather.gc.ca/collections/hydrometric-daily-mean/items?datetime=1972-10-30/2010-07-31
Returns feature collection items that fall within the provided datetime range:
Query by time extent with the open ended .. keyword:
-
Since a specific datetime: https://api.weather.gc.ca/collections/hydrometric-daily-mean/items?f=json&STATION_NUMBER=07DA006&datetime=2012-05-20/..
-
Returns feature collection items from 2015-05-20 onwards.
Until a specific datetime: https://api.weather.gc.ca/collections/hydrometric-daily-mean/items?f=json&STATION_NUMBER=07DA006&datetime=../2015-11-11
Returns feature collection items older than 2015-11-11.
Property
Query by an entity collection property:
https://api.weather.gc.ca/collections/hydrometric-daily-mean/items?STATION_NUMBER=10CD001
Returns feature collection items that have STATION_NUMBER = 10CD001
Paging
-
Offset
The
offsetparameter can be used to specify the record to start at when extracting features. The default value is0(first feature). -
Total number of items
The
resulttype=hitsparameter is used to return the number of features found in a collection, without returning any features from the collection. The number of features is found in the response'snumberMatchedkey.https://api.weather.gc.ca/collections/ltce-stations/items?f=json&resulttype=hits
-
Limit
The
limitparameter is used to define the maximum records to return in the response. The response size is limited to 10 000 features per query. In the following query, a single entity is requested, for example, to inspect its geometry and properties:https://api.weather.gc.ca/collections/hydrometric-daily-mean/items?limit=1
Returns a single feature from the collection:
The
offsetandlimitparameters can be used in tandem to cycle through feature collections. The examples below demonstrate how to adjust and page through query results.-
Query and limit to items 1 to 100:
https://api.weather.gc.ca/collections/hydrometric-daily-mean/items?STATION_NUMBER=10CD001&limit=100
-
Query and limit to items 101-200:
-
-
Strategies for paging
The paging strategy is commonly used when extract large amounts of data. For example, a client could page by 500 features to cycle through an entire station record, starting with the first 500 items and subsquently requesting the next 500 items, until all items are returned. https://api.weather.gc.ca/collections/hydrometric-daily-mean/items?STATION_NUMBER=01AP004&offset=0&limit=500
The above queries return the first 500 items and the next 500 items respectively. The client can then simply cycle through all items until there are no longer any records. This would constitute the entire record. The
/itemsendpoint will also return anextlink in the response, which can be used to retrieve the next page of results. Thenextlink will contain the updatedoffsetvalue for the next page of results.
Filter the list of properties of a queried element
Query a feature collection and return features for which the list of associated properties is restricted to the properties listed in the properties=. For example, the query below returns properties related to atmospheric pressure observations, in JSON format.
Combining filter parameters
Query all hydrometric daily mean data from a single station between 2001 and 2010:
Exporting to CSV format
Results from queries made against feature collections can be exported to the CSV format by adding &f=csv to the request, or by clicking on the CSV top-right icon on the collection page.
https://api.weather.gc.ca/collections/hydrometric-daily-mean/items?f=csv
Sorting
Results from a query can be sorted by using the sortby parameter. By default, the sort order is ascending. For a descending sort, add the - (minus) character prior to the property name. Sorting by multiple properties is supported by providing a comma-separated list of properties to the sortby query parameter.
&sortby=PROPERTY sorts the property in ascending order
&sortby=-PROPERTY sorts the property in descending order.
Examples with the DATE property:
OGC API - Coverages
The OGC API - Coverages standard specifies the fundamental API building blocks for interacting with coverages (raster data). The spatial data community uses the term "coverage" to refer to homogeneous sets of values located in space and time, such as data from spatiotemporal sensors, images, simulation and statistics.
GET collections/{coverageId} - Coverage collection landing page
https://api.weather.gc.ca/collections/climate:cmip5:historical:annual:anomaly
Retrieves detailed coverage metadata for the specified collection. The JSON response for this page displays the collection's geographic extent, grid size, temporal extent (if available), and any other additional dimensions. The collection root page also lists supported other collection-related links such as the collection schema (available fields) and coverage URLs for each format support by the collection.
GET collections/{coverageId}/schema - Coverage schema
https://api.weather.gc.ca/collections/climate:cmip5:historical:annual:anomaly/schema
Retrieves the schema of the coverage collection. The schema provides a list of available fields and their data types. The schema can be returned in JSON or HTML format and can be used to understand the available fields of the coverage data.
GET collections/{coverageId}/coverage - Coverage query with no parameters
https://api.weather.gc.ca/collections/climate:cmip5:historical:annual:anomaly/coverage
Returns the coverage data for the specified collection. The response will include all available data for the collection, which may be a large amount of data. It is recommended to use query parameters to filter the data returned by the coverage query. The data can be returned in different formats, such as CoverageJSON, GRIB, or NetCDF, depending on the collection and its supported output formats. See the links contained in the collection's root page for its available formats.
GET collections/{coverageId}/coverage?{params} - Coverage with query parameters
Queries against coverage collections allow for spatial, temporal, and field filtering against the n-dimensionsal space of the coverage itself//. The first parameter of a query must be specified with ? and subsequent parameters with &.
Spatial
Query by bounding box (minx, miny,maxx, maxy). https://api.weather.gc.ca/collections/climate:cmip5:historical:annual:anomaly/coverage?f=json&bbox=-80.557251,42.561173,-75.135498,44.805224
Temporal
Query for an instantaneous value for a single period if a "time" axis label is available, for this specific coverage: https://api.weather.gc.ca/collections/climate:cmip5:historical:annual:anomaly/coverage?f=json&datetime=1905
Query for a time interval: https://api.weather.gc.ca/collections/climate:cmip5:historical:annual:anomaly/coverage?f=json&datetime=1901/1905
Schema field
Query for a specifc coverage field listed in the coverage schema via the properties query parameter. For example, the query below retrieves coverage data for the schema field named "sit" (sea ice thickness).
https://api.weather.gc.ca/collections/climate:cmip5:historical:annual:anomaly/coverage?properties=sit
Additional dimensions
Query additional dimensions (other than the spatial and temporal, which are queried via the bbox and datetime query parameters respectively) with the subset query parameter. Additional available dimensions are listed under the extent key at the coverage collection root via the JSON response only.
https://api.weather.gc.ca/collections/climate:cmip5:historical:annual:anomaly/coverage?subset=percentile(25)
Combination of filter parameters
Query the coverage with a spatial filter, temporal filter, and for a specific coverage field via the properties query parameter.
Returns in json format by default:
Export to GTiff or NetCDF format
Queries pointing to collections pages can be exported in GTiff format and NetCDF format if available. Consult the root of a given coverage collection to identify the available response data formats. For example:
https://api.weather.gc.ca/collections/climate:cmip5:projected:annual:absolute?f=json
Under the links key, the available values for the ?f parameter can be found. In this case the coverage data can be requested in JSON via ?f=json or as NetCDF via ?f=NetCDF.
OGC API - Advanced filtering with CQL2
GeoMet-OGC-API supports data filtering using Common Query Language (CQL2) though the ?filter query parameter in the HTTP request. CQL2 is an OGC standard designed to support advanced query of geospatial data and perform filtering directly on the server. By using CQL2, clients can reduce the amount of data returned and retrieve only data of interest.
CQL2 allows users to construct flexible and precise queries based on feature attributes and geometry.
- Exact matches:
=match features where an attribute has exact value; - Partial string or keywords:
LIKEwith%wildcard; - Comparison operators:
>,<,>=,<=filter numeric or temporal values; - Logical operators:
AND,OR,NOTcompare conditions; - Spatial predicates:
INTERCEPTS,DISJOINTfilter based on geometry.
The full list of predicates available is accessible from the OGC CQL2 standard.
String values must be enclosed in single quotes '' and the property path (properties.attribute) must exactly match the attribute name defined in the collection schema (is case-sensitive). For GeoMet-OGC-API collections the synthax is:
https://api.weather.gc.ca/collections/{collectionID}/items?filter=<CQL2 expression>&<CQL2 expression>
1. Searching exact matches
1.1) Filter all meteorological bulletin that have the type CA, and are from issuing office AO.
CQL2: ?filter=properties.type=CA AND properties.issuing_office=AO
1.2) Filter Climate - Daily Observations by attribute LOCAL_DAY. Here, LOCAL_DAY must be in capital letters, as defined in schema.
CQL2: ?filter=properties.LOCAL_DAY=15
https://api.weather.gc.ca/collections/climate-daily/items?filter=properties.LOCAL_DAY=15
2. Searching partial keywords
CQL supports partial text matching using the LIKE operator. It allows you to search for substrings within text attributes by using the % character as a wildcard, which can represent zero or more characters. The comparison is case-insensitive, but search string must be entered as lowercase and in between single quote. If the attribute value contains multiple words, the LIKE predicate is applied to the entire text value, matching the pattern against any substring within it.
2.1) Find all STATION_NAME containing the keyword LAKE. In that case, '%lak%' matches case insensitive keywords such as Lake, lak5, SPOKIN LAKE 4E, 2LAKE9, LaK.
CQL2: ?filter=properties.STATION_NAME LIKE '%lak%'
2.2) Filter only specific name of station whose name is (or is NOT) in a given list. The wildcard % can also be used in the list.
CQL2: ?filter=properties.STATION_NAME NOT IN ('lake', 'river')
2.3) To filter for more than a word, you need two different LIKE predicates. In this case, specifying 'quinsam' AND 'river' allows to filter for station name containing both 'quinsam' and 'river' keywords. Using OR would return station name containing either 'quinsam', 'river' or both.
CQL2: ?filter=properties.STATION_NAME LIKE '%quisam%' AND properties.STATION_NAME LIKE '%river%'
2.4) Extracts station name starting with the letter K. You can place the wildcard % accordingly. Here, you see that it is not necessarily the first word of the station name that is a match. The LIKE predicate will compare every keywords of the text.
CQL2: ?filter=properties.STATION_NAME LIKE 'k%'
https://api.weather.gc.ca/collections/climate-normals/items?filter=properties.STATION_NAME LIKE 'k%'
2.5) Extracts station name ending with the letter T. You can place the wildcard % accordingly. Here, you see that it is not necessarily the last word of the station name that is a match. The LIKE predicate will compare every keywords of the text.
CQL2: ?filter=properties.STATION_NAME LIKE '%t'
https://api.weather.gc.ca/collections/climate-normals/items?filter=properties.STATION_NAME LIKE '%t'
2.6) Searching meteorological bulletins by the identifier. It's not possible to manually search by id on the interface. Users can use CQL filters and leverage the bulletin filename nomenclature documentation.
CQL2: ?filter=properties.identifier LIKE '%WWCN70%'
2.7) Search bulletins with partial keyword CYOY in the url.
CQL2: ?filter=properties.type=SA AND properties.issuer_code=CWAO AND properties.url LIKE '%cyoy%'
?q= is the "full text search" capability in the MSC GeoMet OGC API. The equivalent query is:
https://api.weather.gc.ca/collections/bulletins-realtime/items?type=SA&issuer_code=CWAO&q=*CYOY*
2.8) Search bulletins by type and issuer code.
CQL2: ?filter=((properties.type=SA OR properties.type=SP) AND (properties.issuer_code=CWAO OR properties.issuer_code=CYOY))
3. Searching comparison operators
CQL2 supports numeric comparisons using operators.
3.1) Filter for temperature TEMP values less than 10°C.
CQL2: ?filter=properties.TEMP < 10
https://api.weather.gc.ca/collections/climate-hourly/items?filter=properties.TEMP < 10
3.2) Filter for Air Quality Health Index (AQHI) forecasts classified as low risk (1-3).
CQL2: ?filter=properties.aqhi<=3
https://api.weather.gc.ca/collections/aqhi-forecasts-realtime/items?filter=properties.aqhi<=3
3.3) Filter for AQHI observations lower than 2 that is in a specified bounding box (Nova Scotia).
CQL2: ?bbox=-70,42,-60,45&filter=properties.aqhi<2
4. Searching with logical operator
CQL supports range-based filtering using the BETWEEN operator. This operator is inclusive for both boundary values. Using BETWEEN is the equivalent of >= and <=, but improves readability.
synthax: [`NOT`] <attribute> BETWEEN <lower_bound> AND <upper_bound>
4.1) Filter items for temperature TEMP between 10°C and 12.5°C. For numeric values (number), no quotes required.
CQL2: ?filter=properties.TEMP BETWEEN 10 AND 12.5
4.2) Filter for wind direction between north-northeast (NNE 23°) and northeast (NE 45°) and speed above 25 km/h.
CQL2: ?filter=properties.WIND_DIRECTION BETWEEN 23 AND 45 AND properties.WIND_SPEED >= 25
4.3) Filter items between two datetime. Datetime values must be enclosed in single quotes, because it is defined as string in the collection queryables.
CQL2: ?filter=properties.LOCAL_DATE BETWEEN '2023-11-28 14:00:00' AND '2024-11-28 14:00:00'
4.4) Search for Air Quality Health Index observations with moderate risk only (BETWEEN 4 AND 6) in a BBOX.
CQL2: ?filter=BBOX(geometry, -140, 50, -85, 63) AND properties.aqhi BETWEEN 4 AND 6
CQL2: ?filter=BBOX(geometry, -140, 50, -85, 63) AND properties.aqhi>=7 (high risk)
5. Spatial searches
Spatial searches allow users to personalize their queries based on geographical locations.
5.1) Find features that INTERSECTS a polygon. Returns all features that touch, crosses, fall in or overlaps a polygon. A polygon of x sides has to have x+1 points because the last point repeats the first one to close the shape of the polygon. Specify the longitude fist, then the latitude.
CQL2: ?filter=INTERSECTS(geometry, POLYGON((lon_1 lat_1,lon_2 lat_2,lon_3 lat_3,lon_4 lat_4,lon_1 lat_1)))
CQL2: ?filter=INTERSECTS(geometry, POLYGON((-56 47,-55 47,-55 48,-56 48,-56 47)))
https://api.weather.gc.ca/collections/swob-stations/items?filter=INTERSECTS(geometry, POLYGON((-56 47,-55 47,-55 48,-56 48,-56 47))) → Returns all SWOB stations intersecting the polygon that has a width of 1 degree (west-east longitudes) and a height of 1 degree (north-south latitude).
5.2) Find features with DISJOINT. Returns all features outside of a polygon, that do not intersect the geometry. It is the opposite of INTERSECTS.
CQL2: ?filter=DISJOINT(geometry, POLYGON((lon_1 lat_1,lon_2 lat_2,lon_3 lat_3,lon_4 lat_4,lon_1 lat_1)))
CQL2: ?filter=DISJOINT(geometry, POLYGON((-80 42,-50 42,-50 60,-80 60,-80 42)))
https://api.weather.gc.ca/collections/swob-stations/items?filter=DISJOINT(geometry, POLYGON((-80 42,-50 42,-50 60,-80 60,-80 42)))
→ Returns all SWOB stations that do not intersect the polygon that has a width of 30 degrees (longitudes) and 18 degrees (latitudes) in Eastern Canada.
6. Examples of multi-parameter filtering
6.1) Filtering Climate - Daily Observations for a weather station in Lanaudiere for wet and warms summer months for a station where:
- The
CLIMATE_IDENTIFIERis 7014160 (L'Assomption station); - The
MEAN_TEMPERATUREis 18°C and 28°C; - The
TOTAL_PRECIPITATIONis between 5 and 40 mm; - The
LOCAL_YEARis between 2010 and 2020; - The
LOCAL_MONTHis between June and September (JJAS).
CQL2: ?filter=CLIMATE_IDENTIFIER = '7014160' AND properties.MEAN_TEMPERATURE BETWEEN 18 AND 28 AND properties.TOTAL_PRECIPITATION BETWEEN 5 AND 40 AND properties.LOCAL_YEAR BETWEEN 2010 AND 2020 AND properties.LOCAL_MONTH BETWEEN 6 AND 9
6.2) Find all AQHI observations in a polygon around Calgary for today where the AQHI value is greater than 6 (high risk) within a polygon.
CQL2: ?filter=INTERSECTS(geometry,POLYGON((-114.2 50.9, -113.8 50.9, -113.8 51.2, -114.2 51.2, -114.2 50.9))) AND properties.aqhi>6
6.3) Retrieve meteorogical stations meeting all of the following conditions:
- Located within a specified polygon (Western canada region) with
INTERSECTS; - Have a station containing the keyword "LAKE" with
LIKE; - Reports minimum temperature
BETWEEN-20 and -10°C; - Either have maximum speed gusts above 20 km/h.
CQL2: ?filter=INTERSECTS(geometry,POLYGON((-140 48,-95 48,-95 60,-140 60,-140 48))) AND properties.STATION_NAME LIKE '%lake%' AND properties.MIN_TEMPERATURE BETWEEN -20 AND -10 AND properties.SPEED_MAX_GUST >= 20
6.4a) Identify regions of elevated risk of wildfire (dry, hot and windy conditions) for the 2025 wildfire season.
- During the wildfire 2025 season (April to October);
- High temperatures (
>=30°C); - Low relative humidity (
<=30%); - Strong winds (speed
>=30 km/h).
Climate - Hourly Observations
CQL2: ?filter=properties.LOCAL_YEAR=2025 AND properties.LOCAL_HOUR=12 AND properties.LOCAL_MONTH BETWEEN 4 AND 10 AND properties.TEMP >=30 AND properties.RELATIVE_HUMIDITY <=30 AND properties.WIND_SPEED >= 30
6.4b) Using the same conditions as for the hourly observations, now we add the 24h total precipitation amount from Climate - Daily observations to account for the dryness. Total rainfall over 24 hours less than an arbitrary value of 3 mm.
Climate - Daily Observations
CQL2: ?filter=properties.LOCAL_YEAR=2025 AND properties.LOCAL_MONTH BETWEEN 4 AND 10 AND properties.MAX_TEMPERATURE>=30 AND properties.MIN_REL_HUMIDITY<=30 AND properties.SPEED_MAX_GUST>30 AND properties.TOTAL_PRECIPITATION<=3
Curl command and Python
It was shown that it's possible to make HTTP request with CQL filters directly in the search bar of the search engine. It's also possible to retrieve the data via the curl command in the terminal. To save the output of what the request returns, use the -o option.
- Using curl with the full encoded url.
curl [encoded_url] [option]
curl https://api.weather.gc.ca/collections/climate-hourly/items?filter=properties.WIND_DIRECTION%20BETWEEN%2023%20AND%2045%20AND%20properties.WIND_SPEED%20%3E=%2025%20&lang=en -o /path/to/output_file.json
- Encoding the url directly with the
--data-urlencodeand-GGET request options. Users can add multiple--data-urlencodeparameters (equivalent to&).
curl -G [basepath] --data-urlencode [CQL EXPRESSION]
curl -G "https://api.weather.gc.ca/collections/climate-hourly/items?" --data-urlencode "filter=properties.WIND_DIRECTION BETWEEN 23 AND 45 AND properties.WIND_SPEED >=25" --data-urlencode "lang=en"
Query with Python
To parse CQL filters and make HTTP requests, install the pygeofilter and requests Python modules.
pip3 install pygeofilter requests
import requests
import json
from pygeofilter.parsers.ecql import parse
# GeoMet-OGC-API endpoint.
api_url = "https://api.weather.gc.ca/collections/climate-daily/items"
# CQL filters.
cql_filter = (
"properties.LOCAL_YEAR = 2025 AND "
"properties.LOCAL_MONTH BETWEEN 6 AND 9 AND "
"properties.MAX_TEMPERATURE >= 30"
)
# Validate CQL filters before API request.
try:
parse(cql_filter)
except Exception as e:
raise ValueError(f"Invalid CQL filter: {e}")
params = {
"filter": cql_filter,
"lang": "en",
"limit": 100
}
# Sends GET request to the API and encode url.
response = requests.get(api_url, params=params)
print(f"The encoded url is: {response.url}")
# Format json in output file.
data = response.json()
climate_output_file = "climate_daysAbove30deg_JJAS_2025.json"
with open(climate_output_file, "w") as fh:
json.dump(data, fh)