Kriging with external drift for functional data for air quality monitoring

  • Rosaria IgnaccoloEmail author
  • Jorge Mateu
  • Ramon Giraldo


Functional data featured by a spatial dependence structure occur in many environmental sciences when curves are observed, for example, along time or along depth. Recently, some methods allowing for the prediction of a curve at an unmonitored site have been developed. However, the existing methods do not allow to include in a model exogenous variables that, for example, bring meteorology information in modeling air pollutant concentrations. In order to introduce exogenous variables, potentially observed as curves as well, we propose to extend the so-called kriging with external drift—or regression kriging—to the case of functional data by means of a three-step procedure involving functional modeling for the trend and spatial interpolation of functional residuals. A cross-validation analysis allows to choose smoothing parameters and a preferable kriging predictor for the functional residuals. Our case study considers daily PM10 concentrations measured from October 2005 to March 2006 by the monitoring network of Piemonte region (Italy), with the trend defined by meteorological time-varying covariates and orographical constant-in-time variables. The performance of the proposed methodology is evaluated by predicting PM10 concentration curves on 10 validation sites, even with simulated realistic datasets on a larger number of spatial sites. In this application the proposed methodology represents an alternative to spatio-temporal modeling but it can be applied more generally to spatially dependent functional data whose domain is not a time interval.


Functional data modeling Functional linear model Residual kriging Particulate matter Spatial dependence 



The authors are grateful to the Associate Editor and two anonymous referees whose comments and suggestions improved the reading and quality of the manuscript. Rosaria Ignaccolo’s work was partially supported by Regione Piemonte and FIRB 2012 Grant (project no. RBFR12URQJ) provided by the Italian Ministry of Education, Universities and Research. This research was also supported in part by the Spanish Ministry of Education and Science and Bancaja through Grants MTM2010-14961 and P1-1B2012-52.


  1. Caballero W, Giraldo R, Mateu J (2013) A universal kriging approach for spatial functional data. Stoch Environ Res Risk Assess doi: 10.1007/s00477-013-0691-4.
  2. Cameletti M, Ignaccolo R, Bande S (2011) Comparing spatio-temporal models for particulate matter in Piemonte. Environmetrics 22:985–996CrossRefGoogle Scholar
  3. Cameletti M, Lindgren F, Simpson D, Rue H (2012) Spatio-temporal modeling of particulate matter concentration through the SPDE approach. AStA Adv Stat Anal 97(2):109–131CrossRefGoogle Scholar
  4. Crujeiras RM, Van Keilegom I (2010) Least squares estimation of nonlinear spatial trends. Comput Stat Data Anal 54:452–465CrossRefGoogle Scholar
  5. Delicado P, Giraldo R, Comas, C, Mateu J (2010) Statistics for spatial functional data: some recent contributions. Environmentrics 21:224–239CrossRefGoogle Scholar
  6. EEA (2012) Air quality in Europe 2012 report. Report No 4/2012. European Environment Agency, Copenhagen.Google Scholar
  7. Ferraty F, Laksaci A, Tadj A, Vieu P (2011) Kernel regression with functional response. Electron J Stat 5:150–171CrossRefGoogle Scholar
  8. Ferraty F, Vieu P (2006) Nonparametric functional data analysis: theory and practice. Springer, New York.Google Scholar
  9. Finardi S, DeMaria R, D’Allura A, Cascone C, Calori G, Lollobrigida F (2008) A deterministic air quality forecasting system for Torino urban area, Italy. Environ Model Softw 23(3):344–355CrossRefGoogle Scholar
  10. Giraldo R, Delicado P, Mateu J (2009) Geostatistics with infinite dimensional data: a generalization of cokriging and multivariable spatial prediction. Reporte Interno de Investigacion No. 14, Universidad Nacional de ColombiaGoogle Scholar
  11. Giraldo R, Delicado P, Mateu J (2010) Continuous time-varying kriging for spatial prediction of functional data: an environmental application. J Agric Biol Environ Stat 15(1):66–82CrossRefGoogle Scholar
  12. Giraldo R, Mateu J, Delicado P (2012) geofd: an R package for function-valued geostatistical prediction. Revista Colombiana de Estadstica 35(3):383–405Google Scholar
  13. Giraldo R, Delicado P, Mateu J (2011) Ordinary kriging for function-valued spatial data. Environ Ecol Stat 18(3):411–426CrossRefGoogle Scholar
  14. Gromenko O, Kokoszka P (2013) Nonparametric inference in small data sets of spatially indexed curves with application to ionospheric trend determination. Comput Stat Data Anal 59:82–94CrossRefGoogle Scholar
  15. Hengl T, Heuvelink GBM, Rossiter DG (2007) About regression-kriging: from equations to case studies. Comput Geosci 33(10):1301–1315CrossRefGoogle Scholar
  16. Horváth L, Kokoszka P (2012) Inference for functional data with applications. Springer, New York.Google Scholar
  17. Ivanescu AE, Staicu AM, Greven S, Scheipl F, Crainiceanu CM (2012) Penalized function-on-function regression (April 2012). Dept. of Biostatistics Working Papers, Johns Hopkins University, Working Paper 240. Available at
  18. Kokoszka P (2012) Dependent functional data. ISRN Probab Stat 1–30. doi: 10.5402/2012/958254.
  19. Lahiri SN, Leea Y, Cressie N (2002) On asymptotic distribution and asymptotic efficiency of least squares estimators of spatial variogram parameters. J Stat Plan Inference 103:65–85CrossRefGoogle Scholar
  20. Molnar S (1985) On the convergence of the kriging method. Annales Univ Sci Budapest Sect Comput 6:81–90Google Scholar
  21. Nerini D, Monestiez P, Mantè C (2010) Cokriging for spatial functional data. J Multivar Anal 101:409–418CrossRefGoogle Scholar
  22. Pebesma EJ (2004) Multivariable geostatistics in S: the gstat package. Comput Geosci 30:683–691CrossRefGoogle Scholar
  23. Poli A, Cirillo M (1993) On the use of the normalized mean square error in evaluating dispersion model performance. Atmos Environ 27:2427–2434CrossRefGoogle Scholar
  24. R Development Core Team (2012) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna.
  25. Ramsay JO, Silverman BW (2002) Applied functional data analysis. Springer, New York.Google Scholar
  26. Ramsay JO, Silverman BW (2005) Functional data analysis. Springer, New York.Google Scholar
  27. Ramsay JO, Hooker G, Graves S (2009) Functional data analysis with R and Matlab. Springer, New York.Google Scholar
  28. Ramsay JO, Wickham H, Graves S, Hooker G (2012) FDA: functional data analysis, R package version 2.3.2.Google Scholar
  29. Ruiz-Medina MD, Fernández-Pascual R (2010) Spatiotemporal filtering from fractal spatial functional data sequence. Stoch Environ Res Risk Assess 24:527–538CrossRefGoogle Scholar
  30. Ruiz-Medina MD, Salmerón R (2010) Functional maximum-likelihood estimation of arh(p) models. Stoch Environ Res Risk Assess 24:131146.Google Scholar
  31. Ruiz-Medina MD (2012) New challenges in spatial and spatiotemporal functional statistics for high-dimensional data. Spat Stat 1:82–91CrossRefGoogle Scholar
  32. Sakata S, Ashida F, Tanaka H (2010) Stabilization of parameter estimation for kriging-based approximation with empirical semivariogram. Comput Methods Appl Mech Eng 199:1710–1721CrossRefGoogle Scholar
  33. Salmerón R, Ruiz-Medina MD (2009) Multi-spectral decomposition of functional autoregressive models. Stoch Environ Res Risk Assess 23(3):289–297CrossRefGoogle Scholar
  34. Temiyasathit C, Kim SB, Park SK (2009) Spatial prediction of ozone concentration profiles. Comput Stat Data Anal 53:3892–3906CrossRefGoogle Scholar
  35. Vazquez E, Bect J (2010) Pointwise consistency of the kriging predictor with known mean and covariance functions. In mODa 9 advances in model-oriented design and analysis. Contributions to statistics, pp 221–228.Google Scholar
  36. Ver Hoef J, Cressie N (1993) Multivariable spatial prediction. Math Geol 25(2):219–240CrossRefGoogle Scholar
  37. Wackernagel H (1995) Multivariable geostatistics: an introduction with applications. Springer, Berlin.Google Scholar
  38. Wood SN (2004) Stable and efficient multiple smoothing parameter estimation for generalized additive models. J Am Stat Assoc 99:673–686CrossRefGoogle Scholar
  39. Wood SN (2011) Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J R Stat Soc B 73(1):3–36CrossRefGoogle Scholar
  40. Wood SN (2012a) mgcv: mixed GAM computation vehicle with GCV/AIC/REML smoothness estimation, R package version 1.7-22Google Scholar
  41. Wood SN (2012b) On p-values for smooth components of an extended generalized additive model. Biometrika 1–8. doi: 10.1093/biomet/ass048.
  42. Yu S, Eder B, Dennis R, Chu S, Schwartz S (2006) New unbiased symmetric metrics for evaluation of air quality models. Atmos Sci Lett 7:26–34CrossRefGoogle Scholar
  43. Zhang X, Zheng Y (2012) A note on spatial-temporal lattice modeling and maximum likelihood estimation. Stat Probab Lett 82:2145–2155CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  1. 1.Università degli Studi di TorinoTurinItaly
  2. 2.Universitat Jaume ICastelloSpain
  3. 3.Universidad Nacional de ColombiaBogotaColombia

Personalised recommendations