River Ice Classification

SAR satellite images are used to monitor river ice conditions by classifying the ice cover into open water, water and slush on ice surface, non-consolidated ice, and consolidated ice.

File Provider (EO Team)#

Product Description#

Information required from the file provider for product development.

Service Name	Description	Data Owners	File Type
River Ice	River Ice classification of SAR data	Pradeep, Kelley, Mike	.shp (shapefile)

Ancilliary Information#

Product set-up prerequisites:

Naming Convention#

The following naming convention applies to all River Ice Classification files sent to Coresight:

YYYY-MM-DD_RRR_ PPP_SS.jpg
YYYY - Year
MM - Month
DD - Day
RRR - River
PPP – Products (Ice Cover ICOV, Ice Classification CLS, Change Detection CHG)
SS - Sensor

Example:

2020-03-02_CHU_CLS_R2.shp

Symbology Map#

River ice classification is the categorization of backscatter into one of 4 classes during the breakup season vs. 3 classes during the regular season (class 32 removed). Classes are defined as:

Label	Color	Gridcode Value
Open water	#004D95	31
Wet ice or pooling water on ice	#60C0F3	32
Smooth ice	#FDFA00	33
Rough ice	#EB3D00	34

File Uploading Instructions#

All files should be uploaded to Coresight’s Amazon S3 Bucket:
- S3 link here.
- Contact Coresight for account credentials (alias, username and password)
Fill out the Coresight Product Onboarding Checklist for new products.

Example Input Image#

Example shapefile uploaded to Coresight cloud storage.

River Ice Classification Input

Coresight#

Developers#

Jeremy Hetzel
Chris Boyce.

Processing#

Description#

Data processing takes place using an ingester that imports, converts, creates and stores files.

Ingester Script#

The bulk of the ingestion work on each file is done in the cloud and the script can be found in the river-ice-ingester repository.

Ingester Functions#

File conversion - shapefile to geoJSON to MapBox Vector Tiles
File storage - Vector Tiles and geoJSON files on AWS cloud storage
STAC item creation - Automatic naming based on date, product type and AOI matching.
STAC catalog:
- Automatic STAC item sorting based on item name.
- Automatic storage in a Firestore collection based on product type and AOI matching.
Summary statistics - We transform dataframe geometries to a new coordinate reference system and divide them with gridcode(“31,32,33,34”) as an index to get categorized data into 4 sections.
Notification - emails developers upon product publication.

Real time API Access#

The FastAPI OpenAPI endpoint for the river-ice-ingester can be found here.

Service Interface#

River Ice Web interfaces:
- Retrieves product information through the firestore collection.
- Displays the most recent satellite ice classification analysis.
River Ice Classification on the Churchill interface (figure below):
- The upper left graphic displays all of the image analysis dates for a particular ice season.
- The date fields can be used to select seasons, month, and day.
- The upper left graphic also displays ice cover types and area percentages.
- The bottom right area of the map shows the ice classification legend.
- Not all ice types will be displayed with all classification dates.
- Various map elements are included such as river segment numbers, place names, hydrometric and climate data stations, and scale bar.
- The upper right toolbar allows the user to switch between the ice classification and the change detection layers.

River Ice Classification Output