Abstract
A multivariate spatial sampling design that uses spatial vine copulas is presented that aims to simultaneously reduce the prediction uncertainty of multiple variables by selecting additional sampling locations based on the multivariate relationship between variables, the spatial configuration of existing locations and the values of the observations at those locations. Novel aspects of the methodology include the development of optimal designs that use spatial vine copulas to estimate prediction uncertainty and, additionally, use transformation methods for dimension reduction to model multivariate spatial dependence. Spatial vine copulas capture non-linear spatial dependence within variables, whilst a chained transformation that uses non-linear principal component analysis captures the non-linear multivariate dependence between variables. The proposed design methodology is applied to two environmental case studies. Performance of the proposed methodology is evaluated through partial redesigns of the original spatial designs. The first application is a soil contamination example that demonstrates the ability of the proposed methodology to address spatial non-linearity in the data. The second application is a forest biomass study that highlights the strength of the methodology in incorporating non-linear multivariate dependence into the design.
Similar content being viewed by others
References
Aas K, Czado C, Frigessi A, Bakken H (2009) Pair-copula constructions of multiple dependence. Insurance 44(2):182–198. doi:10.1016/j.insmatheco.2007.02.001
Atteia O, Dubois JP, Webster R (1994) Geostatistical analysis of soil contamination in the Swiss Jura. Environ Pollut 86(3):315–327. doi:10.1016/0269-7491(94)90172-4
Bandarian EM, Bloom LM, Mueller UA (2008) Direct minimum/maximum autocorrelation factors within the framework of a two structure linear model of coregionalisation. Comput Geosci 34(3):190–200. doi:10.1016/j.cageo.2007.03.015
Bárdossy A (2006) Copula-based geostatistical models for groundwater quality parameters. Water Resour Res 42(11):W11,416. doi:10.1029/2005WR004754
Bárdossy A, Li J (2008) Geostatistical interpolation using copulas. Water Resour Res 44(7):W07,412. doi:10.1029/2007WR006115
Barnett RM, Deutsch CV (2012) Practical implementation of non-linear transforms for modeling geometallurgical variables. In: Abrahamsen P, Hauge R, Kolbjørnsen O (eds) Geostatistics Oslo 2012, Springer, Dordrecht, pp 409–422. doi:10.1007/978-94-007-4153-9_33
Barnett RM, Manchuk JG, Deutsch CV (2014) Projection pursuit multivariate transform. Math Geosci 46(3):337–359. doi:10.1007/s11004-013-9497-7
Bayraktar H, Turalioglu FS (2005) A Kriging-based approach for locating a sampling site—in the assessment of air quality. Stoch Environ Res Risk Assess 19(4):301–305. doi:10.1007/s00477-005-0234-8
Bohorquez M, Giraldo R, Mateu J (2016) Multivariate functional random fields: prediction and optimal sampling. Stoch Environ Res Risk Assess. doi:10.1007/s00477-016-1266-y
Bonneau M, Gaba S, Peyrard N, Sabbadin R (2014) Reinforcement learning-based design of sampling policies under cost constraints in Markov random fields: application to weed map reconstruction. Computat Stat Data Anal 72:30–44. doi:10.1016/j.csda.2013.10.002
Brown PJ, Le ND, Zidek JV (1994) Multivariate spatial interpolation and exposure to air pollutants. Can J Stat 22(4):489–509. doi:10.2307/3315406
Bueso MC, Angulo JM, Cruz-Sanjulián J, García-Aróstegui JL (1999) Optimal spatial sampling design in a multivariate framework. Math Geol 31(5):507–525. doi:10.1023/A:1007511923053
Desbarats AJ, Dimitrakopoulos R (2000) Geostatistical simulation of regionalized pore-size distributions using min/max autocorrelation factors. Math Geol 32(8):919–942. doi:10.1023/A:1007570402430
Diggle P, Lophaven S (2006) Bayesian geostatistical design. Scand J Stat 33(1):53–64
Diggle PJ, Ribeiro PJ (2007) Geostatistical design. Model-based geostatistics. Springer, New York, pp 199–212
Erhardt TM, Czado C, Schepsmeier U (2015a) R-vine models for spatial time series with an application to daily mean temperature. Biometrics 71(2):323–332. doi:10.1111/biom.12279
Erhardt TM, Czado C, Schepsmeier U (2015b) Spatial composite likelihood inference using local C-vines. J Multivar Anal 138(C):74–88. doi:10.1016/j.jmva.2015.01.021
Finley AO, Banerjee S, Carlin P (2007) spBayes: an R package for univariate and multivariate hierarchical point-referenced spatial models. J Stat Softw 19(4):1–24. doi:10.18637/jss.v019.i04
Goovaerts P (1997) Geostatistics for natural resources evaluation. Oxford University Press, New York
Gräler B (2014) Modelling skewed spatial random fields through the spatial vine copula. Spat Stat 10:87–102. doi:10.1016/j.spasta.2014.01.001
Gräler B, Pebesma E (2011) The pair-copula construction for spatial data: a new approach to model spatial dependency. Procedia Environ Sci 7:206–211. doi:10.1016/j.proenv.2011.07.036
Harris P, Clarke A, Juggins S, Brunsdon C, Charlton M (2014) Geographically weighted methods and their use in network re-designs for environmental monitoring. Stoch Environ Res Risk Assess 28(7):1869–1887. doi:10.1007/s00477-014-0851-1
Haslauer CP, Li J, Bárdossy A (2010) Application of copulas in geostatistics. In: Atkinson PM, Lloyd CD (eds) geoENV VII—geostatistics for environmental applications, Springer, Dordrecht, pp 395–404. doi:10.1007/978-90-481-2322-3_34
Hassanipak AA, Sharafodin M (2004) GET: A function for preferential site selection of additional borehole drilling. Explor Mining Geol 13(1–4):139–146. doi:10.2113/gsemg.13.1-4.139
Joe H (1996) Families of m-variate distributions with given margins and m(m-1)/2 bivariate dependencee parameters. In: Distributions with fixed marginal and related topics, lecture notes-monograph series, vol 28. Institute of Mathematical Statistics, pp 120–141
Kramer MA (1991) Nonlinear principal component analysis using autoassociative neural networks. AIChE J 37(2):233–243. doi:10.1002/aic.690370209
Kurowicka D, Cooke R (2006) Uncertainty analysis with high dimensional dependence modelling. Wiley, New York, NY, doi:10.1002/0470863072
Lark RM (2002) Optimized spatial sampling of soil for estimation of the variogram by maximum likelihood. Geoderma 105(1–2):49–80. doi:10.1016/S0016-7061(01)00092-1
Leuangthong O, Deutsch CV (2003) Stepwise conditional transformation for simulation of multiple variables. Math Geol 35(2):155–173. doi:10.1023/A:1023235505120
Li J, Zimmerman DL (2015) Model-based sampling design for multivariate geostatistics. Technometrics 57(1):75–86. doi:10.1080/00401706.2013.873003
Li J, Bárdossy A, Guenni L, Liu M (2011) A copula based observation network design approach. Environ Modell Softw 26(11):1349–1357. doi:10.1016/j.envsoft.2011.05.001
Lischeid G (2009) Non-linear visualization and analysis of large water quality data sets: a model-free basis for efficient monitoring and risk assessment. Stoch Environ Res Risk Assess 23(7):977–990. doi:10.1007/s00477-008-0266-y
Melles SJ, Heuvelink GBM, Twenhöfel CJW, van Dijk A, Hiemstra PH, Baume O, Stöhlker U (2011) Optimizing the spatial pattern of networks for monitoring radioactive releases. Comput Geosci 37(3):280–288. doi:10.1016/j.cageo.2010.04.007
Moon CJ, Whateley MKG (2006) From prospect to prefeasibility. In: Moon CJ, Whateley MKG, Evans AM (eds) Introduction to mineral exploration, 2nd edn. Wiley-Blackwell, New York, pp 70–103
Musafer GN, Thompson MH (2016) Optimal adaptive sequential sampling of soil using pair-copulas. Geoderma 271:124–133
Musafer GN, Thompson MH, Wolff RC, Kozan E (2015) Non-linear multivariate spatial modelling using NLPCA and pair-copulas. Technical report, Queensland University of Technology
Peyrard N, Sabbadin R, Spring D, Brook B, Mac Nally R (2011) Model-based adaptive spatial sampling for occurrence map construction. Stat Comput 23(1):29–42. doi:10.1007/s11222-011-9287-3
Quessy J, Rivest L, Toupin M (2014) Semi-parametric pariwise inference methods in spatial models based on copulas. Spatial Stat 14(Part C):472–490. doi:10.1016/j.spasta.2015.08.002
R Core Team (2016) R: a language and environment for statistical computing. R foundation for statistical computing, Vienna. http://www.R-project.org/
Ritchie MW, Zhang J, Hamilton TA (2013) Aboveground tree biomass for Pinus ponderosa in Northeastern California. Forests 4(1):179–196. doi:10.3390/f4010179
Shapiro A (2003) Monte Carlo sampling methods. In: Ruszczyriski A, Shapiro A (eds) Stochastic programming, handbooks in operations research and management science, vol 10, Elsevier, Amsterdam, pp 353–425. doi:10.1016/S0927-0507(03)10006-0
Spöck G, Pilz J (2010) Spatial sampling design and covariance-robust minimax prediction based on convex design ideas. Stoch Environ Res Risk Assess 24(3):463–482. doi:10.1007/s00477-009-0334-y
Vašát R, Heuvelink GBM, Borůvka L (2010) Sampling design optimization for multivariate soil mapping. Geoderma 155(3–4):147–153. doi:10.1016/j.geoderma.2009.07.005
Wackernagel H (2003) Multivariate geostatistics, 3rd edn. Springer, London
Wang J, Stein A, Gao B, Ge Y (2012) A review of spatial sampling. Spat Stat 2:1–14. doi:10.1016/j.spasta.2012.08.001
Webster R, Oliver MA (1992) Sample adequately to estimate variograms of soil properties. J Soil Sci 43(1):177–192. doi:10.1111/j.1365-2389.1992.tb00128.x
Zhou X, Hemstrom MA (2009) Estimating aboveground tree biomass on forest land in the Pacific Northwest: a comparison of approaches. Research paper PNW-RP-584, United States Department of Agriculture Forest Service, Pacific Southwest Research Station. http://www.fs.fed.us/pnw/pubs/pnw_rp584.pdf
Zhu Z, Stein ML (2006) Spatial sampling design for prediction with estimated parameters. J Agric Biol Environ Stat 11(1):24–44. doi:10.1198/108571106X99751
Zimmerman DL (2006) Optimal network design for spatial prediction, covariance parameter estimation, and empirical prediction. Environmetrics 17(6):635–652. doi:10.1002/env.769
Acknowledgments
The authors would like to acknowledge the support of the Australian Government’s Cooperative Research Centre for Optimising Resource Extraction (Grant P3C-030). The authors would also like to acknowledge the United States Department of Agriculture Forest Service for assistance with the Bartlett Experimental Forest data. Thanks also to the reviewers, whose comments helped improve the introduction and to further clarify the model dependent nature of the designs presented.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Musafer, G.N., Thompson, M.H. Non-linear optimal multivariate spatial design using spatial vine copulas. Stoch Environ Res Risk Assess 31, 551–570 (2017). https://doi.org/10.1007/s00477-016-1307-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00477-016-1307-6