Integrating Airborne Hyperspectral, Topographic, and Soil Data for Estimating Pasture Quality Using Recursive Feature Elimination with Random Forest Regression

Accurate and efficient monitoring of pasture quality on hill country farm systems is crucial for pasture management and optimizing production. Hyperspectral imaging is a promising tool for mapping a wide range of biophysical and biochemical properties of vegetation from leaf to canopy scale. In this study, the potential of high spatial resolution and airborne hyperspectral imaging for predicting crude protein (CP) and metabolizable energy (ME) in heterogeneous hill country farm was investigated. Regression models were developed between measured pasture quality values and hyperspectral data using random forest regression (RF). The results proved that pasture quality could be predicted with hyperspectral data alone; however, accuracy was improved after combining the hyperspectral data with environmental data (elevation, slope angle, slope aspect, and soil type) where the prediction accuracy for CP was RCV (cross-validated coefficient of determination) = 0.70, RMSECV (cross-validated root mean square error) = 2.06%, RPDCV (cross-validated ratio to prediction deviation) = 1.82 and ME: RCV = 0.75, RMSECV = 0.65 MJ/kg DM, RPDCV = 2.11. Interestingly, the accuracy was further out-performed by considering important hyperspectral and environmental variables using RF combined with recursive feature elimination (RFE) (CP: RCV = 0.80, RMSECV = 1.68%, RPDCV = 2.23; ME: RCV = 0.78, RMSECV = 0.61 MJ/kg DM, RPDCV = 2.19). Similar performance trends were noticed with validation data. Utilizing the best model, spatial pasture quality maps were created across the farm. Overall, this study showed the potential of airborne hyperspectral data for producing accurate pasture quality maps, which will help farm managers to optimize decisions to improve environmental and economic benefits.

[1]  Ian J. Yule,et al.  Mapping of macro and micro nutrients of mixed pastures using airborne AisaFENIX hyperspectral imagery , 2016 .

[2]  Rasmus Bro,et al.  Variable selection in regression—a tutorial , 2010 .

[3]  L. Kumar,et al.  Estimating and mapping grass phosphorus concentration in an African savanna using hyperspectral image data , 2007 .

[4]  Elizabeth M. Middleton,et al.  Selection of Hyperspectral Narrowbands (HNBs) and Composition of Hyperspectral Twoband Vegetation Indices (HVIs) for Biophysical Characterization and Discrimination of Crop Types Using Field Reflectance and Hyperion/EO-1 Data , 2013, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[5]  Balázs Deák,et al.  Classification of Herbaceous Vegetation Using Airborne Hyperspectral Imagery , 2015, Remote. Sens..

[6]  Onisimo Mutanga,et al.  Forage quality of savannas - Simultaneously mapping foliar protein and polyphenols for trees and grass using hyperspectral imagery , 2010 .

[7]  R. Kokaly,et al.  Characterizing canopy biochemistry from imaging spectroscopy and its application to ecosystem studies , 2009 .

[8]  Onisimo Mutanga,et al.  High density biomass estimation for wetland vegetation using WorldView-2 imagery and random forest regression algorithm , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[9]  Zou Xiaobo,et al.  Variables selection methods in near-infrared spectroscopy. , 2010, Analytica chimica acta.

[10]  Jan G. P. W. Clevers,et al.  Optical remote sensing and the retrieval of terrestrial vegetation bio-geophysical properties - A review , 2015 .

[11]  R. Richter,et al.  Geo-atmospheric processing of airborne imaging spectrometry data. Part 2: Atmospheric/topographic correction , 2002 .

[12]  Georg Bareth,et al.  Evaluating hyperspectral vegetation indices for estimating nitrogen concentration of winter wheat at different growth stages , 2010, Precision Agriculture.

[13]  C. Furlanello,et al.  Recursive feature elimination with random forest for PTR-MS analysis of agroindustrial products , 2006 .

[14]  Simon D. Jones,et al.  Hyperspectral determination of feed quality constituents in temperate pastures: Effect of processing methods on predictive relationships from partial least squares regression , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[15]  Marco Heurich,et al.  Canopy foliar nitrogen retrieved from airborne hyperspectral imagery by correcting for canopy structure effects , 2017, Int. J. Appl. Earth Obs. Geoinformation.

[16]  A. Skidmore,et al.  Integrating imaging spectroscopy and neural networks to map grass quality in the Kruger National Park, South Africa , 2004 .

[17]  R. Pullanagari,et al.  Quantification of dead vegetation fraction in mixed pastures using AisaFENIX imaging spectroscopy data , 2017, Int. J. Appl. Earth Obs. Geoinformation.

[18]  Ian J. Yule,et al.  Developing variable rate application technology: Economic impact for farm owners and topdressing operators , 2007 .

[19]  P. Curran.  Remote sensing of foliar chemistry , 1989 .

[20]  K. Soder,et al.  Effect of soil type and fertilization level on mineral concentration of pasture: potential relationships to ruminant performance and health. , 2003, Journal of animal science.

[21]  Andrew K. Skidmore,et al.  Dry season mapping of savanna forage quality, using the hyperspectral Carnegie Airborne Observatory sensor , 2011 .

[22]  Mohamed Medhat Gaber,et al.  Random forests: from early developments to recent advancements , 2014 .

[23]  M. Cho,et al.  Non-linear partial least square regression increases the estimation accuracy of grass nitrogen and phosphorus using in situ hyperspectral and environmental data , 2013 .

[24]  V. Kakani,et al.  Estimation of bioenergy crop yield and N status by hyperspectral canopy reflectance and partial least square regression , 2017, Precision Agriculture.

[25]  M. P. Tuohy,et al.  Seasonal prediction of in situ pasture macronutrients in New Zealand pastoral systems using hyperspectral data , 2013 .

[26]  Matthias Rothmund,et al.  Precision agriculture on grassland : Applications, perspectives and constraints , 2008 .

[27]  P. Burrough,et al.  Principles of geographical information systems , 1998 .

[28]  Yoshio Inoue,et al.  Estimating forage biomass and quality in a mixed sown pasture based on partial least squares regression with waveband selection , 2008 .

[29]  Raymond F. Kokaly,et al.  Investigating a Physical Basis for Spectroscopic Estimates of Leaf Nitrogen Concentration , 2001 .

[30]  Elfatih M. Abdel-Rahman,et al.  Random forest regression and spectral band selection for estimating sugarcane leaf nitrogen concentration using EO-1 Hyperion hyperspectral data , 2013 .

[31]  A. Skidmore,et al.  Narrow band vegetation indices overcome the saturation problem in biomass estimation , 2004 .

[32]  John R. Miller,et al.  Hyperspectral vegetation indices and novel algorithms for predicting green LAI of crop canopies: Modeling and validation in the context of precision agriculture , 2004 .

[33]  M. P. Tuohy,et al.  Multi-spectral radiometry to estimate pasture quality components , 2012, Precision Agriculture.

[34]  Michael Wachendorf,et al.  Development of canopy reflectance models to predict forage quality of legume-grass mixtures. , 2009 .

[35]  Hao Zhou,et al.  Structure damage detection based on random forest recursive feature elimination , 2014 .

[36]  Ian J. Yule,et al.  Developing variable rate application technology: Scenario development and agronomic evaluation , 2007 .

[37]  A. Skidmore,et al.  Red edge shift and biochemical content in grass canopies , 2007 .

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

[39]  D. I. Gray,et al.  Controlling pasture quality on hill country-key decisions and techniques , 2015 .

[40]  J. Glover,et al.  Milk production from pasture , 1961, The Journal of Agricultural Science.

[41]  F. Scrimgeour,et al.  Pathways ahead for New Zealand hill country farming , 2016 .

[42]  Aixia Guo,et al.  Gene Selection for Cancer Classification using Support Vector Machines , 2014 .

[43]  M. P. Tuohy,et al.  In-field hyperspectral proximal sensing for estimating quality parameters of mixed pasture , 2011, Precision Agriculture.

[44]  J. Pereira,et al.  Modeling spatial patterns of fire occurrence in Mediterranean Europe using Multiple Regression and Random Forest , 2012 .

[45]  John G. Hodgson,et al.  New zealand pasture and crop science , 1999 .

[46]  F. Kırkpınar,et al.  Chemical Composition, In vivo Digestibility and Metabolizable Energy Values of Caramba (Lolium multiflorum cv. caramba) Fresh, Silage and Hay , 2015, Asian-Australasian journal of animal sciences.