Electrical imaging of soil water availability to grapevine: a benchmark experiment of several machine-learning techniques

Electrical resistivity (ER) can be used to assess soil water in the field. This study investigated the possibility of extending the use of ER to measure plant available soil water variables, i.e. available soil water (ASW), total transpirable SW (TTSW), and fraction of transpirable SW (FTSW) using a pedotransfer approach. In a vineyard, 224 electrical resistivity tomography (ERT) transects and 672 time domain reflectometry (TDR) soil water profiles were acquired over 2 years. Soil physical–chemical properties were measured on 73 soil samples from eight different sites. To estimate the amount of soil water available to plants, grapevine (Vitis vinifera L.) water status was monitored by means of leaf water potentials. A benchmark experiment was carried out to compare four machine-learning techniques: multivariate adaptive regression splines (MARS), k-nearest neighbours (KNN), random forest (RF), and gradient boosting machine (GBM). Model interpretation led to a deeper understanding of the relationships between electrical resistivity and soil properties when predicting soil water availability for the plant. The models assessed had good predictive performance and were therefore used to map ASW, TTSW and FTSW in the vineyard. ER coupled to machine-learning algorithms was shown to be a good proxy for quantification and visualisation of plant available soil water with low disturbance.

[1]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[2]  Eric Lebon,et al.  A model-based diagnosis tool to evaluate the water stress experienced by grapevine in field sites , 2006 .

[3]  Jacques Wery,et al.  RELATIONSHIPS BETWEEN PLANT AND SOIL WATER STATUS IN FIVE FIELD-GROWN COTTON (GOSSYPIUM HIRSUTUM L.) CULTIVARS , 1998 .

[4]  Liuwen－Zhao,et al.  Crop Yield Response to Water and Fertilizer in Loess Tableland of China：A Field Research , 1995 .

[5]  Gérard Barbeau,et al.  Contribution of soil electric resistivity measurements to the studies on soil/grapevine water relations , 2006 .

[6]  Vincent Dumas,et al.  Modelling the seasonal dynamics of the soil water balance of vineyards. , 2003, Functional plant biology : FPB.

[7]  S. Pellegrini,et al.  Relevance of the Lin's and Host hydropedological models to predict grape yield and wine quality , 2009 .

[8]  J. Maindonald.  Statistical Learning from a Regression Perspective , 2008 .

[9]  B. Nicoullaud,et al.  Spatial and temporal monitoring of soil water content with an irrigated corn crop cover using surface electrical resistivity tomography , 2003 .

[10]  M. Loke.  Tutorial : 2-D and 3-D electrical imaging surveys , 2001 .

[11]  Brian D. Ripley,et al.  Modern applied statistics with S, 4th Edition , 2002, Statistics and computing.

[12]  J. Freidman,et al.  Multivariate adaptive regression splines , 1991 .

[13]  D. Hillel.  Environmental soil physics , 1998 .

[14]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[15]  G. Richard,et al.  Electrical resistivity survey in soil science: a review . , 2005 .

[16]  Manuel J. A. Eugster,et al.  Benchmark experiments: a tool for analyzing statistical learning algorithms , 2011 .

[17]  H. Vereecken,et al.  High-resolution imaging of a vineyard in south of France using ground penetrating radar, electromagnetic induction and electrical resistivity tomography , 2012 .

[18]  L. A. Richards,et al.  Change of Electrical Conductivity With Temperature and the Relation of Osmotic Pressure to Electrical Conductivity and Ion Concentration for Soil Extracts , 1949 .

[19]  BrillanteLuca,et al.  Investigating the use of gradient boosting machine, random forest and their ensemble to predict skin flavonoid content from berry physical-mechanical characteristics in wine grapes , 2015 .

[20]  L. Williams,et al.  The effects of applied water at various fractions of measured evapotranspiration on reproductive growth and water productivity of Thompson Seedless grapevines , 2010, Irrigation Science.

[21]  Recovery of Populus tremuloides seedlings following severe drought causing total leaf mortality and extreme stem embolism. , 2010, Physiologia plantarum.

[22]  T. Hothorn,et al.  Simultaneous Inference in General Parametric Models , 2008, Biometrical journal. Biometrische Zeitschrift.

[23]  Said Attia Al Hagrey,et al.  Geophysical imaging of root-zone, trunk, and moisture heterogeneity. , 2007, Journal of experimental botany.

[24]  P. Kramer.  Soil moisture in relation to plant growth , 1944, The Botanical Review.

[25]  D. Bates,et al.  Mixed-Effects Models in S and S-PLUS , 2001 .

[26]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[27]  Diane M. McKnight,et al.  Transport and cycling of iron and hydrogen peroxide in a freshwater stream: Influence of organic acids , 2003 .

[28]  Rate of leaf production in response to soil water deficits in field pea , 1998 .

[29]  R. Perria,et al.  The use of the ARP© system to reduce the costs of soil survey for precision viticulture , 2013 .

[30]  Simone Giacosa,et al.  Investigating the use of gradient boosting machine, random forest and their ensemble to predict skin flavonoid content from berry physical-mechanical characteristics in wine grapes , 2015, Comput. Electron. Agric..

[31]  H. Jones,et al.  Using plant physiology to improve the water use efficiency of horticultural crops. , 2000 .

[32]  P. Dry,et al.  Hormonal changes induced by partial rootzone drying of irrigated grapevine. , 2000, Journal of experimental botany.

[33]  Walter Krämer,et al.  Review of Modern applied statistics with S, 4th ed. by W.N. Venables and B.D. Ripley. Springer-Verlag 2002 , 2003 .

[34]  Giovanni Bitella,et al.  Electrical resistivity tomography as a non‐destructive method for mapping root biomass in an orchard , 2011 .

[35]  P. F. Scholander,et al.  Sap Pressure in Vascular Plants , 1965, Science.

[36]  A. Pellegrino,et al.  Relationships between plant and soil water status in vine (Vitis vinifera L.) , 2005, Plant and Soil.

[37]  William N. Venables,et al.  Modern Applied Statistics with S , 2010 .

[38]  C. Doussan,et al.  Estimation of the spatial variability of root water uptake of maize and sorghum at the field scale by electrical resistivity tomography , 2009, Plant and Soil.

[39]  Jones Arwyn,et al.  World reference base for soil resources 2014International soil classification system for naming soils and creating legends for soil maps , 2015 .

[40]  N. Holbrook,et al.  Daily transpiration rates of woody species on drying soil. , 2005, Tree physiology.

[41]  Jean Lévêque,et al.  The use of soil electrical resistivity to monitor plant and soil water relationships in vineyards. , 2014 .

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

[43]  J. Lhomme,et al.  Modelling the daily course of capitulum temperature in a sunflower canopy , 2006 .

[44]  P. Dry,et al.  Effects of partial root-zone drying on grapevine vigour, yield, composition of fruit and use of water , 1996 .

[45]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[46]  Jean Lévêque,et al.  Monitoring soil volume wetness in heterogeneous soils by electrical resistivity. A field-based pedotransfer function , 2014 .

[47]  J. Deckers,et al.  World Reference Base for Soil Resources , 1998 .

[48]  Y. Pachepsky,et al.  Pedotransfer functions in soil electrical resistivity estimation , 2011 .

[49]  Christian Gary,et al.  WaLIS--A simple model to simulate water partitioning in a crop association: The example of an intercropped vineyard , 2010 .

[50]  Joe T. Ritchie,et al.  Water dynamics in the soil-plant-atmosphere system , 1981, Plant and Soil.

[51]  D. Bates,et al.  Linear Mixed-Effects Models using 'Eigen' and S4 , 2015 .