An interpolation method is an algorithm for estimating values at unsampled locations from values at known locations, and SWEET models it as a kind of Algorithm tied to the Interpolation data service.12
SWEET’s Turtle file makes InterpolationMethod a subclass of Algorithm with a restriction on the Interpolation data service, and lists distance weighting and nearest neighbour as individual methods.1 The USACE HEC documentation describes nearest neighbour as assigning each grid cell the value of the closest gauge, and inverse distance weighting as computing a weight for each contributing gauge.3 Both sources name these methods, one as ontology individuals and the other as computations.
A textbook chapter explains why more than one method exists: infinitely many functions pass through the given points, so additional conditions must be imposed, and these conditions define the character of each technique.4 QGIS’s manual draws the practical consequence that the choice should depend on the sample data, the type of surface to be generated and the tolerance of estimation errors.2
Resampling a raster is a related but distinct operation, since the same chapter lists it among the uses of interpolation.4 Retained sources do not establish a ranking of methods, accuracy comparisons or error bounds. SWEET’s file defines the class by structure only and adds no definition text.1
Footnotes
-
ESIP Federation, “SWEET reprDataServiceAnalysis.ttl”. https://raw.githubusercontent.com/ESIPFed/sweet/master/src/reprDataServiceAnalysis.ttl ↩ ↩2 ↩3
-
QGIS Project, “Spatial Analysis (Interpolation)”. https://docs.qgis.org/3.44/en/docs/gentle_gis_introduction/spatial_analysis_interpolation.html ↩ ↩2
-
USACE Hydrologic Engineering Center, “Spatial Interpolation Method”. https://www.hec.usace.army.mil/confluence/metdoc/metum/3.2/processing-data/metinterp/computational-methods/spatial-interpolation-method ↩
-
L. Mitas and H. Mitasova, “Spatial interpolation (GIS book chapter 34)”. https://www.geos.ed.ac.uk/~gisteac/gis_book_abridged/files/ch34.pdf ↩ ↩2