Skip to content

Legacy API

The legacy climakitae.core interface is the original function-based API for climakitae. It is maintained for backward compatibility only — new work should use climakitae.new_core.user_interface.ClimateData.

Warning

climakitae.core is maintained for backward compatibility only. It still powers older notebooks and internal workflows, but it is no longer the home for new features. New processors, validators, catalogs, and analysis features land in climakitae.new_core.

What belongs here

The legacy API is centered on a small set of modules that work together. Each has a dedicated reference page:

Module Purpose Reference
climakitae.core.data_interface The main legacy entry point. Defines DataParameters and get_data(). Data Interface
climakitae.core.boundaries Legacy boundary loader used for named geographic clipping. Boundaries
climakitae.core.data_load Internal helpers that assemble and load legacy datasets. Data Loading
climakitae.core.data_export Legacy export helpers for NetCDF, CSV, Zarr, and GeoTIFF. Data Export
climakitae.core.constants Shared sentinels and lookup constants used across the legacy stack. Constants
climakitae.core.paths File and catalog path constants used by the legacy loaders. Paths

How to read this section

How the legacy workflow fits together

The primary entry point is get_data(). It accepts keyword arguments, builds its own DataParameters object internally, and returns an xarray object:

  1. Call get_data() with GUI-style keyword arguments.
  2. get_data() constructs and validates a DataParameters object for you.
  3. It executes the query and returns an xarray.DataArray (or None on bad input).
  4. Use load() or export() when you need to materialize or persist the result.
from climakitae.core.data_interface import get_data

data = get_data(
    variable="Air Temperature at 2m",
    resolution="9 km",
    timescale="hourly",
    downscaling_method="Dynamical",
    scenario=["Historical Climate", "SSP 3-7.0"],
    area_subset="CA counties",
    cached_area=["Los Angeles County"],
)

For GUI-style workflows you can still build a DataParameters object directly and call .retrieve(), but get_data() with keyword arguments is the recommended path. See Core Concepts for the field-name conventions and the legacy → modern mapping.

Status and roadmap

Legacy support remains available for backward compatibility, but the project direction is clear:

  • Existing code continues to work.
  • New documentation and tutorials use ClimateData.
  • New feature work lands in climakitae.new_core only.

If you are starting new code, use the modern interface and treat this section as a compatibility reference.