GNGTS 2022 - Atti del 40° Convegno Nazionale

506 GNGTS 2022 Sessione 3.3 Discussion. The result in Fig. 3 shows the performance of function (6) which considers the spatial structure between the data (semivariograms), for the NN interpolation scheme. In the case shown, we do not consider the effect of the nugget (b in the formulas), which agrees with the property that the original data values are maintained at the reference data points (the colours start from blue = 0 to red = maximum value). As might to be expected, the maximum variance values are found at the edges of V i polygons, particularly in less information-dense areas. The formula can be applied to other interpolation methods based on convex combination. Given the promise of the method, we will show further developments, like an exponential semivariogram model and testing on experimental datasets in the future. References Chilès J. P. and Delfiner P.; 1999: Geostatistics . Modeling Spatial Uncertainty. Wiley Series is Probability and Statistics, 695. Etherington T. R.; 2020. Discrete natural neighbour interpolation with uncertainty using cross-validation error- distance fields . PeerJ Comput. Sci. 6 :e282, 1-16, http://doi.org/10.7717/peerj-cs.282 Iurcev M., Pettenati F., Diviacco P.; 2021; Gridding, boundary definition and interpolation methods for near real-time spatial data. Bull. Geoph. and Oceanog., 62 , 3, 427-454, September 2021; doi:10.4430/bgta0360. Sibson R.; 1980. A vector identity for Dirichlet tessellation . Math. Proc. Camb. Philos. Soc. 87 , 151–155. Fig. 3 - the result of function (6) that considers the structure of data in Fig. 1. (the colours start from blue = 0 to red = maximum value).