CHC (CHIRPS) — catalog explorer (no network)¶

The chc backend reaches the full Climate Hazards Center product line — CHIRPS / CHIRP precipitation, CHIRTS temperature & humidity, CHIRPS-GEFS forecasts, CHPclim climatology, WBGT, and the SPI / SPEI drought indices — over anonymous FTP. This notebook explores the bundled catalog offline: no FTP access, so it runs deterministically at docs-build time.

Setup¶

Load the bundled Catalog. It reads chc_data_catalog.yaml from package data, so no network call is made. The counts below summarise the curated datasets map, the informational available_datasets index, and the named region sub-grids.

In [ ]:

from earthlens.chc import Catalog

cat = Catalog()
print(f'datasets in catalog: {len(cat.datasets)}')
print(f'available_datasets index: {len(cat.available_datasets)}')
print(f'named regions: {len(cat.available_regions)}')

from earthlens.chc import Catalog cat = Catalog() print(f'datasets in catalog: {len(cat.datasets)}') print(f'available_datasets index: {len(cat.available_datasets)}') print(f'named regions: {len(cat.available_regions)}')

Datasets by temporal resolution¶

Each catalog entry knows its native cadence (daily, monthly, dekadal, pentadal, …) and the pandas frequency the backend builds its date axis from. Counting entries per cadence shows where the product line is concentrated.

In [ ]:

from collections import Counter

by_res = Counter(d.temporal_resolution for d in cat.datasets.values())
for res, n in sorted(by_res.items(), key=lambda kv: -kv[1]):
    print(f'{n:>3}  {res}')

from collections import Counter by_res = Counter(d.temporal_resolution for d in cat.datasets.values()) for res, n in sorted(by_res.items(), key=lambda kv: -kv[1]): print(f'{n:>3} {res}')

Inspect one dataset¶

global-daily is the headline CHIRPS-2.0 product — 0.05° global daily rainfall back to 1981. Pull its entry out of the datasets map and print the grid, date, and format metadata the backend uses to shape a download.

In [ ]:

d = cat.datasets['global-daily']
print('region            :', d.region)
print(
    'temporal_resolution:', d.temporal_resolution, '(pandas freq', d.pandas_freq + ')'
)
print('spatial_resolution :', d.spatial_resolution, 'deg')
print('lat boundaries     :', d.lat_boundaries)
print('lon boundaries     :', d.lon_boundaries)
print('start_date         :', d.start_date)
print('formats            :', d.formats)
print('variables          :', list(d.variables))

d = cat.datasets['global-daily'] print('region :', d.region) print( 'temporal_resolution:', d.temporal_resolution, '(pandas freq', d.pandas_freq + ')' ) print('spatial_resolution :', d.spatial_resolution, 'deg') print('lat boundaries :', d.lat_boundaries) print('lon boundaries :', d.lon_boundaries) print('start_date :', d.start_date) print('formats :', d.formats) print('variables :', list(d.variables))

Look up a variable¶

Every dataset exposes a variables map of Variable metadata — name, units (via type), and a human description. Here we read the precipitation variable off global-daily.

In [ ]:

var = cat.datasets['global-daily'].variables['precipitation']
print('name       :', var.name)
print('type        :', var.types)
print('description :', var.description)

var = cat.datasets['global-daily'].variables['precipitation'] print('name :', var.name) print('type :', var.types) print('description :', var.description)

Named regions¶

CHC ships several regional sub-grids; the catalog expands per-region dataset variants from this block. Each entry maps a region name to its latitude / longitude bounding box.

In [ ]:

for name, box in cat.available_regions.items():
    print(f'{name:<28} {box}')

for name, box in cat.available_regions.items(): print(f'{name:<28} {box}')