We analyse the Dioxin (TCDD) pollution of the Seveso (Milan, Italy) te
rritory, seeking to statistically parametrize it in terms of Universal
Multifractals. The data set contains the measurements collected from
1976 up to 1981. We apply the Double Trace Moment (DTM) technique in o
rder to estimate both a (the degree of multifractality) and C-1 (the c
odimension of the mean field) and (with the help of spectral analysis)
we also calculate H (the degree of non-conservation of the process).
We then discuss the effects introduced by statistical undersampling an
d network sparseness and provide a way to statistically correct for th
ese effects. We conclude that the ground distribution of Dioxin shows
clear multifractal features and can be classified as an unconditionall
y hard universal multifractal process.