The ConMap approach for terrain-based digital soil mapping

We present a new digital terrain analysis framework for digital soil mapping, referred to as contextual elevation mapping (ConMap). In contrast to common regression approaches based on features from digital terrain analysis, ConMap is not based on standard terrain attributes, but on elevation differences from the centre pixel to each pixel in circular neighbourhoods only. These differences are used as features in random forest regressions. We applied and validated the framework by predicting topsoil silt content in a loess region of 1150 km2 in Rhineland-Palatinate and Hesse, Germany. Three hundred and forty-two samples and a 20-m resolution digital elevation model were used for this illustration and validation. We compared ConMap with standard and multi-scale terrain analysis approaches as well as with ordinary kriging interpolations. Cross-validation root mean square error (RMSE) decreased from 16.1 when the standard digital terrain analysis was used to 11.2 when ConMap was used. This corresponds to an increase in variance explained (R2) from 15 to 61%. Even though ordinary kriging out-performed standard terrain analysis as well, the variance explained was 6% smaller compared with that using ConMap. The results show that the geomorphic settings in the study area must have induced the spatial trend, which can be accounted for by ConMap over different scales. We conclude that ConMap shows great potential for digital soil mapping studies.

[1]  Philippe Lagacherie,et al.  Digital Soil Mapping: A State of the Art , 2008 .

[2]  James E. Burt,et al.  The Impact of Neighbourhood Size on Terrain Derivatives and Digital Soil Mapping , 2008 .

[3]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[4]  Thorsten Behrens,et al.  Digital soil mapping using artificial neural networks , 2005 .

[5]  H. Jenny,et al.  Factors of Soil Formation , 1941 .

[6]  Timothy C. Coburn,et al.  Geostatistics for Natural Resources Evaluation , 2000, Technometrics.

[7]  A-Xing Zhu,et al.  Multi-scale digital terrain analysis and feature selection for digital soil mapping , 2010 .

[8]  Pierre Goovaerts,et al.  Predicting Ordinary Kriging Errors Caused by Surface Roughness and Dissectivity , 2005 .

[9]  Achim Zeileis,et al.  BMC Bioinformatics BioMed Central Methodology article Conditional variable importance for random forests , 2008 .

[10]  Achim Zeileis,et al.  Bias in random forest variable importance measures: Illustrations, sources and a solution , 2007, BMC Bioinformatics.

[11]  Elisabeth N. Bui,et al.  Spatial data mining for enhanced soil map modelling , 2002, Int. J. Geogr. Inf. Sci..

[12]  H. Jenny.  Factors of Soil Formation: A System of Quantitative Pedology , 2011 .

[13]  Thorsten Behrens,et al.  Digital soil mapping in Germany—a review , 2006 .

[14]  Bernard De Baets,et al.  Random forests as a tool for predictive ecohydrological modelling , 2005 .

[15]  Tin Kam Ho,et al.  The Random Subspace Method for Constructing Decision Forests , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[16]  J. Peters,et al.  Random forests as a tool for ecohydrological distribution modelling , 2007 .

[17]  D. R. Cutler,et al.  Utah State University From the SelectedWorks of , 2017 .

[18]  A. Papritz,et al.  An Empirical Comparison of Kriging Methods for Nonlinear Spatial Point Prediction , 2002 .

[19]  R. Tibshirani,et al.  An introduction to the bootstrap , 1993 .

[20]  A. J. Gerrard.  Soils and landforms: An integration of geomorphology and pedology , 1982 .

[21]  Edzer J. Pebesma,et al.  Multivariable geostatistics in S: the gstat package , 2004, Comput. Geosci..

[22]  Wei-Yin Loh,et al.  Classification and regression trees , 2011, WIREs Data Mining Knowl. Discov..

[23]  Ron Begleiter,et al.  A Generic Tool for Performance Evaluation of Supervised Learning Algorithms , 2008 .

[24]  Dominique Arrouays,et al.  Extrapolating regional soil landscapes from an existing soil map: Sampling intensity, validation procedures, and integration of spatial context , 2008 .

[25]  A-Xing Zhu,et al.  Soil Mapping Using GIS, Expert Knowledge, and Fuzzy Logic , 2001 .

[26]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[27]  C. Thorne,et al.  Quantitative analysis of land surface topography , 1987 .

[28]  David W. Opitz,et al.  An Empirical Evaluation of Bagging and Boosting , 1997, AAAI/IAAI.

[29]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[30]  H. Elsenbeer,et al.  Soil organic carbon concentrations and stocks on Barro Colorado Island — Digital soil mapping using Random Forests analysis , 2008 .

[31]  Budiman Minasny,et al.  Chapter 21 Soil Prediction with Spatially Decomposed Environmental Factors , 2006 .

[32]  Alex B. McBratney,et al.  Spatial prediction of soil properties from landform attributes derived from a digital elevation model , 1994 .

[33]  I. Evans.  Statistical Characterization of Altitude Matrices by Computer. Report 6. An Integrated System of Terrain Analysis and Slope Mapping. , 1979 .

[34]  Guoan Tang,et al.  Advances in Digital Terrain Analysis , 2008 .

[35]  A. Stewart Fotheringham,et al.  Geographically Weighted Regression: A Method for Exploring Spatial Nonstationarity , 2010 .

[36]  A. Prasad,et al.  Newer Classification and Regression Tree Techniques: Bagging and Random Forests for Ecological Prediction , 2006, Ecosystems.

[37]  P. A. Shary,et al.  Fundamental quantitative methods of land surface analysis , 2002 .

[38]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[39]  M. Hutchinson,et al.  Digital terrain analysis. , 2008 .

[40]  Budiman Minasny,et al.  On digital soil mapping , 2003 .

[41]  Thorsten Behrens,et al.  Instance selection and classification tree analysis for large spatial datasets in digital soil mapping , 2008 .

[42]  A-Xing Zhu,et al.  The effects of DEM resolution and neighborhood size on digital soil survey , 2006 .

[43]  Francis D. Hole,et al.  An approach to landscape analysis with emphasis on soils , 1978 .