Cloud Score+ is a quality assessment (QA) processor for medium-to-high
resolution optical satellite imagery. The Cloud Score+ S2_HARMONIZED
dataset is being operationally produced from the
harmonized Sentinel-2 L1C collection,
and Cloud Score+ outputs can be used to identify relatively clear pixels and
effectively remove clouds and cloud shadows from L1C (Top-of-Atmosphere)
or L2A (Surface Reflectance)
imagery.
The Cloud Score+ S2_HARMONIZED dataset includes two QA bands, cs and
cs_cdf, that both grade the usability of individual pixels with respect to
surface visibility on a continuous scale between 0 and 1, where 0 represents
"not clear" (occluded), while 1 represents "clear" (unoccluded) observations.
The cs band scores QA based on a spectral distance between the observed
pixel and a (theoretical) clear reference observation, while the cs_cdf band
represents the likelihood an observed pixel is clear based on an estimated
cumulative distribution of scores for a given location through time. In
other words, cs can be thought of as a more instantaneous atmospheric
similarity score (i.e., how similar is this pixel to what we'd expect to
see in a perfectly a clear reference), while cs_cdf captures an expectation
of the estimated score through time (i.e., if we had all the scores for this
pixel through time, how would this score rank?).
Images in the Cloud Score+ S2_HARMONIZED collection have the same id and
system:index properties as the individual Sentinel-2 L1C
assets from which they were produced such that Cloud Score+ bands can be
linked to source images based on their shared system:index.
Cloud Score+ backfill for the entire Sentinel-2 archive is currently in
progress and Dataset Availability dates will be updated periodically as new
results are added to the Cloud Score+ collection.
For more information about the Cloud Score+ dataset and modelling
approach, see
this Medium post.
Bands
Pixel Size 10 meters
Bands
Name
Units
Min
Max
Pixel Size
Description
cs
Dimensionless
0
1
meters
Pixel quality score based on spectral distance from a
(theoretical) clear reference
cs_cdf
Dimensionless
0
1
meters
Value of the cumulative distribution function of possible
cs values for the estimated cs value
Image Properties
Image Properties
Name
Type
Description
DATE_PRODUCT_GENERATED
STRING
Production date.
MGRS_TILE
STRING
Sentinel-2 Military Grid Reference System ID.
MODEL_VERSION
STRING
Cloud Score+ model version.
NO_CONTEXT_FRACTION
DOUBLE
Fraction of subtiles processed with no temporal context.
Pasquarella, V. J., Brown, C. F., Czerwinski, W., & Rucklidge, W. J. (2023)
Comprehensive Quality Assessment of Optical Satellite Imagery Using Weakly
Supervised Video Learning. In Proceedings of the IEEE/CVF Conference on
Computer Vision and Pattern Recognition (pp. 2124-2134).
PDF
Cloud Score+ is a quality assessment (QA) processor for medium-to-high resolution optical satellite imagery. The Cloud Score+ S2_HARMONIZED dataset is being operationally produced from the harmonized Sentinel-2 L1C collection, and Cloud Score+ outputs can be used to identify relatively clear pixels and effectively remove clouds and cloud shadows from L1C …
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],[],[[["\u003cp\u003eCloud Score+ is a quality assessment tool for optical satellite imagery, aiding in the identification and removal of clouds and cloud shadows.\u003c/p\u003e\n"],["\u003cp\u003eIt offers two QA bands (\u003ccode\u003ecs\u003c/code\u003e and \u003ccode\u003ecs_cdf\u003c/code\u003e) to grade pixel usability, where 0 signifies occlusion and 1 represents clear observations.\u003c/p\u003e\n"],["\u003cp\u003eThe dataset is derived from the harmonized Sentinel-2 L1C collection and shares the same ID and system index for easy linking.\u003c/p\u003e\n"],["\u003cp\u003eCloud Score+ backfill for the Sentinel-2 archive is ongoing, with dataset availability dates updated as new data is added.\u003c/p\u003e\n"],["\u003cp\u003eUsers can access the dataset through Google Earth Engine and apply it to both L1C and L2A collections for clearer imagery analysis.\u003c/p\u003e\n"]]],["The Cloud Score+ S2_HARMONIZED dataset, provided by Google Earth Engine, assesses the quality of satellite imagery from 2015-06-27 to 2025-02-23. It grades pixel usability for surface visibility using `cs` and `cs_cdf` bands, with scores ranging from 0 (not clear) to 1 (clear). Cloud Score+ identifies clear pixels and removes clouds/shadows from Sentinel-2 imagery. It can be used on L1C or L2A imagery. Users can link Cloud Score+ bands to source images.\n"],null,["Dataset Availability\n: 2015-06-27T00:00:00Z--2025-09-04T15:18:33.642000Z\n\nDataset Provider\n:\n\n\n [Google Earth Engine](https://earthengine.google.com/)\n\nTags\n:\n[cloud](/earth-engine/datasets/tags/cloud) [google](/earth-engine/datasets/tags/google) [satellite-imagery](/earth-engine/datasets/tags/satellite-imagery) [sentinel2-derived](/earth-engine/datasets/tags/sentinel2-derived) \n\nDescription \nCloud Score+ is a quality assessment (QA) processor for medium-to-high\nresolution optical satellite imagery. The Cloud Score+ S2_HARMONIZED\ndataset is being operationally produced from the\n[harmonized Sentinel-2 L1C collection](https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S2_HARMONIZED),\nand Cloud Score+ outputs can be used to identify relatively clear pixels and\neffectively remove clouds and cloud shadows from [L1C (Top-of-Atmosphere)](https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S2_HARMONIZED)\nor [L2A (Surface Reflectance)](https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S2_SR_HARMONIZED)\nimagery.\n\nThe Cloud Score+ S2_HARMONIZED dataset includes two QA bands, `cs` and\n`cs_cdf`, that both grade the usability of individual pixels with respect to\nsurface visibility on a continuous scale between 0 and 1, where 0 represents\n\"not clear\" (occluded), while 1 represents \"clear\" (unoccluded) observations.\nThe `cs` band scores QA based on a spectral distance between the observed\npixel and a (theoretical) clear reference observation, while the `cs_cdf` band\nrepresents the likelihood an observed pixel is clear based on an estimated\ncumulative distribution of scores for a given location through time. In\nother words, `cs` can be thought of as a more instantaneous atmospheric\nsimilarity score (i.e., how similar is this pixel to what we'd expect to\nsee in a perfectly a clear reference), while `cs_cdf` captures an expectation\nof the estimated score through time (i.e., if we had all the scores for this\npixel through time, how would this score rank?).\n\nImages in the Cloud Score+ S2_HARMONIZED collection have the same id and\n`system:index` properties as the individual [Sentinel-2 L1C](https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S2_HARMONIZED)\nassets from which they were produced such that Cloud Score+ bands can be\nlinked to source images based on their shared `system:index`.\n\nCloud Score+ backfill for the entire Sentinel-2 archive is currently in\nprogress and Dataset Availability dates will be updated periodically as new\nresults are added to the Cloud Score+ collection.\n\nFor more information about the Cloud Score+ dataset and modelling\napproach, see\n[this Medium post](https://medium.com/google-earth/all-clear-with-cloud-score-bd6ee2e2235e).\n\nBands\n\n\n**Pixel Size**\n\n10 meters\n\n**Bands**\n\n| Name | Units | Min | Max | Pixel Size | Description |\n|----------|---------------|-----|-----|------------|----------------------------------------------------------------------------------------------------|\n| `cs` | Dimensionless | 0 | 1 | meters | Pixel quality score based on spectral distance from a (theoretical) clear reference |\n| `cs_cdf` | Dimensionless | 0 | 1 | meters | Value of the cumulative distribution function of possible `cs` values for the estimated `cs` value |\n\nImage Properties\n\n**Image Properties**\n\n| Name | Type | Description |\n|-----------------------------|--------|----------------------------------------------------------|\n| DATE_PRODUCT_GENERATED | STRING | Production date. |\n| MGRS_TILE | STRING | Sentinel-2 Military Grid Reference System ID. |\n| MODEL_VERSION | STRING | Cloud Score+ model version. |\n| NO_CONTEXT_FRACTION | DOUBLE | Fraction of subtiles processed with no temporal context. |\n| PROCESSING_SOFTWARE_VERSION | STRING | Cloud Score+ processing software version. |\n| SOURCE_ASSET_ID | STRING | Earth Engine Asset ID for source image. |\n| SOURCE_PRODUCT_ID | STRING | Sentinel-2 Product ID for source image. |\n\nTerms of Use\n\n**Terms of Use**\n\n[CC-BY-4.0](https://spdx.org/licenses/CC-BY-4.0.html)\n\nCitations \nCitations:\n\n- Pasquarella, V. J., Brown, C. F., Czerwinski, W., \\& Rucklidge, W. J. (2023)\n Comprehensive Quality Assessment of Optical Satellite Imagery Using Weakly\n Supervised Video Learning. In Proceedings of the IEEE/CVF Conference on\n Computer Vision and Pattern Recognition (pp. 2124-2134).\n [PDF](https://openaccess.thecvf.com/content/CVPR2023W/EarthVision/html/Pasquarella_Comprehensive_Quality_Assessment_of_Optical_Satellite_Imagery_Using_Weakly_Supervised_CVPRW_2023_paper.html)\n\nExplore with Earth Engine **Important:** Earth Engine is a platform for petabyte-scale scientific analysis and visualization of geospatial datasets, both for public benefit and for business and government users. Earth Engine is free to use for research, education, and nonprofit use. To get started, please [register for Earth Engine access.](https://console.cloud.google.com/earth-engine)\n\nCode Editor (JavaScript) \n\n```javascript\n// Harmonized Sentinel-2 Level 2A collection.\nvar s2 = ee.ImageCollection('COPERNICUS/S2_SR_HARMONIZED');\n\n// Cloud Score+ image collection. Note Cloud Score+ is produced from Sentinel-2\n// Level 1C data and can be applied to either L1C or L2A collections.\nvar csPlus = ee.ImageCollection('GOOGLE/CLOUD_SCORE_PLUS/V1/S2_HARMONIZED');\n\n// Region of interest.\nvar ROI = ee.Geometry.Point(-119.9087, 37.4159);\n\n// Use 'cs' or 'cs_cdf', depending on your use case; see docs for guidance.\nvar QA_BAND = 'cs_cdf';\n\n// The threshold for masking; values between 0.50 and 0.65 generally work well.\n// Higher values will remove thin clouds, haze & cirrus shadows.\nvar CLEAR_THRESHOLD = 0.60;\n\n// Make a clear median composite.\nvar composite = s2\n .filterBounds(ROI)\n .filterDate('2023-01-01', '2023-02-01')\n .linkCollection(csPlus, [QA_BAND])\n .map(function(img) {\n return img.updateMask(img.select(QA_BAND).gte(CLEAR_THRESHOLD));\n })\n .median();\n\n// Sentinel-2 visualization parameters.\nvar s2Viz = {bands: ['B4', 'B3', 'B2'], min: 0, max: 2500};\n\nMap.addLayer(composite, s2Viz, 'median composite');\nMap.centerObject(ROI, 11);\n```\n[Open in Code Editor](https://code.earthengine.google.com/?scriptPath=Examples:Datasets/GOOGLE/GOOGLE_CLOUD_SCORE_PLUS_V1_S2_HARMONIZED) \n[Cloud Score+ S2_HARMONIZED V1](/earth-engine/datasets/catalog/GOOGLE_CLOUD_SCORE_PLUS_V1_S2_HARMONIZED) \nCloud Score+ is a quality assessment (QA) processor for medium-to-high resolution optical satellite imagery. The Cloud Score+ S2_HARMONIZED dataset is being operationally produced from the harmonized Sentinel-2 L1C collection, and Cloud Score+ outputs can be used to identify relatively clear pixels and effectively remove clouds and cloud shadows from L1C ... \nGOOGLE/CLOUD_SCORE_PLUS/V1/S2_HARMONIZED, cloud,google,satellite-imagery,sentinel2-derived \n2015-06-27T00:00:00Z/2025-09-04T15:18:33.642000Z \n-90 -180 90 180 \nGoogle Earth Engine \nhttps://developers.google.com/earth-engine/datasets\n\n- [](https://doi.org/https://earthengine.google.com/)\n- [](https://doi.org/https://developers.google.com/earth-engine/datasets/catalog/GOOGLE_CLOUD_SCORE_PLUS_V1_S2_HARMONIZED)"]]
