DESCRIPTION

Removing unreliable pixels is a fundamental and one of the first steps in remote sensing. Landsat8 provides a Quality Assessment (QA) band which can be used for this purpose.

The i.landsat8.qc module generates reclassification rule files which can be used in r.reclass for filtering the QA band according to pixel quality characteristics the user defines as unacceptable.

Values defined as unacceptable for a given condition will be set to NULL in the output raster map. All other values will be set to 1.

The Quality Assessment (QA) band from Landsat8 contains 16bit integer values that represent "bit-packed combinations of surface, atmosphere, and sensor conditions that can affect the overall usefulness of a given pixel".
The QA band is currently only available for Landsat 8 OLI/TIRS data.

The following quality relevant conditions are represented as "single bits" in the Landsat8 QA band:

Condition QA Bit Position
Designated Fill 0
Dropped Frame 1
Terrain Occlusion 2
Reserved (not currently used) 3

Possible choices for the "single bits" are:

Value Description Bit representation
No This condition does not exist 0
Yes This condition exists 1

The following quality relevant conditions are represented as "double bits" in the Landsat8 QA band:

Condition QA Bit Position
Water Confidence 4-5
Reserved (not currently used) 6-7
Vegetation Confidence 8-9
Snow/Ice Confidence 10-11
Cirrus Confidence 12-13
14-15 Cloud Confidence

Possible choices for the "double bits" are:

Value Description Bit representation
Not Determined Algorithm did not determine the status of this condition 00
No Algorithm has low to no confidence that this condition exists (0-33 percent confidence) 01
Maybe Algorithm has medium confidence that this condition exists (34-66 percent confidence) 10
Yes Algorithm has high confidence that this condition exists (67-100 percent confidence) 11

NOTES

The Landsat Quality Assessment band is an artificial band which represents an analysis based on defined algorithms. The USGS provides the users with the following note on how the QA band should be used:

"Rigorous science applications seeking to optimize the value of pixels used in a study will find QA bits useful as a first level indicator of certain conditions. Otherwise, users are advised that this file contains information that can be easily misinterpreted and it is not recommended for general use."

EXAMPLE

#Create a cloud mask:
i.landsat8.qc --overwrite --verbose cloud="Maybe,Yes" \
    output=./Cloud_Mask_rules.txt
r.reclass input=LC81980182015183LGN00_BQA \
    output=LC81980182015183LGN00_Cloud_Mask rules=./Cloud_Mask_rules.txt

#Create a water mask:
i.landsat8.qc --overwrite --verbose water="Maybe,Yes" \
    output=./Water_Mask_rules.txt
r.reclass input=LC81980182015183LGN00_BQA \
    output=LC81980182015183LGN00_Water_Mask rules=./Water_Mask_rules.txt

SEE ALSO

r.reclass, i.modis.qc, r.bitpattern, i.landsat8.swlst

REFERENCES

http://landsat.usgs.gov/qualityband.php https://landsat.usgs.gov/documents/LDCM_CVT_ADD.pdf

AUTHOR

Stefan Blumentrath, Norwegian Institute for Nature Research, Oslo (Norway)

Last changed: $Date: 2016-05-20 21:01:30 +0000 (Fri, 20 May 2016) $