Estimating spatial variations in soil organic carbon using satellite hyperspectral data and map algebra

This study evaluated the effectiveness of using Hyperion hyperspectral data in improving existing remote-sensing methodologies for estimating soil organic carbon (SOC) content on farmland. The study area is Big Creek Watershed in Southern Illinois, USA. Several data-mining techniques were tested to calibrate and validate models that could be used for predicting SOC content using Hyperion bands as predictors. A combined model of stepwise regression followed by a five hidden nodes artificial neural network was selected as the best model, with a calibration coefficient of determination (R 2) of 78.9% and a root mean square error (RMSE) of 3.3 tonnes per hectare (t ha−1). The validation RMSE, however, was found to be 11.3 t ha−1. Map algebra was implemented to extrapolate this model and produce a SOC map for the watershed. Hyperspectral data improved marginally the predictability of SOC compared to multispectral data under natural field conditions. They could not capture small annual variations in SOC, but could measure decadal variations with moderate error. Satellite-based hyperspectral data combined with map algebra can measure total SOC pools in various ecosystem or soil types to within a few per cent error.

[1]  K. Shepherd,et al.  Development of Reflectance Spectral Libraries for Characterization of Soil Properties , 2002 .

[2]  Jane R. Foster,et al.  Predicting tropical forest carbon from EO-1 hyperspectral imagery in Noel Kempff Mercado National Park, Bolivia , 2002, IEEE International Geoscience and Remote Sensing Symposium.

[3]  R. Henry,et al.  Simultaneous Determination of Moisture, Organic Carbon, and Total Nitrogen by Near Infrared Reflectance Spectrophotometry , 1986 .

[4]  Edward A. McBean,et al.  Statistical procedures for analysis of environmental monitoring data and risk assessment , 2000 .

[5]  R. Lal,et al.  Soil carbon dynamics in cropland and rangeland. , 2002, Environmental pollution.

[6]  J. Rantanen,et al.  In-line moisture measurement during granulation with a four-wavelength near-infrared sensor: an evaluation of process-related variables and a development of non-linear calibration model , 2001 .

[7]  C. Mallows.  More comments on C p , 1995 .

[8]  K. Smith,et al.  SOILS AND THE GREENHOUSE EFFECT , 1997 .

[9]  Holger R. Maier,et al.  An Evaluation of Methods for the Selection of Inputs for an Artificial Neural Network Based River Model , 2006 .

[10]  Ronald Amundson,et al.  The Carbon Budget in Soils , 2001 .

[11]  Philip K. Hopke,et al.  Application of PLS and Back-Propagation Neural Networks for the estimation of soil properties , 2005 .

[12]  D. C. Howell.  Statistical Methods for Psychology , 1987 .

[13]  C. L. Mallows.  Some Comments onCp , 1973 .

[14]  B. Kowalski,et al.  Partial least-squares regression: a tutorial , 1986 .

[15]  A. Bouwman,et al.  Soils and the greenhouse effect. , 1990 .

[16]  L. Duram,et al.  A Local Example of Land-Use Change: Southern Illinois—1807, 1938, and 1993* , 2004 .

[17]  L. West,et al.  Field-Scale Mapping of Surface Soil Organic Carbon Using Remotely Sensed Imagery , 2000 .

[18]  F. R. Troeh,et al.  Introductory Soil Science, Laboratory Manual , 1978 .

[19]  T. G. Crowe,et al.  Reflectance models for predicting organic carbon in Saskatchewan soils , 2000 .

[20]  M. Forina,et al.  Multivariate calibration. , 2007, Journal of chromatography. A.

[21]  Marvin H. Hall,et al.  Carbon and Nitrogen Analysis of Soil Fractions Using Near-Infrared Reflectance Spectroscopy , 1991 .

[22]  H. Akaike.  A new look at the statistical model identification , 1974 .

[23]  Mark Peters,et al.  Economics of Sequestering Carbon in the U.S. Agricultural Sector , 2004 .

[24]  Anne-Johan Annema.  Feed-forward neural networks - vector decomposition analysis, modelling and analog implementation , 1995, The Kluwer international series in engineering and computer science.

[25]  C. Mallows.  Some Comments on Cp , 2000, Technometrics.

[26]  John R. Jensen,et al.  Introductory Digital Image Processing: A Remote Sensing Perspective , 1986 .

[27]  Fernando Bação,et al.  Self-organizing Maps as Substitutes for K-Means Clustering , 2005, International Conference on Computational Science.

[28]  P. Gong,et al.  Quantity and spatial variability of soil carbon in the conterminous United States , 2006 .

[29]  Jan-Tai Kuo,et al.  USING ARTIFICIAL NEURAL NETWORK FOR RESERVOIR EUTROPHICATION PREDICTION , 2007 .

[30]  Can Ozan Tan,et al.  Methodological issues in building, training, and testing artificial neural networks in ecological applications , 2005, q-bio/0510017.

[31]  P. A. Agbu,et al.  Soil Property Relationships with SPOT Satellite Digital Data in East Central Illinois , 1990 .

[32]  Margaret J. Robertson,et al.  Design and Analysis of Experiments , 2006, Handbook of statistics.

[33]  R. V. Rossel,et al.  Soil organic carbon prediction by hyperspectral remote sensing and field vis-NIR spectroscopy: An Australian case study , 2008 .

[34]  T. M. Lillesand,et al.  Remote Sensing and Image Interpretation , 1980 .

[35]  S. Grunwald,et al.  Tree-based modeling of complex interactions of phosphorus loadings and environmental factors. , 2009, The Science of the total environment.

[36]  B. Frazier,et al.  Relationship between soil organic carbon and Landsat TM data in eastern Washington , 1994 .

[37]  E. Davidson,et al.  Estimating regional carbon stocks and spatially covarying edaphic factors using soil maps at three scales , 1993 .

[38]  C. Lee Burras,et al.  Equations for Predicting Soil Organic Carbon Using Loss‐on‐Ignition for North Central U.S. Soils , 2002 .

[39]  Holger R. Maier,et al.  Input determination for neural network models in water resources applications. Part 1—background and methodology , 2005 .

[40]  James R. Anderson,et al.  A land use and land cover classification system for use with remote sensor data , 1976 .

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

[42]  Qiming Zhou,et al.  Estimating and Analyzing the Spatial Distribution of Soil Organic Carbon in China , 2003, Ambio.

[43]  Frédérique Seyler,et al.  Land cover mapping and carbon pools estimates in Rondonia, Brazil , 1998 .

[44]  W. Post,et al.  Soil organic carbon sequestration rates by tillage and crop rotation : A global data analysis , 2002 .

[45]  Karen A. F. Copeland.  Experiments: Planning, Analysis, and Parameter Design Optimization , 2002 .

[46]  Jeffrey S. Kern,et al.  Spatial Patterns of Soil Organic Carbon in the Contiguous United States , 1994 .

[47]  C. Walthall,et al.  Artificial neural networks for corn and soybean yield prediction , 2005 .

[48]  D. W. Nelson,et al.  Total Carbon, Organic Carbon, and Organic Matter , 1983, SSSA Book Series.

[49]  Mark Guetersloh,et al.  Big Creek Watershed Restoration Plan: A Component of the Cache River Watershed Resource Plan , 2002 .

[50]  M. Gevrey,et al.  Review and comparison of methods to study the contribution of variables in artificial neural network models , 2003 .

[51]  Roberto C. Izaurralde,et al.  Monitoring and Verifying Changes of Organic Carbon in Soil , 2001 .

[52]  Desire L. Massart,et al.  Artificial neural networks in classification of NIR spectral data: Selection of the input , 1996 .

[53]  S. Kultti,et al.  Upscaling soil organic carbon estimates for the Usa Basin (Northeast European Russia) using GIS-based landcover and soil classification schemes , 2002 .

[54]  Leandro Nunes de Castro,et al.  Fundamentals of natural computing: an overview , 2007 .

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

[56]  F. Blaine Metting,et al.  Science Needs and New Technology for Increasing Soil Carbon Sequestration , 2001 .

[57]  Raja Sengupta,et al.  Evaluating the Impact of Policy-induced Land Use Management Practices on Non-point Source Pollution Using a Spatial Decision Support System , 2000 .

[58]  Wsd Wong,et al.  Statistical Analysis of Geographic Information with ArcView GIS And ArcGIS , 2005 .