Abstract
Frequently, regionalized positive variables are treated by preliminarily applying a logarithm, and kriging estimates are back-transformed using classical formulae for the expectation of a lognormal random variable. This practice has several problems (lack of robustness, non-optimal confidence intervals, etc.), particularly when estimating block averages. Therefore, many practitioners take exponentials of the kriging estimates, although the final estimations are deemed as non-optimal. Another approach arises when the nature of the sample space and the scale of the data are considered. Since these concepts can be suitably captured by an Euclidean space structure, we may define an optimal kriging estimator for positive variables, with all properties analogous to those of linear geostatistical techniques, even for the estimation of block averages. In this particular case, no assumption on preservation of lognormality is needed. From a practical point of view, the proposed method coincides with the median estimator and offers theoretical ground to this extended practice. Thus, existing software and routines remain fully applicable.
Similar content being viewed by others
References
Aitchison J, Brown JAC (1957) The lognormal distribution. Cambridge University Press, Cambridge, p 176
Armstrong M (1992) Positive definiteness is not enough. Math Geol 24(1):135–143
Carrera J, Neuman S (1986) Estimation of aquifer parameters under steady-state and transient conditions: I. background and statistical framework. Water Resources Res 22(2):199–210
Chilès J-P, Delfiner P (1999) Geostatistics—modeling spatial uncertainty. Wiley, New York, p 695
Clark I, Harper WV (2000) Practical geostatistics 2000. Ecosse North America Llc, Columbus, p 342
Cressie N (1991) Statistics for spatial data. Wiley, New York, p 900
David M (1977) Geostatistical ore reserve estimation. Elsevier, New York, p 364
Deutsch C, Journel A (1992) GSLIB—geostatistical software library and user’s guide. Oxford University Press, New York, p 340
Dowd PA (1982) Lognormal kriging—the general case. Math Geol 14(5):475–499
Eaton ML (1983) Multivariate statistics: a vector space approach. Wiley, New York, p 512
Emery X (2004) On the consistency of the indirect lognormal correction. Stoch Envir Res Risk Ass 18(4):258–264
Journel AG (1980) The lognormal approach to predicting local distributions of selective mining unit grades. Math Geol 12(4):285–303
Journel AG, Huijbregts CJ (1978) Mining geostatistics. Academic, London, p 600
Marcotte D, Groleau P (1997) A simple and robust lognormal estimator. Math Geol 29(8):993–1009
Mateu-Figueras G, Pawlowsky-Glahn V, Martín-Fernández JA (2002) Normal in ℜ + vs lognormal in ℜ. In: Bayer U, Burger H, Skala W (eds) Proceedings of IAMG’02—the eighth annual conference of the international association for mathematical geology. Selbstverlag der Alfred-Wegener-Stiftung, Berlin, p 1106
McAlister D (1879) The law of the geometric mean. Proc Roy Soc Lond 29(1):367–376
Myers DE (1982) Matrix formulation of co-kriging. Math Geol 14(3):249–257
Nielsen OA (1997) An introduction to integration and measure theory. Wiley, New York, p 473
Pardo-Igúzquiza E, Dowd P (2001) Variance-covariance matrix of the experimental variogram: assessing variogram uncertainty. Math Geol 33(4):397–419
Pawlowsky-Glahn V (2003) Statistical modelling on coordinates. In: Thió-Henestrosa S, Martín-Fernández JA (eds) Compositional data analysis workshop—CoDaWork’03, proceedings, University of Girona, ISBN 84-8458-111-X, http://ima.udg.es/Activitats/CoDaWork03/
Pawlowsky-Glahn V, Egozcue JJ (2001) Geometric approach to statistical analysis on the simplex. Stoch Environ Res Risk Assess (SERRA) 15(5):384–398
Pawlowsky-Glahn V, Olea RA (2004) Geostatistical analysis of compositional data. Oxford University Press, New York, p 181
Pukelsheim F (1994) The three sigma rule. Am Stat 48(2):88–91
Rendu J-M (1979) Normal and lognormal estimation. Math Geol 11(4):407–422
Rivoirard J (1990) A review of lognormal estimators for in situ reserves (teacher’s aid). Math Geol 22(2):213–221
Roth C (1998) Is lognormal kriging suitable for local estimation? Math Geol 30(8):999–1009
Sichel H (1971) On a family of discrete distributions particularly suited to represent long-tailed frequency data. In: Laubscher NF (ed) Third sym mathem statist, Pretoria, South Africa, p 51–97
Tolosana-Delgado R (2005) Geostatistics for constrained variables: positive data, compositions and probabilities. Applications to environmental hazard monitoring. PhD thesis, IMAMB-Institut de Medi Ambient. University of Girona, Spain, p 214. Available online at http://www.tdx.cesca.es/TDX-0123106-122444/index_an.html
van den Boogaart KG, Brenning A (2001) Why is universal kriging better than irfk-kriging: estimation of the variogram in the presence of trend. In: Ross, G (ed) Proceedings of IAMG’01—the sixth annual conference of the international association for mathematical geology (CD-ROM)
Verly G (1983) The multiGaussian approach and its applications to the estimation of local reserves. Math Geol 15(2):259–286
Vysochanskij DF, Petunin YI (1980) Justification of the 3 sigma rule for unimodal distributions. Theory Probab Math Stat 21:25–36
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Tolosana-Delgado, R., Pawlowsky-Glahn, V. Kriging Regionalized Positive Variables Revisited: Sample Space and Scale Considerations. Math Geol 39, 529–558 (2007). https://doi.org/10.1007/s11004-007-9107-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11004-007-9107-7