NodataWerten

NodataWerten, often written as NoData values in English, refer to missing or undefined values in raster and gridded datasets. They indicate locations where data are unavailable, irrelevant, or outside the capture extent. Rather than real numeric data, a NoData value serves as a sentinel that informs analysis and visualization routines how to treat those cells.

In raster formats such as GeoTIFF, NetCDF, or HDF, a specific value is designated as the NoData value for each band or for the dataset as a whole. This value is stored in the file's metadata and is used by software to mask or ignore those cells during processing. While common sentinel values have included numbers like -9999, any value can be chosen as long as it does not collide with valid data. Some formats allow per-band NoData definitions.

NoData values are conceptually distinct from NaN (Not a Number) in floating-point arithmetic or from SQL NULL. NaN is a result of a computation, whereas NoData is a dataset-level indicator of missing information. Software often renders NoData as transparent or uses it to exclude cells from statistics, histogram calculations, resampling, and classification, unless a mask overrides this behavior.

Handling NoData involves masking, data cleaning, or interpolation that respects missing data, as well as using tools that compute statistics or render imagery while ignoring NoData. Proper use and documentation of NoData values are essential for accurate spatial analysis and data integrity.