Skip to main content
ARS Home » Research » Publications at this Location » Publication #215283

Title: Statistical models for the prediction of field scale, spatial salinity patterns from soil conductivity survey data

Author
item LESCH, SCOTT - UC, RIVERSIDE

Submitted to: Book Chapter
Publication Type: Book / Chapter
Publication Acceptance Date: 12/31/2007
Publication Date: 1/2/2012
Citation: Lesch, S.M. 2012. Statistical models for the prediction of field scale, spatial salinity patterns from soil conductivity survey data. In: Wallender, W.W. and Tanji, K.K. (eds.) ASCE Manual and Report on Engineering Practice No. 71 Agricultural Salinity Assessment and Management. 2nd Edition. ASCE, Reston, VA. p. 461-482.

Interpretive Summary: A practical, regression based methodology for the prediction of field scale, spatial salinity patterns from soil conductivity survey data is presented in this chapter. The connection between an ordinary linear regression model and the geostatistical mixed linear model is reviewed, along with linear regression model parameter estimates and survey predictions. These regression based techniques are applicable to a wide variety of spatial calibration problems encountered in soil salinity survey work and precision farming applications. Two case studies are presented that highlight and demonstrate how one can perform and interpret the various modeling techniques discussed throughout the chapter.

Technical Abstract: A practical linear regression based methodology for the prediction of field scale, spatial salinity patterns from soil conductivity survey data is presented in this chapter. The connection between an ordinary linear regression model and the geostatistical mixed linear model is reviewed, along with the concepts of best linear unbiased estimation and prediction. The Moran test for detecting spatial correlation in linear regression model residuals is also discussed, in addition to three useful regression model validation tests. Two case studies are presented that highlight and demonstrate the various parameter estimation, salinity prediction, and model validation techniques discussed throughout the chapter.