Abstract
In many studies, the distribution of soil attributes depends on both spatial location and environmental factors, and prediction and process identification are performed using existing methods such as kriging. However, it is often too restrictive to model soil attributes as dependent on a known, parametric function of environmental factors, which kriging typically assumes. This paper investigates a semiparametric approach for identifying and modeling the nonlinear relationships of spatially dependent soil constituent levels with environmental variables and obtaining point and interval predictions over a spatial region. Frequentist and Bayesian versions of the proposed method are applied to measured soil nitrogen levels throughout Florida, USA and are compared to competing models, including frequentist and Bayesian kriging, based an array of point and interval measures of out-of-sample forecast quality. The semiparametric models outperformed competing models in all cases. Bayesian semiparametric models yielded the best predictive results and provided empirical coverage probability nearly equal to nominal.
Similar content being viewed by others
References
Banerjee S, Carlin B, Gelfand A (2003) Hierarchical modeling and analysis for spatial data. Chapman & Hall/CRC, Boca Raton
Bliss CM, Comerford NB, Graetz DA, Grunwald S, Stoppe AM (2014) Land use influence on carbon, nitrogen, and phosphorus in size fractions of sandy surface soils. Soil Sci J 178:654–661
Bliznyuk N, Ruppert D, Shoemaker C, Regis R, Wild S, Mugunthan P (2008) Bayesian calibration and uncertainty analysis for computationally expensive models using optimization and radial basis function approximation. J Comput Gr Stat 17(2):270–294
Bliznyuk N, Ruppert D, Shoemaker CA (2011) Efficient interpolation of computationally expensive posterior densities with variable parameter costs. J Comput Gr Stat 20(3):636–655
Bliznyuk N, Ruppert D, Shoemaker CA (2012) Local derivative-free approximation of computationally expensive posterior densities. J Comput Gr Stat 21(2):476–495
Bliznyuk N, Paciorek CJ, Schwartz J, Coull B (2014) Nonlinear predictive latent process models for integrating spatio-temporal exposure data from multiple sources. Ann Appl Stat 8(3):1538–1560
Buhmann MD (2003) Radial basis functions: theory and implementations. Cambridge Monogr Appl Comput Math 12:147–165
Crainiceanu CM, Ruppert D, Wand MP (2005) Bayesian analysis for penalized spline regression using WinBUGS. J Stat Softw 14(14):1–24
Ding X, Zhou H, Lei X et al (2013) Hydrological and associated pollution load simulation and estimation for the Three Gorges Reservoir of China. Stoch Environ Res Risk Assess. 27:617. doi:10.1007/s00477-012-0627-4
Florida Fish and Wildlife Conservation Commission (FFWCC) (2003) Land cover/land use map of Florida derived from Landsat satellite imagery. 30 meter resolution. http://www.fgdl.org
Grunwald S, Thompson JA, Boettinger JL (2011) Digital soil mapping and modeling at continental scales—finding solutions for global issues. Soil Sci Soc Am J 75(4):1201–1213
Guan Y, Shao C, Gu Q et al (2016) Study of a comprehensive assessment method of the environmental quality of soil in industrial and mining gathering areas. Stoch Environ Res Risk Assess. 30:91
Hastie T, Tibshirani R, Friedman J (2009) The elements of statistical learning, 2nd ed. Springer, New York
Kasiviswanathan KS, Sudheer KP (2013) Quantification of the predictive uncertainty of artificial neural network based river flow forecast models. Stoch Environ Res Risk Assess. 27:137. doi:10.1007/s00477-012-0600-2
Kaufman L, Rousseeuw PJ (2005) Finding groups in data: an introduction to cluster analysis. Wiley, New York
Kim J, Grunwald S, Rivero RG (2014) Soil phosphorus and nitrogen predictions across spatial escalating scales in an aquatic ecosystem using remote sensing images. IEEE Trans Geosci Remote Sens J 52(10):6724–6737
Nash JE, Sutcliffe JV (1970) River flow forecasting through conceptual models part I—a discussion of principles. J Hydrol 10(3):282–290
Natural Resources Conservation Service (NRCS) (2006) Soil Data Mart: Soil Survey Geographic Database, SSURGO. Map scale 1:24000. http://www.nrcs.usda.gov/wps/ portal/nrcs/detail/soils/survey/?cid = nrcs142p2_053627
Plummer M (2003) JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling. http://mcmc-jags.sourceforge.net
Rabalais NN, Turner RE, Wiseman WJ (2002) Gulf of Mexico Hypoxia, A.K.A. “The Dead Zone”. Ann Rev Ecol Syst 33:235–263
R Core Team (2014) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. http://www.R-project.org/.Rabalais
Ruppert D, Wand MP, Carroll RJ (2003) Semiparametric regression. Cambridge University Press, New York
Seber GAF, Lee AJ (2003) Linear regression analysis. Wiley and Sons, Hoboken
Shmueli G (2010) To explain or to predict? Stat Sci 25(3):289–310
Su YS, Yajima M (2014) R2jags: A package for running jags from R. http://CRAN.R-project.org/package=R2jags
Sun Y, Bowman KP, Genton MG, Tokay A (2015) A Matern model of the spatial covariance structure of point rain rates. Stoch Env Res Risk Assess 29:411–416
Turner RE, Rabalais NN, Justic D (2006) Predicting summer hypoxia in the northern Gulf of Mexico: Riverine N, P, and Si loading. Mar Pollut Bull 52:139–148
Vasques GM, Grunwald S, Myers DB (2012) Associations between soil carbon and ecological landscape variables at escalating spatial scales in Florida, USA. Landsc Ecol J. 27:355–367. doi:10.1007/s10980-011-9702-3
Weese DJ, Heath KD, Dentinger M, Lau JA (2015) Long-term nitrogen addition causes the evolution of less-cooperative mutualists. Int J Org Evol. doi:10.1111/evo.12594
Wood SN (2004) Stable and efficient multiple smoothing parameter estimation for generalized additive models. J Am Stat Assoc 99:673–686
Wood SN (2006) Generalized additive models: an introduction with R. Chapman & Hall/CRC, Boca Raton
Wood SN (2011) Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J Royal Stat Soc (B) 73(1):3–36
Wood SN, Goude Y, Shaw S (2015) Generalized additive models for large data sets. J Royal Stat Soc (C) 64(1):139–155
Xiong X, Grunwald S, Myers DB, Kim J, Harris WG, Comerford NB (2014a) Holistic environmental soil-landscape modeling of soil organic carbon. Environ Model Software J. 57:202–215
Xiong X, Grunwald S, Myers DB, Ross CW, Harris WG, Comerford NB (2014b) Interaction effects of climate and land use/land cover change on soil organic carbon sequestration. Sci Total Environ J. 493:974–982. doi:10.1016/j.scitotenv.2014.06.088
Xiong X, Grunwald S, Myers DB, Kim J, Harris WG, Bliznyuk N (2015) Assessing uncertainty in soil organic carbon modeling across a highly heterogeneous landscape. Geoderma 251:105–116
Yang Y, Christakos G (2015) Uncertainty assessment of heavy metal soil contamination mapping using spatiotemporal sequential indicator simulation with multi-temporal sampling points. Environ Monit Assess 187(9):1–15. doi:10.1007/s10661-015-4785-y
Ye M, Meyer PD, Lin Y et al (2010) Quantification of model uncertainty in environmental modeling. Stoch Environ Res Risk Assess. 24:807. doi:10.1007/s00477-010-0377-0
Acknowledgements
Bliznyuk’s effort was partially supported by National Institutes of Health grants U54GM111274 and R21AI119773. Data collection was funded by USDA-CSREES-NRI grant award 2007-35107-18368 ‘Rapid Assessment and Trajectory Modeling of Changes in Soil Carbon across a Southeastern Landscape’ (National Institute of Food and Agriculture (NIFA)—Agriculture and Food Research Initiative (AFRI)). This project is a Core Project of the North American Carbon Program. The authors would like to thank Aja Stoppe, Brenton D. Myers, Christopher Wade Ross, Elena Azuaje, Samiah Moustafa, Lisa Stanley, Adriana Comerford, and Anne Quidez for their hard work in field soil sampling and laboratory analyses. In addition, we like to thank Xiong Xiong, Christopher Wade Ross, Gustavo M. Vasques, and Brenton D. Myers for the preparation of geospatial environmental data using GIS methods. Other thanks go to: Dr. N.B. Comerford and Dr. W.G. Harris.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Merrill, H.R., Grunwald, S. & Bliznyuk, N. Semiparametric regression models for spatial prediction and uncertainty quantification of soil attributes. Stoch Environ Res Risk Assess 31, 2691–2703 (2017). https://doi.org/10.1007/s00477-016-1337-0
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00477-016-1337-0