Annual NLCD Land Cover Confidence
The land cover confidence represents the probability value for the land cover class derived from the classification method.
Annual NLCD land cover product generation strongly relies on supervised classification that is implemented with a series of deep learning models. The final result from the classification system is a discrete probability distribution across the output classes. The land cover confidence product provides the probability value for the final output land cover class. The index ranges from low (1) to high (100) model confidence.
Data Access
Products can be accessed via the data access page.
Documents
Additional information on Annual NLCD and Annual NLCD science products can be found in the Science Product User Guide.
Annual NLCD Citation
Annual NLCD has no restrictions on the use of science products. Annual NLCD does ask that if you use the data as part of a publication or presentation that you use the following citation below; however, it is not a requirement to use the data.
U.S. Geological Survey (USGS), 2024, Annual NLCD Collection 1 Science Products: U.S. Geological Survey data release, https://doi.org/10.5066/P94UXNTS
Land Cover Confidence Characteristics, Constraints, and Caveats
The following artifacts were discovered in the Land Cover Confidence product.
- Value/Range Truncation: Linear regression predictions and the nature of the ensemble approach can give values greater than 100, which were not properly truncated.
-
Confidence Value Misrepresentations: The wrong class can be referenced for the associated confidence value due to modeling on an expanded set of classes which get cross-walked to the final land cover calls. This often results in a much lower confidence value than intended.
Back to the Annual NLCD Product Suite page.
The land cover confidence represents the probability value for the land cover class derived from the classification method.
Annual NLCD land cover product generation strongly relies on supervised classification that is implemented with a series of deep learning models. The final result from the classification system is a discrete probability distribution across the output classes. The land cover confidence product provides the probability value for the final output land cover class. The index ranges from low (1) to high (100) model confidence.
Data Access
Products can be accessed via the data access page.
Documents
Additional information on Annual NLCD and Annual NLCD science products can be found in the Science Product User Guide.
Annual NLCD Citation
Annual NLCD has no restrictions on the use of science products. Annual NLCD does ask that if you use the data as part of a publication or presentation that you use the following citation below; however, it is not a requirement to use the data.
U.S. Geological Survey (USGS), 2024, Annual NLCD Collection 1 Science Products: U.S. Geological Survey data release, https://doi.org/10.5066/P94UXNTS
Land Cover Confidence Characteristics, Constraints, and Caveats
The following artifacts were discovered in the Land Cover Confidence product.
- Value/Range Truncation: Linear regression predictions and the nature of the ensemble approach can give values greater than 100, which were not properly truncated.
-
Confidence Value Misrepresentations: The wrong class can be referenced for the associated confidence value due to modeling on an expanded set of classes which get cross-walked to the final land cover calls. This often results in a much lower confidence value than intended.
Back to the Annual NLCD Product Suite page.