Improving the spatial prediction of soil organic carbon stocks in a complex tropical mountain landscape by methodological specifications in machine learning approaches

PLoS ONE

Volumn 11, Issue 4, 2016, Pages

  Ließ, Mareike ^a,b   Schmidt, Johannes ^c   Glaser, Bruno ^d  

^a HELMHOLTZ CENTRE FOR ENVIRONMENTAL RESEARCH UFZ   (Germany)

^b UNIVERSITY OF BAYREUTH   (Germany)

^c UNIVERSITY OF LEIPZIG   (Germany)

^d MARTIN LUTHER UNIVERSITY HALLE WITTENBERG   (Germany)

Author keywords

[No Author keywords available]

Indexed keywords

ORGANIC CARBON; SOIL CARBON STOCK; UNCLASSIFIED DRUG; CARBON; SOIL;

ARTICLE; ARTIFICIAL NEURAL NETWORK; BOOSTED REGRESSION TREE; CARBON CYCLE; CHEMICAL ANALYSIS; CONCENTRATION (PARAMETERS); CONTROLLED STUDY; ENVIRONMENTAL MONITORING; GEOGRAPHIC DISTRIBUTION; GEOGRAPHIC INFORMATION SYSTEM; INFORMATION PROCESSING; INTERMETHOD COMPARISON; LANDSCAPE; MACHINE LEARNING; MULTIVARIATE ADAPTIVE REGRESSION SPLINE; PREDICTION; RANDOM FOREST; SOIL ANALYSIS; SUPPORT VECTOR MACHINE; BIOLOGICAL MODEL; CARBON SEQUESTRATION; CHEMISTRY; ECUADOR; RAIN FOREST; REGRESSION ANALYSIS; SOIL;

CARBON; CARBON CYCLE; CARBON SEQUESTRATION; ECUADOR; MACHINE LEARNING; MODELS, BIOLOGICAL; RAINFOREST; REGRESSION ANALYSIS; SOIL; SUPPORT VECTOR MACHINE;

EID: 84966269208     PISSN: None     EISSN: 19326203     Source Type: Journal    
DOI: 10.1371/journal.pone.0153673     Document Type: Article

References (108)

1. Cerri CEP, Easter M, Paustian K, Killian K, Coleman K, Bernoux M, et al. Predicted soil organic carbon stocks and changes in the Brazilian Amazon between 2000 and 2030. Agric Ecosyst Environ. 2007; 122: 58-72. doi: 10.1016/j.agee.2007.01.008
2. Pan Y, Birdsey R, Fang J, Houghton R, Kauppi PE, Kurz W, et al. A large and persistent carbon sink in the world's forests. Science. 2011; 333: 988-993. doi: 10.1126/science.1201609 PMID: 21764754
3. Post WM, Kwon KC. Soil carbon sequestration and land-use change: processes and potential. Glob Chang Biol. 2000; 6: 317-327. doi: 10.1046/j.1365-2486.2000.00308.x
4. Don A, Schumacher J, Freibauer A. Impact of tropical land-use change on soil organic carbon stocks - a meta-analysis. Glob Chang Biol. 2011; 17: 1658-1670. doi: 10.1111/j.1365-2486.2010.02336.x
5. Watson RT. Land Use, Land-Use Change, and Forestry [Internet]. 2000. p. 392. Available: http://books.google.com/books?id=EyXQW-gRey0C
6. FAO. State of the World's Forests. 2014.
7. Tapia-Armijos MF, Homeier J, Espinosa CI, Leuschner C, De La Cruz M. Deforestation and forest fragmentation in south Ecuador since the 1970s - Losing a hotspot of biodiversity. PLoS One. 2015; 10: 1-18. doi: 10.1371/journal.pone.0133701
8. Beck E, Hartig K, Roos K. Forest clearing by slash and burn. In: Beck E.; Bendix J.; Kottke I.; Makeschi F.; Mosandl R, editor. Gradients in a tropical mountain ecosystem of Ecuador Ecological Studies 198. Heidelberg: Springer; 2008.
9. Bahr E, Chamba Zaragocin D, Makeschin F. Soil nutrient stock dynamics and land-use management of annuals, perennials and pastures after slash-and-burn in the Southern Ecuadorian Andes. Agric Ecosyst Environ. Elsevier B.V.; 2014; 188: 275-288. doi: 10.1016/j.agee.2014.03.005
10. Ließ M, Hitziger M, Huwe B. The sloping mire soil-landscape of southern ecuador: Influence of predictor resolution and model tuning on random forest predictions. Appl Environ Soil Sci. 2014; 2014. doi: 10.1155/2014/603132
11. Jenny H. Factors of Soil Formation. A System of Quantitative Pedology. New York: Dover Publications, Inc.; 1941.
12. Ryan PJ, McKenzie NJ, O'Connell D, Loughhead AN, Leppert PM, Jacquier D, et al. Integrating forest soils information across scales: Spatial prediction of soil properties under Australian forests. For Ecol Manage. 2000; 138: 139-157. doi: 10.1016/S0378-1127(00)00393-5
13. Ließ M. Sampling for regression-based digital soil mapping: Closing the gap between statistical desires and operational applicability. Spat Stat. Elsevier Ltd; 2015; 13: 106-122. doi: 10.1016/j.spasta.2015.06.002
14. Cavazzi S, Corstanje R, Mayr T, Hannam J, Fealy R. Are fine resolution digital elevation models always the best choice in digital soil mapping? Geoderma. Elsevier B.V.; 2013; 195-196: 111-121. doi: 10.1016/j.geoderma.2012.11.020
15. Behrens T, Zhu AX, Schmidt K, Scholten T. Multi-scale digital terrain analysis and feature selection for digital soil mapping. Geoderma. Elsevier B.V.; 2010; 155: 175-185. doi: 10.1016/j.geoderma.2009.07.010
16. Maynard JJ, Johnson MG. Scale-dependency of LiDAR derived terrain attributes in quantitative soil-landscape modeling: Effects of grid resolution vs. neighborhood extent. Geoderma. Elsevier B.V.; 2014; 230-231: 29-40. doi: 10.1016/j.geoderma.2014.03.021
17. Samuel-Rosa A, Heuvelink GBM, Vasques GM, Anjos LHC. Do more detailed environmental covariates deliver more accurate soil maps? Geoderma. Elsevier B.V.; 2015; 243-244: 214-227. doi: 10.1016/j.geoderma.2014.12.017
18. Witten I, Frank E. Data mining-Practical machine learning tools and techniques. 2nd ed. Amsterdam: Elsevier; 2005.
19. Larose DT. Data mining methods and models. Hoboken: Wiley-Interscience/ John Wiley & Sons, Inc.; 2006.
20. Xu G, Zong Y, Yang Z. Applied data mining. Booca Raton: CRC Press; 2013.
21. John G, Kohavi R, Pfleger K. Irrelevant Features and the Subset Selection Problem. Proceedings of the Eleventh International Conference on Machine Learning. 1994. pp. 121-129.
22. Kohavi R, John H. Artificial Intelligence Wrappers for feature subset selection. 2011; 97: 273-324.
23. Kuhn M, Johnson K. Applied predictive modeling. New York: Springer Science; 2013. doi: 10.1007/978-1-4614-6849-319
24. Hastie T, Tibshirani R, Friedman J. The elements of statistical learning. 2nd ed. Springer; 2009. doi: 10.1007/978-0-387-98135-2
25. Grimm R, Behrens T, Märker M, Elsenbeer H. Soil organic carbon concentrations and stocks on Barro Colorado Island - Digital soil mapping using Random Forests analysis. Geoderma. 2008; 146: 102-113. doi: 10.1016/j.geoderma.2008.05.008
26. Guo PT, Li MF, Luo W, Tang QF, Liu ZW, Lin ZM. Digital mapping of soil organic matter for rubber plantation at regional scale: An application of random forest plus residuals kriging approach. Geoderma. Elsevier B.V.; 2015; 237-238: 49-59. doi: 10.1016/j.geoderma.2014.08.009
27. Gessler PE, Chadwick OA, Chamron F, Holmes K, Althouse L. Modeling soil-landscape and ecosystem properties using terrain attributes. Soil Sci Soc Am J. 2000; 2046-2056.
28. Gasparini P, Di Cosmo L. Forest carbon in Italian forests: Stocks, inherent variability and predictability using NFI data. For Ecol Manage. Elsevier B.V.; 2015; 337: 186-195. doi: 10.1016/j.foreco.2014.11.012
29. Zhang S, Huang Y, Shen C, Ye H, Du Y. Spatial prediction of soil organic matter using terrain indices and categorical variables as auxiliary information. Geoderma. Elsevier B.V.; 2012;171-172: 35-43. doi: 10.1016/j.geoderma.2011.07.012
30. Martin MP, Wattenbach M, Smith P, Meersmans J, Jolivet C, Boulonne L, et al. Spatial distribution of soil organic carbon stocks in France. Biogeosciences. 2011; 8: 1053-1065. doi: 10.5194/bg-8-1053-2011
31. Martin MP, Orton TG, Lacarce E, Meersmans J, Saby NPA, Paroissien JB, et al. Evaluation of modelling approaches for predicting the spatial distribution of soil organic carbon stocks at the national scale. Geoderma. Elsevier B.V.; 2014; 223-225: 97-107. doi: 10.1016/j.geoderma.2014.01.005
32. Pastick NJ, Rigge M, Wylie BK, Jorgenson MT, Rose JR, Johnson KD, et al. Distribution and landscape controls of organic layer thickness and carbon within the Alaskan Yukon River Basin. Geoderma. Elsevier B.V.; 2014; 230-231: 79-94. doi: 10.1016/j.geoderma.2014.04.008
33. Bou Kheir R, Bøcher PK, Greve MB, Greve MH. The application of GIS based decision-tree models for generating the spatial distribution of hydromorphic organic landscapes in relation to digital terrain data. Hydrol Earth Syst Sci. 2010; 14: 847-857. doi: 10.5194/hess-14-847-2010
34. Dai F, Zhou Q, Lv Z, Wang X, Liu G. Spatial prediction of soil organic matter content integrating artificial neural network and ordinary kriging in Tibetan Plateau. Ecol Indic. Elsevier Ltd; 2014; 45: 184-194. doi: 10.1016/j.ecolind.2014.04.003
35. Were K, Bui DT, Dick ØB, Singh BR. A comparative assessment of support vector regression, artificial neural networks, and random forests for predicting and mapping soil organic carbon stocks across an Afromontane landscape. Ecol Indic. Elsevier Ltd; 2015; 52: 394-403. doi: 10.1016/j.ecolind.2014.12.028
36. Basheer IA, Hajmeer M. Artificial neural networks: fundamentals, computing, design, and application. 2000; 43: 3-31. PMID: 11084225
37. Hsu C, Chang C, Lin C. A Practical Guide to Support Vector Classification. BJU Int. 2008; 101: 1396-400.
38. Cherkassky V, Ma Y. Practical selection of SVM parameters and noise estimation for SVM regression. Neural Networks. 2004; 17: 113-126. doi: 10.1016/S0893-6080(03)00169-2 PMID: 14690712
39. Hitziger M, Ließ M. Comparison of three supervised learning methods for digital soil mapping: Application to a complex terrain in the Ecuadorian Andes. Appl Environ Soil Sci. Hindawi Publishing Corporation; 2014; 2014.
40. Fries A, Rollenbeck R, Göttlicher D, Nauss T, Homeier J, Peters T, et al. Thermal structure of a megadiverse Andean mountain ecosystem in southern Ecuador. Erdkd - Arch Sci Geogr. 2009; 63: 321-335.
41. Rollenbeck R. Variability of precipitation in the Reserva Biólogica San Francisco / Southern Ecuador. Lyonia. 2006; 9: 43-51.
42. Litherland M, Aspen JA Jamielita R. The metamorphic belts of Ecuador. Overseas Mem Br Geol Surv. 1994; 11: 1-147.
43. Ließ M, Glaser B, Huwe B. Uncertainty in the spatial prediction of soil texture. Comparison of regression tree and Random Forest models. Geoderma. 2012; 170: 70-79.
44. Bauer FC. Water flow paths in soils of an undisturbed and landslide affected mature montane rainforest in South Ecuador. 2010; 150.
45. Ließ M, Glaser B, Huwe B. Digital soil mapping in Southern Ecuador. Erdkunde. 2009; 63: 309-319. doi: 10.3112/erdkunde.2009.04.02
46. Schmidt J. Statistische Modellierung der Kohlenstoff- und Stickstoffvorräte der organischen Auflage unter Bergregenwald in Südecuador. Universität Leipzig. 2012.
47. USGS. Using the USGS Landsat 8 product [Internet]. 2013. Available: http://landsat.usgs.gov/Landsat8-Using-Product.php.
48. Jackson TJ, Chen D, Cosh M, Li F, Anderson M, Walthall C, et al. Vegetation water content mapping using Landsat data derived normalized difference water index for corn and soybeans. Remote Sens Environ. 2004; 92: 475-482. doi: 10.1016/j.rse.2003.10.021
49. Richardson A, Wiegand C. Distinguishing vegetation from soil background information. Photogramm Eng Remote Sens. 1977; 43: 1541-1552.
50. Fox G, Sabbagh G, Searcy S, Yang C. An Automated Soil Line Identification Routine for Remotely Sensed Images. Soil Sci Soc Am J. 2004; 68: 1326. doi: 10.2136/sssaj2004.1326
51. Baret F, Guyot G. Potentials and limits of vegetation indices for LAI and APAR assessment. Remote Sens Environ. 1991; 35: 161-173. doi: 10.1016/0034-4257(91)90009-U
52. Baret F, Guyot G, Major DJ. TSAVI: A Vegetation Index Which Minimizes Soil Brightness Effects On LAI And APAR Estimation. 12th Can Symp Remote Sens Geosci Remote Sens Symp. 1989; 3: 1355-1358. doi: 10.1109/IGARSS.1989.576128
53. Brenning A. RSAGA: SAGA geoprocessing and terrain analysis in R [Internet]. 2015. Available: http://cran.r-project.org/web/packages/RSAGA/index.html.
54. Association SUG. SAGA-System for Automated Geoscientific Analysis [Internet]. 2014. Available: http://www.saga-gis.org.
55. Nauss T, Göttlicher D, Dobbermann M, Bendix J. Central data services in multidisciplinary environmental research projects. e-Zeitschrift für Agrarinformatik. e-Zeitschrift für Agrar. 2007; 2. Available: http://www.preagro.de/ezai/index.php/eZAI/article/view/28/28
56. Lyle WZ, Colin RT. Quantitative analysis of land surface topography. Earth Surf Process Landforms. 1987; 12: 47-56. doi: 10.1002/esp.3290120107
57. Friedrich K. Digitale Reliefgliederungsverfahren zur Ableitung bodenkundlich relevanter Flächeneinheiten. Frankfurt/M: Frankfurter Geowissenschaftliche Arbeiten D21; 1996.
58. Friedrich K. Multivariate distance methods of geomorphographic relief classification. In: Heinicke H, Eckelmannn W, Thomasson A, Jones J, Montarella L BB, editor. Land information systems-Developments for planning the sustainable use of land resources European Soil Bureau-Research Report 4. EUR 17729. Office for official publications of the European Communities, Ispra; 1998. pp. 259-266.
59. Möller M, Volk M, Friedrich K, Lymburner L. Placing soil-genesis and transport processes into a landscape context: A multiscale terrain-analysis approach. J Plant Nutr Soil Sci. 2008; 171: 419-430. doi: 10.1002/jpln.200625039
60. Rodriguez F. The Black Top Hat function applied to a DEM: A tool to estimate recent incision in a mountainous watershed (Estibère Watershed, Central Pyrenees). Geophys Res Lett. 2002; 29: 2-5. doi: 10.1029/2001GL014412
61. Böhner J, Antonic O. Land surface parameters specific to topo-climatology. In: Hengl T.; Reuter HI, editor. Geomorphometry-Concepts, Software, Applications Developments in Soil Science, Vol 33. Amsterdam: Elsevier; 2009. pp. 195-226.
62. Oke TR. Boundary layer climates. London: Taylor & Francis; 1987.
63. Wilson JP, Gallant JC, editors. Terrain analysis-Principles and Applications. New York: John Wiley & Sons, Inc.; 2000.
64. Häntzschel J, Goldberg V Bernhofer C. GIS-based regionalisation of radiation, temperature and coupling measures in complex terrain for low mountain ranges. Meteorol Appl. 2005; 12: 33-42.
65. Anders N, Seijmonsbergen A, Bouten W. Multi-Scale and Object-Oriented Image Analysis of High-Res LiDAR Data for Geomorphological Mapping in Alpine Mountains. Geomophometry 2009. 2009; 61-65.
66. Prima ODA, Echigo A, Yokoyama R, Yoshida T. Supervised landform classification of Northeast Honshu from DEM-derived thematic maps. Geomorphology. 2006; 78: 373-386. doi: 10.1016/j.geomorph.2006.02.005
67. Yokoyama R, Shlrasawa M, Pike RJ. Visualizing Topography by Openness: A New Application of Image Processing to Digital Elevation Models. Photogramm Eng Remote Sens. 2002; 68: 257-265.
68. Lea NL. An aspect driven kinematic routing algorithm. In: Parsons A.J.; Abrahams AD, editor. Overland flow: hydraulics and erosion mechanics. London: CRC Press; 1992.
69. Böhner J, Köthe R, Conrad O, Gross J, Ringeler A, Selige T. Soil regionalisation by means of terrain analysis and process parametrisation. In: Micheli E, Nachtergaele F ML, editor. Soil Classification 2001 European Soil Bureau, Research Report No 7, EUR 20398 EN. 2002.
70. Beven KJ, Kirkby MJ. Physically based, variable contibution area model of basin hydrology [Internet]. Hydrological Sciences Bulletin. 1979. pp. 43-69. doi: 10.1080/02626667909491834
71. Böhner J, Selige T. Spatial prediction of soil attributes using terrain analysis and climate regionalisation. SAGA - Anal Model Appl Göttinger Aeographische Abhandlungen. 2006; 115: 13-28.
72. Moore ID, Grayson RB, Ladson AR. Digital Terrain Modelling: a Review of Hydrological, Geomorphological, and Biological Applications. Rev Lit Arts Am. 1991; 5: 3-30.
73. Desmet P, Govers G. A GIS procedure for automatically calculating the USLE LS factor on topographically complex landscape units. J Soil Water Conserv. 1996; 51: 427-433.
74. Kinnell PIA. Alternative Approaches for Determining the Usle-M Slope Length Factor for Grid Cells. Soil Sci Soc Am J. 2005; 69: 674. doi: 10.2136/sssaj2004.0047
75. Köthe R, Lehmeier F. SARA-System zur automatischen Relief-Analyse. User Manual. unpublished; 1996.
76. Riley S, De Gloria S, Elliot R. A terrain ruggedness that quantifies topographic heterogeneity. Intermt J Sci. 1999; 5: 23-27.
77. Iwahashi J, Pike RJ. Automated classifications of topography from DEMs by an unsupervised nestedmeans algorithm and a three-part geometric signature. Geomorphology. 2007; 86: 409-440. doi: 10.1016/j.geomorph.2006.09.012
78. Wood J. The geomorphological chracteristics of digital elevation models. University of Leicester. 1996.
79. Wood J. Geomorphometry in LandSerf. In: Heng T, Reuter H editors. Geomorphometry: Concepts, Software, Applications Developments in Soils Science, 33. Amsterdam: Elsevier; 2009. pp. 333-349.
80. Breiman L, Friedman J, Olshen R, Stone C. Classification and Regression Trees. Belmont: Wadsworth International Group; 1984.
81. Breiman L, Cutler A. Breiman and Cutler's random forests for classification and regression. Packag "randomForest." 2012; 29. Available: https://cran.r-project.org/web/packages/randomForest/randomForest.pdf
82. Lawrence J. Neuronale Netze-Computersimulation biologischer Intelligenz. München: Systhema Verlag; 1992.
83. Scherer A. Neuronale Netze-Grundlagen und Anwendungen. Vieweg, editor. Braunschweig; 1997.
84. Masters T. Practical neural network recipies in C++. California: Academic Press; 1993.
85. Foresee FD, Hagan MT. GAUSS-NEWTON APPROXIMATION TO BAYESIAN LEARNING ∗∗ School of Electrical and Computer Engineering. Network. 1930; 1930-1935.
86. Perez Rodriguez P, Gianola D. Package "brnn." 2015
87. Friedman J. Multivariate Adaptive Regression Splines. Ann Stat. 1991; 19: 1-67.
88. Milborrow S. Package "earth". Multivariate Adaptive Regression Splines [Internet]. 2015. Available: http://cran.r-project.org/web/packages/earth/index.html.
89. Friedman JH. Fast MARS. Technical Repoert No. 110 [Internet]. 1993. Available: http://www.milbo.users.sonic.net/earth/Friedman-FastMars.pdf.
90. Elith J, Leathwick JR, Hastie T, Elith R. Leathwick J., Leathwick & Hastie A working guide to boosted regression trees - Online Appendices Page 1. J Anim Ecol. 2008; 77: 802-13. doi: 10.1111/j.1365-2656.2008.01390.x PMID: 18397250
91. Ridgeway G. Package "gbm". Generalized Boosted Regression Models. 2015; Available: http://cran.r-project.org/web/packages/gbm/index.html.
92. Freund Y, Schapire RE. A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting. J Comput Syst Sci. 1997; 55: 119-139. doi: 10.1006/jcss.1997.1504
93. Friedman JH. Stochastic Gradient Boosting. Comput Stat Data Anal. 2002; 38: 367-378.
94. Bennett K, Campbell C. Support vector machines: Hype or hallelujah? SIGKDD Explor. 2000; 2. Available: http://www.acm.org/sigs/sigkdd/explorations/issue2-2/bennett.pdf.
95. Vapnik VN. The nature of statistical learning theory. New York: Springer; 1995.
96. Smola A, Schölkopf B. A tutorial on support vector regression. Stat Comput. 2004; 14: 199-222. doi: 10.1023/B:STCO.0000035301.49549.88
97. Olsen D, Delen D. Advanced data mining techniques. Berlin: Springer; 2008.
98. Meyer D, Dimitriadou E, Hornik K, Weingessel A, Leisch F, Chang C, et al. Package "e1071". Misc Functions of the Department of Statistics (e1071) [Internet]. E1071. 2015. Available: http://cran.r-project.org/package=e1071
99. Mattera D, Haykin S. Support vector machines for dynamic reconstruction of a chaotic System. In: Schölkopf B.; Burges C.J.C.; Smola A.., editor. Advances in kernel methods-Support vector learning. Massachusets Institute of Technology; 1999.
100. Breiman L, Spector P. Submodel selection and evaluation in regression. The X-random case. Int Stat Rev. 1992; 60: 291-319. doi: 10.2307/1403680
101. Kohavi R. A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection. Int Jt Conf Artif Intell. 1995; 14: 1137-1143.
102. Weston S. Package 'doParallel'. Foreach Parallel Adaptor for the "parallel" Package [Internet]. 2015. Available: https://cran.r-project.org/web/packages/doParallel/doParallel.pdf.
103. Weston S. Package "foreach". Provides Foreach Looping Construct for R [Internet]. 2015. Available: https://cran.r-project.org/web/packages/foreach/foreach.pdf.
104. Moser G, Leuschner C, Hertel D, Graefe S, Soethe N, Iost S. Elevation effects on the carbon budget of tropical mountain forests (S Ecuador): The role of the belowground compartment. Glob Chang Biol. 2011; 17: 2211-2226. doi: 10.1111/j.1365-2486.2010.02367.x
105. Roman L, Scatena F, Bruijnzeel L. Global and local variations in tropical montane cloud forest soils. In: Brujinzeel LA, Scatena FN HL, editor. Tropical Montane Cloud Forests International Hydrology Series. New York: Cambridge University Press; 2010.
106. Dielemann W, Venter M, Ramachandra A, Krockenberger A, Bird M. Soil carbon stocks vary predictably with altitude in tropical forests: Implications for soil carbon storage. Geoderma. Elsevier B.V.; 2013; 204-205: 59-67. doi: 10.1016/j.geoderma.2013.04.005
107. Wilcke W, Boy J, Goller R, Fleischbein K, Valarezo C, Zech W. Effect of topography on soil fertility and water folw in an Ecuadorian lower montane forest. In: Brujinzeel LA, Scatena FN HL, editor. Tropical Montane Cloud Forests International Hydrology Series. New York: Cambridge University Press; 2010.
108. Waring R, Running S. Forest Ecosystems: Analysis at Multiple Scales. San Diego: Academic Press; 1998.