tams.identify

Contents

tams.identify#

tams.identify(ctt, *, ctt_threshold=235, ctt_core_threshold=219, size_filter=True, size_threshold=4000, convex_hull=True, parallel=False)#

Identify clouds in 2-D (lat/lon) or 3-D (lat/lon + time) cloud-top temperature data ctt. The returned list of polygon dataframes serves to identify cloud elements (CEs). In a given frame from this list, each row corresponds to a certain CE.

This is the first step in a TAMS workflow.

Parameters:

  • ctt (xarray.DataArray) – Cloud-top temperature array.

  • ctt_threshold (float) – Used to identify the edges of cloud elements (CEs).

  • ctt_core_threshold (float) – Used to identify deep convective cloud regions within larger cloud areas (cold cores). This is used to determine whether or not a system is eligible for being classified as an organized system. It helps target raining clouds.

  • size_filter (bool) –

    Whether to apply size-filtering (using CE and cold-core areas to filter out CEs that are not MCS material). Only CEs with enough cold-core area (size_threshold) are kept.

    Deprecated since version 0.2.0: Set size_threshold=0 instead to disable size filtering.

  • size_threshold (float) – Cold-core area threshold (units: km²). CEs with total cold-core area below this threshold are considered not MCS material and are filtered out. Set to 0 to disable size filtering (e.g. in order to do your own). Note that filtering at this stage makes TAMS more computationally efficient overall.

  • convex_hull (bool) –

    Apply convex hull to the CE polygons to simplify the shapes.

    Note

    • This is done before size filtering / area computation.

    • This fills in any holes the CE polygons may have.

    Added in version 0.2.0: In v0.1.x it was not possible to disable convex hulling.

  • parallel (bool) – Identify in parallel along 'time' dimension for 3-D ctt (requires joblib).

Returns:

ces (list of geopandas.GeoDataFrame) – List of dataframes of CE polygons. Columns:

  • geometry – geometry, the CE polygons

  • area_km2 – float, area of the CE polygons (km²)

  • core – geometry, the cold cores within each CE (MultiPolygon, Polygon, or None if no cores)

  • area_core_km2 – float, the CE’s cold-core area (km²)

Return type:

list[geopandas.GeoDataFrame]

See also

Identify

Demonstrating the impacts of options.

contour()

A lower-level and more general routine for producing shapes by contouring a threshold.

Contents