TiTiler-CMR Compatibility Report¶

What datasets are compatible with TiTiler-CMR?¶

This report summarizes the 2026-07-01 compatibility assessment for TiTiler-CMR tiling endpoints. It distinguishes three outcomes:

Compatible: the assessment found a tileable asset.
Incompatible: the assessment found a concrete data, asset, access, render, request, or service failure.
Inconclusive: the assessment could not make a compatibility determination. These datasets are probably incompatible with TiTiler-CMR as currently configured, but they may be worth exploring further when the blocker is operational, sampling-related, or caused by incomplete metadata.

The current assessment uses standard access paths. Authentication and credential failures are therefore visible in this report and should not be compared directly with older reports that bypassed some access checks.

This report provides:

A searchable listing of NASA EOSDIS datasets and their tiling compatibility outcome.
An overview of the dataset and failure categories observed in the assessment.

See titiler-cmr-compatibility's METHODOLOGY documentation for a detailed explanation of how datasets were tested.

What datasets are compatible?¶

Out of 9665 datasets tested, 931 were found to be compatible. The assessment also found 5434 incompatible datasets and 3300 inconclusive datasets.

Searchable Dataset Table¶

To search across all datasets (where a granule was found) and check compatibility status, view the interactive table:

→ View Interactive Dataset Table

To determine if a particular dataset is compatible, search for it by name and check the Tiling Compatible column.

To regenerate the table, run the generate_table.py script.

What types of datasets are compatible?¶

Compatibility is concentrated in datasets whose assets can be opened by TiTiler-CMR's rasterio or xarray-backed readers and whose metadata identifies a tileable variable and spatial coordinates. The charts below summarize compatible datasets by data center, processing level, and file extension.

Compatible Datasets by Data Center¶

No description has been provided for this image

Percentage of compatible datasets by data center¶

Out[3]:

	Data Center	% Tiling-Compatible Datasets
0	ASF	14.86%
1	AVISO	0.00%
2	Aquarius Project NASA/CONAE	0.00%
3	Aquarius Project NASA/CONAE, JPL	100.00%
4	BYU/SCP	0.00%
...	...	...
62	UTX-AUSTIN/CSR	0.00%
63	UWI-MAD/SSEC	0.00%
64	UWI-MAD/SSEC/ASIPS	0.00%
65	UWI-MAD/SSEC/CIMSS	100.00%
66	University of Colorado, Boulder	100.00%

67 rows × 2 columns

Compatible datasets by backend¶

This table shows which backend found a tileable asset for each compatible dataset.

Out[4]:

	Backend	Compatible Datasets	Share of Compatible Datasets
0	xarray	771	82.8%
1	rasterio	160	17.2%

Compatible Datasets by Processing Level¶

Compatible Datasets by File Extension¶

Not all datasets have the format attribute, so we assess format using the extension attribute.

What datasets are incompatible or inconclusive, and why?¶

The 2026 assessment reports normalized categories derived from assessment_status, failure_category, failure_subcategory, error_code, HTTP status fields, endpoints, and selected raw_error_body patterns. Service errors and access failures are reported separately from proven data incompatibilities so operational issues do not distort the data-compatibility picture. In practice, inconclusive datasets should be treated as likely incompatible until a targeted follow-up confirms otherwise. The Unsupported asset category usually means the assessment found a granule data asset, but TiTiler-CMR could not select a supported reader for it because the asset has an unsupported or unrecognized file type, media type, file signature, or because the asset href/extension metadata needed to identify the file type was missing.

Assessment status breakdown¶

assessment_status remains visible because an inconclusive result is not the same as a proven incompatible dataset. Most inconclusive rows are probably incompatible with TiTiler-CMR today, but they are also the rows most worth revisiting when the blocker is missing granules, missing or invalid bbox metadata, service errors, or sampling/probe limits.

Inconclusive raw error summary¶

These are the most common raw error bodies among inconclusive operational outcomes. Metadata gaps are excluded here so service, render, request, and access errors are easier to inspect.

Out[9]:

	Category	Reason Detail	Raw Error Snippet	Datasets
350	Service error	Internal NoneType access error	{"detail":"'NoneType' object has no attribute ...	973
352	Service error	Invalid media type surfaced as service error	{"detail":"0 has an invalid media type"}	111
353	Service error	Service unavailable response	{"message":"Service Unavailable"}	53
294	Render error	Bbox probe attempt limit exceeded	{"detail":"'air_pres'"}	27
321	Render error	HTTP 500 response	{"message":"Internal Server Error"}	25
349	Service error	Internal CRS authority lookup error	{"detail":"'NoneType' object has no attribute ...	18
351	Service error	Internal bytes/string media-type check error	{"detail":"'in <string>' requires string as le...	13
296	Render error	Bbox probe attempt limit exceeded	{"detail":"'cloud_frac_pres_layer'"}	9
295	Render error	Bbox probe attempt limit exceeded	{"detail":"'air_pres_stand'"}	6
299	Render error	Bbox probe attempt limit exceeded	{"detail":"'land_bands'"}	6
304	Render error	Bbox probe attempt limit exceeded	{"detail":"invalid literal for int() with base...	5
302	Render error	Bbox probe attempt limit exceeded	{"detail":"Could not automatically create Pand...	4
318	Render error	HTTP 500 response	{"detail":"index must be monotonic increasing ...	3
298	Render error	Bbox probe attempt limit exceeded	{"detail":"'emissivity_band'"}	3
300	Render error	Bbox probe attempt limit exceeded	{"detail":"'surf_wnum_ir'"}	3

Data incompatibilities¶

Data incompatibilities are cases where the granule asset had an acceptable file type, but the structure or contents of the data were not tileable by TiTiler-CMR. In other words, TiTiler-CMR could choose a reader for the file, but the opened dataset did not satisfy the assumptions needed to render map tiles. A basic requirement for TiTiler-CMR compatibility is gridded data with usable spatial coordinates, either X/Y or latitude/longitude coordinates. Many incompatible datasets do not meet that requirement. Common examples include missing X/Y or lat/lon spatial coordinate metadata, no tileable variables, unsupported dimensionality, unsupported file signatures, and decode errors.

Metadata, access, and service outcomes¶

Metadata gaps include missing granules and invalid or missing granule bounding boxes. Inaccessible/authentication failures include S3 credential lookup failures, including IMDS token lookup failures observed in raw_error_body. Service, render, and request errors are kept as operational categories; they should not be treated as compatible just because they may be transient. They are better read as likely incompatible results that deserve targeted investigation if the dataset is important.

Summary and next steps¶

The current report is regenerated from assessment-results-2026-07-01.parquet and uses shared categorization code in reporting.py. Individual datasets may still need targeted testing before being advertised as supported, but the normalized categories make future assessment updates easier to review.

Potential follow-up work:

File endpoint issues for recurring server-side errors, such as NoneType failures returned as HTTP 500 responses.
Re-run the assessment after endpoint or credential-environment fixes.
Continue improving metadata-based identification of datasets that should work with TiTiler-CMR.