How to use the Compatibility API¶

The /compatibility endpoint helps you understand how to call TiTiler-CMR for a collection. It returns the sampled granule_ur and accepts granule_ur when you want to inspect a specific granule. For hierarchical HDF5 and NetCDF datasets it can return a lightweight list of candidate group paths from the root request, then let you inspect variables by making a follow-up request with group=....

This notebook demonstrates that workflow for the NISAR Beta Geocoded Polarimetric Covariance (GCOV) collection, which is a known example of a hierarchical HDF5 dataset that needs a group parameter for xarray requests.

Setup¶

In [1]:

Copied!





import json
import os

import earthaccess
import httpx2 as httpx

titiler_endpoint = os.getenv(
    "TITILER_CMR_ENDPOINT", "https://staging.openveda.cloud/api/titiler-cmr"
)
import json
import os

import earthaccess
import httpx2 as httpx

titiler_endpoint = os.getenv(
    "TITILER_CMR_ENDPOINT", "https://staging.openveda.cloud/api/titiler-cmr"
)

Identify the dataset¶

You can find the collection with earthaccess.search_datasets.

In [2]:

Copied!





datasets = earthaccess.search_datasets(concept_id="C3622214170-ASF")
ds = datasets[0]

collection_concept_id = ds["meta"]["concept-id"]
print("CollectionConcept-Id:", collection_concept_id)
print("Short name:", ds["umm"]["ShortName"])
print("Version:", ds["umm"]["Version"])
datasets = earthaccess.search_datasets(concept_id="C3622214170-ASF")
ds = datasets[0]

collection_concept_id = ds["meta"]["concept-id"]
print("CollectionConcept-Id:", collection_concept_id)
print("Short name:", ds["umm"]["ShortName"])
print("Version:", ds["umm"]["Version"])

CollectionConcept-Id: C3622214170-ASF
Short name: NISAR_L2_GCOV_BETA_V1
Version: 1

Explore the collection using `/compatibility`¶

In [3]:

Copied!





compatibility_response = httpx.get(
    f"{titiler_endpoint}/compatibility",
    params={"collection_concept_id": collection_concept_id},
    timeout=None,
).json()

print(json.dumps(compatibility_response, indent=2))
compatibility_response = httpx.get(
    f"{titiler_endpoint}/compatibility",
    params={"collection_concept_id": collection_concept_id},
    timeout=None,
).json()

print(json.dumps(compatibility_response, indent=2))

{
  "concept_id": "C3622214170-ASF",
  "backend": "xarray",
  "datetime": [
    {
      "RangeDateTimes": [
        {
          "BeginningDateTime": "2025-10-01T00:00:00.000Z"
        }
      ],
      "EndsAtPresentFlag": true
    }
  ],
  "variables": {},
  "dimensions": {},
  "coordinates": {},
  "compatible_groups": [
    "science/LSAR/GCOV/grids/frequencyA",
    "science/LSAR/GCOV/grids/frequencyB"
  ],
  "example_assets": "s3://sds-n-cumulus-prod-nisar-products/NISAR_L2_GCOV_BETA_V1/NISAR_L2_PR_GCOV_002_109_D_063_4005_DHDH_A_20251012T182508_20251012T182531_X05010_N_P_J_001/NISAR_L2_PR_GCOV_002_109_D_063_4005_DHDH_A_20251012T182508_20251012T182531_X05010_N_P_J_001.h5",
  "sample_asset_raster_info": null,
  "links": []
}

Interpret the response¶

For hierarchical datasets, a root /compatibility request stays lightweight. The most important fields are:

granule_ur: the sampled granule used for this compatibility check
compatible_groups: candidate group paths to try in a follow-up request

At this stage the response may not include variable metadata or xarray links yet, because TiTiler-CMR has not inspected a nested group.

In [4]:

Copied!





compatible_groups = compatibility_response.get("compatible_groups", [])
variables = list((compatibility_response.get("variables") or {}).keys())
tilejson_link = next(
    (
        link["href"]
        for link in compatibility_response.get("links", [])
        if link["rel"] == "tilejson"
    ),
    None,
)

print("compatible_groups:")
for group in compatible_groups:
    print(" -", group)
compatible_groups = compatibility_response.get("compatible_groups", [])
variables = list((compatibility_response.get("variables") or {}).keys())
tilejson_link = next(
    (
        link["href"]
        for link in compatibility_response.get("links", [])
        if link["rel"] == "tilejson"
    ),
    None,
)

print("compatible_groups:")
for group in compatible_groups:
    print(" -", group)

compatible_groups:
 - science/LSAR/GCOV/grids/frequencyA
 - science/LSAR/GCOV/grids/frequencyB

Inspect a specific group¶

Once you choose a group from compatible_groups, make a follow-up /compatibility request with group=... to inspect variables and get xarray links.

In [5]:

Copied!





selected_group = compatible_groups[1]

grouped_compatibility_response = httpx.get(
    f"{titiler_endpoint}/compatibility",
    params={
        "collection_concept_id": collection_concept_id,
        "group": selected_group,
    },
    timeout=30,
).json()

print(json.dumps(grouped_compatibility_response, indent=2))
selected_group = compatible_groups[1]

grouped_compatibility_response = httpx.get(
    f"{titiler_endpoint}/compatibility",
    params={
        "collection_concept_id": collection_concept_id,
        "group": selected_group,
    },
    timeout=30,
).json()

print(json.dumps(grouped_compatibility_response, indent=2))

{
  "concept_id": "C3622214170-ASF",
  "backend": "xarray",
  "datetime": [
    {
      "RangeDateTimes": [
        {
          "BeginningDateTime": "2025-10-01T00:00:00.000Z"
        }
      ],
      "EndsAtPresentFlag": true
    }
  ],
  "variables": {
    "HHHH": {
      "shape": [
        4590,
        4635
      ],
      "dtype": "float32",
      "min": 1.9753079380513405e-15,
      "max": 126.9072265625,
      "mean": 0.2488117665052414,
      "p01": 5.214990765611826e-15,
      "p05": 1.1158001606004219e-14,
      "p95": 0.572845458984375,
      "p99": 1.0888000488281195
    },
    "HVHV": {
      "shape": [
        4590,
        4635
      ],
      "dtype": "float32",
      "min": 3.3623819874206706e-16,
      "max": 1.6359405517578125,
      "mean": 0.05236886069178581,
      "p01": 9.670644276690842e-16,
      "p05": 2.27081808218399e-15,
      "p95": 0.13129768371582023,
      "p99": 0.24061668395996075
    },
    "listOfCovarianceTerms": {
      "shape": [
        2
      ],
      "dtype": "|S4",
      "min": null,
      "max": null,
      "mean": null,
      "p01": null,
      "p05": null,
      "p95": null,
      "p99": null
    },
    "listOfPolarizations": {
      "shape": [
        2
      ],
      "dtype": "|S2",
      "min": null,
      "max": null,
      "mean": null,
      "p01": null,
      "p05": null,
      "p95": null,
      "p99": null
    },
    "mask": {
      "shape": [
        4590,
        4635
      ],
      "dtype": "float32",
      "min": 0.0,
      "max": 3.0,
      "mean": 1.6180167198181152,
      "p01": 0.0,
      "p05": 0.0,
      "p95": 3.0,
      "p99": 3.0
    },
    "numberOfLooks": {
      "shape": [
        4590,
        4635
      ],
      "dtype": "float32",
      "min": 0.036182284355163574,
      "max": 68.21331024169922,
      "mean": 36.98814010620117,
      "p01": 20.3447624206543,
      "p05": 27.77091426849365,
      "p95": 45.24116630554199,
      "p99": 50.53271461486816
    },
    "numberOfSubSwaths": {
      "shape": [],
      "dtype": "uint8",
      "min": 3.0,
      "max": 3.0,
      "mean": 3.0,
      "p01": 3.0,
      "p05": 3.0,
      "p95": 3.0,
      "p99": 3.0
    },
    "rtcGammaToSigmaFactor": {
      "shape": [
        4590,
        4635
      ],
      "dtype": "float32",
      "min": 0.2671063244342804,
      "max": 0.9898551106452942,
      "mean": 0.7490795254707336,
      "p01": 0.5234366035461426,
      "p05": 0.6213086247444153,
      "p95": 0.8584897398948669,
      "p99": 0.9072614645957946
    },
    "xCoordinateSpacing": {
      "shape": [],
      "dtype": "float64",
      "min": 80.0,
      "max": 80.0,
      "mean": 80.0,
      "p01": 80.0,
      "p05": 80.0,
      "p95": 80.0,
      "p99": 80.0
    },
    "yCoordinateSpacing": {
      "shape": [],
      "dtype": "float64",
      "min": -80.0,
      "max": -80.0,
      "mean": -80.0,
      "p01": -80.0,
      "p05": -80.0,
      "p95": -80.0,
      "p99": -80.0
    }
  },
  "dimensions": {
    "yCoordinates": 4590,
    "xCoordinates": 4635,
    "phony_dim_5": 2
  },
  "coordinates": {
    "projection": {
      "size": 1,
      "dtype": "uint32",
      "min": 32633.0,
      "max": 32633.0
    },
    "xCoordinates": {
      "size": 4635,
      "dtype": "float64",
      "min": 148360.0,
      "max": 519080.0
    },
    "yCoordinates": {
      "size": 4590,
      "dtype": "float64",
      "min": 5392840.0,
      "max": 5759960.0
    }
  },
  "compatible_groups": null,
  "example_assets": "s3://sds-n-cumulus-prod-nisar-products/NISAR_L2_GCOV_BETA_V1/NISAR_L2_PR_GCOV_002_109_D_063_4005_DHDH_A_20251012T182508_20251012T182531_X05010_N_P_J_001/NISAR_L2_PR_GCOV_002_109_D_063_4005_DHDH_A_20251012T182508_20251012T182531_X05010_N_P_J_001.h5",
  "sample_asset_raster_info": null,
  "links": [
    {
      "rel": "tilejson",
      "href": "https://openveda.cloud/api/titiler-cmr/xarray/WebMercatorQuad/tilejson.json?collection_concept_id=C3622214170-ASF&variables=HHHH&group=science/LSAR/GCOV/grids/frequencyB&temporal={temporal}",
      "title": "TileJSON",
      "type": "application/json"
    },
    {
      "rel": "map",
      "href": "https://openveda.cloud/api/titiler-cmr/xarray/WebMercatorQuad/map.html?collection_concept_id=C3622214170-ASF&variables=HHHH&group=science/LSAR/GCOV/grids/frequencyB&temporal={temporal}",
      "title": "Map viewer",
      "type": "text/html"
    },
    {
      "rel": "tile",
      "href": "https://openveda.cloud/api/titiler-cmr/xarray/tiles/WebMercatorQuad/{z}/{x}/{y}?collection_concept_id=C3622214170-ASF&variables=HHHH&group=science/LSAR/GCOV/grids/frequencyB&temporal={temporal}",
      "title": "Map tile",
      "type": "image/png"
    }
  ]
}

In [6]:

Copied!





group_variables = list((grouped_compatibility_response.get("variables") or {}).keys())
group_tilejson_link = next(
    link["href"]
    for link in grouped_compatibility_response.get("links", [])
    if link["rel"] == "tilejson"
)

print("selected group:", selected_group)
print("variables in selected group (first 10):", group_variables[:10])
print("tilejson link for selected group:", group_tilejson_link)
group_variables = list((grouped_compatibility_response.get("variables") or {}).keys())
group_tilejson_link = next(
    link["href"]
    for link in grouped_compatibility_response.get("links", [])
    if link["rel"] == "tilejson"
)

print("selected group:", selected_group)
print("variables in selected group (first 10):", group_variables[:10])
print("tilejson link for selected group:", group_tilejson_link)

selected group: science/LSAR/GCOV/grids/frequencyB
variables in selected group (first 10): ['HHHH', 'HVHV', 'listOfCovarianceTerms', 'listOfPolarizations', 'mask', 'numberOfLooks', 'numberOfSubSwaths', 'rtcGammaToSigmaFactor', 'xCoordinateSpacing', 'yCoordinateSpacing']
tilejson link for selected group: https://openveda.cloud/api/titiler-cmr/xarray/WebMercatorQuad/tilejson.json?collection_concept_id=C3622214170-ASF&variables=HHHH&group=science/LSAR/GCOV/grids/frequencyB&temporal={temporal}

For a grouped dataset like NISAR GCOV, the workflow is:

Call /compatibility without group to get candidate compatible_groups.
Choose one of those group paths.
Call /compatibility again with group=... to inspect variables and dimensions for that group.
Pick a variable name from variables.
Reuse the returned xarray links from the grouped response.

That keeps the root compatibility response lightweight while still giving you a practical path to the next xarray request.