Vineyard water status estimation using multispectral imagery from an UAV platform and machine learning algorithms for irrigation scheduling management

A machine learning model was developed to estimate stem water potential from UAV.An additional model was developed to detect three levels of water stress.Models were developed at pixel-by-pixel and plant by plant scales.Spatial distribution maps were constructed for a grapevine region for irrigation.High accuracy models developed could assist irrigation and grapevine management. Remote sensing can provide a fast and reliable alternative for traditional in situ water status measurement in vineyards. Several vegetation indices (VIs) derived from aerial multispectral imagery were tested to estimate midday stem water potential (stem) of grapevines. The experimental trial was carried out in a vineyard in the Shangri-La region, located in Yunnan province in China. Statistical methods and machine learning algorithms were used to evaluate the correlations between stem and VIs. Results by simple regression between VIs individually and stem showed no significant relationships, with coefficient of determination (R2) for linear fitting smaller than 0.3 for almost all the indices studied, except for the Optimal Soil Adjusted Vegetation Index (OSAVI); R2=0.42 with statistical significance (p0.001). However, results from a model obtained by fitting using Artificial Neural Network (ANN), using all VIs calculated as inputs and real stem from plants within the study site (n=90) as targets (Model 1), showed high correlation between the estimated water potential through ANN (stem ANN) and the actual measured stem. Training, validation and testing data sets presented individual correlations of R=0.8, 0.72 and 0.62 respectively. The models obtained from the study site were then applied to a wider area from the vineyard studied and compared to further stem measured obtained from different sites (n=23) showing high correlation values between stem ANN and real stem (R2=0.83; slope=1; p0.001). Finally, a pattern recognition ANN model (Model 2) was developed for irrigation scheduling purposes using the same stem measured in the study site as inputs and with the following thresholds as outputs: stem below 1.2MPa considered as severe water stress (SS), stem between 0.8 to 1.2MPa as moderate stress (MS) and stem over 0.8MPa with no water stress (NS). This model can be applied to analyze on a plant by plant basis to identify sectors of stress within the vineyard for optimal irrigation management and to identify spatial variability within the vineyards.

[1]  Bruno Tisseyre,et al.  Temporal stability of within-field patterns of NDVI in non irrigated Mediterranean vineyards , 2011 .

[2]  M. A. Skewes,et al.  Night-time sap flow is parabolically linked to midday water potential for field-grown almond trees , 2013, Irrigation Science.

[3]  J. Peñuelas,et al.  Estimation of plant water concentration by the reflectance Water Index WI (R900/R970) , 1997 .

[4]  J. Roujean,et al.  Estimating PAR absorbed by vegetation from bidirectional reflectance measurements , 1995 .

[5]  J. Flexas,et al.  A new instrument for passive remote sensing 1. Measurements of sunlight-induced chlorophyll fluorescence , 2004 .

[6]  G. Birth,et al.  Measuring the Color of Growing Turf with a Reflectance Spectrophotometer1 , 1968 .

[7]  Yuxuan Wang,et al.  A Leaf Recognition Algorithm for Plant Classification Using Probabilistic Neural Network , 2007, 2007 IEEE International Symposium on Signal Processing and Information Technology.

[8]  P. F. Scholander,et al.  HYDROSTATIC PRESSURE AND OSMOTIC POTENTIAL IN LEAVES OF MANGROVES AND SOME OTHER PLANTS. , 1964, Proceedings of the National Academy of Sciences of the United States of America.

[9]  Stefano Amaducci,et al.  Assessing canopy PRI from airborne imagery to map water stress in maize , 2013 .

[10]  L. Williams,et al.  Correlations among Predawn Leaf, Midday Leaf, and Midday Stem Water Potential and their Correlations with other Measures of Soil and Plant Water Status in Vitis vinifera , 2002 .

[11]  Mario Minacapilli,et al.  Detecting crop water status in mature olive groves using vegetation spectral measurements , 2014 .

[12]  L. S. Pereira,et al.  Predicting Grapevine Water Status Based on Hyperspectral Reflectance Vegetation Indices , 2015, Remote. Sens..

[13]  H. Jones.  Monitoring plant and soil water status: established and novel methods revisited and their relevance to studies of drought tolerance. , 2006, Journal of experimental botany.

[14]  J. Peñuelas,et al.  The red edge position and shape as indicators of plant chlorophyll content, biomass and hydric status. , 1994 .

[15]  W. E. Larson,et al.  Coincident detection of crop water stress, nitrogen status and canopy density using ground-based multispectral data. , 2000 .

[16]  John E. Gilley,et al.  Artificial Neural Network estimation of soil erosion and nutrient concentrations in runoff from land application areas , 2008 .

[17]  Alan A. Ager,et al.  Broadband, red-edge information from satellites improves early stress detection in a New Mexico conifer woodland , 2011 .

[18]  R. De Bei,et al.  Night-time responses to water supply in grapevines (Vitis vinifera L.) under deficit irrigation and partial root-zone drying , 2014 .

[19]  J. Everitt,et al.  Using spectral vegetation indices to estimate rangeland productivity , 1992 .

[20]  Janet Franklin,et al.  A Neural Network Method for Efficient Vegetation Mapping , 1999 .

[21]  John R. Miller,et al.  Assessing vineyard condition with hyperspectral indices: Leaf and canopy reflectance simulation in a row-structured discontinuous canopy , 2005 .

[22]  Pau Martí,et al.  An artificial neural network approach to the estimation of stem water potential from frequency domain reflectometry soil moisture measurements and meteorological data , 2013 .

[23]  Marie-Françoise Courel,et al.  Utilisation des bandes spectrales du vert et du rouge pour une meilleure évaluation des formations végétales actives , 1991 .

[24]  D. Smart,et al.  Evaluation of Hyperspectral Reflectance Indexes to Detect Grapevine Water Status in Vineyards , 2007, American Journal of Enology and Viticulture.

[25]  P. Zarco-Tejada,et al.  A PRI-based water stress index combining structural and chlorophyll effects: Assessment using diurnal narrow-band airborne imagery and the CWSI thermal index , 2013 .

[26]  Carlos Lopes,et al.  Modern viticulture in southern Europe: Vulnerabilities and strategies for adaptation to water scarcity , 2016 .

[27]  Kuolin Hsu,et al.  Artificial Neural Network Modeling of the Rainfall‐Runoff Process , 1995 .

[28]  Benjamin Bois,et al.  Vine water status is a key factor in grape ripening and vintage quality for red Bordeaux wine. How can it be assessed for vineyard management purposes , 2009 .

[29]  J. Baluja,et al.  Assessment of vineyard water status variability by thermal and multispectral imagery using an unmanned aerial vehicle (UAV) , 2012, Irrigation Science.

[30]  John A. Gamon,et al.  Assessing leaf pigment content and activity with a reflectometer , 1999 .

[31]  J. A. Schell,et al.  Monitoring vegetation systems in the great plains with ERTS , 1973 .

[32]  Souad Riad,et al.  Rainfall-runoff model usingan artificial neural network approach , 2004, Math. Comput. Model..

[33]  Laurent Tits,et al.  Stem Water Potential Monitoring in Pear Orchards through WorldView-2 Multispectral Imagery , 2013, Remote. Sens..

[34]  Michael E. Schaepman,et al.  A review on reflective remote sensing and data assimilation techniques for enhanced agroecosystem modeling , 2007, Int. J. Appl. Earth Obs. Geoinformation.

[35]  Rüdiger Henrich,et al.  Late Quaternary climatic events and sea-level changes recorded by turbidite activity, Dakar Canyon, NW Africa , 2010, Quaternary Research.

[36]  N. Turner.  Measurement of plant water status by the pressure chamber technique , 1988, Irrigation Science.

[37]  Pablo J. Zarco-Tejada,et al.  High-resolution airborne hyperspectral and thermal imagery for early detection of Verticillium wilt of olive using fluorescence, temperature and narrow-band spectral indices , 2013 .

[38]  Vladimir Alexandrov,et al.  Artificial neural networks and their application in biological and agricultural research , 2014 .

[39]  P. Curran,et al.  Technical Note Grass chlorophyll and the reflectance red edge , 1996 .

[40]  Finn Plauborg,et al.  Estimating plant root water uptake using a neural network approach , 2010 .

[41]  N. Conceição,et al.  Transpiration and water stress effects on water use, in relation to estimations from NDVI: Application in a vineyard in SE Portugal , 2012 .

[42]  B. Tisseyre,et al.  Spatial extrapolation of the vine (Vitis vinifera L.) water status: a first step towards a spatial prediction model , 2009, Irrigation Science.

[43]  Cornelis van Leeuwen,et al.  Stem Water Potential is a Sensitive Indicator of Grapevine Water Status , 2001 .

[44]  Jérôme Grimplet,et al.  Water deficit alters differentially metabolic pathways affecting important flavor and quality traits in grape berries of Cabernet Sauvignon and Chardonnay , 2009, BMC Genomics.

[45]  B. Debska,et al.  Application of artificial neural network in food classification. , 2011, Analytica chimica acta.

[46]  Sigfredo Fuentes,et al.  Digital surface model applied to unmanned aerial vehicle based photogrammetry to assess potential biotic or abiotic effects on grapevine canopies , 2016 .

[47]  K. P. Sudheer,et al.  Modelling evaporation using an artificial neural network algorithm , 2002 .

[48]  Sigfredo Fuentes,et al.  An automated procedure for estimating the leaf area index (LAI) of woodland ecosystems using digital imagery, MATLAB programming and its application to an examination of the relationship between remotely sensed and field measurements of LAI. , 2008, Functional plant biology : FPB.

[49]  Gregory A. Carter,et al.  Responses of leaf spectral reflectance to plant stress. , 1993 .

[50]  Samuel Ortega-Farías,et al.  Is it possible to assess the spatial variability of vine water status , 2008 .

[51]  Anatoly A. Gitelson,et al.  Non‐destructive detection of water stress and estimation of relative water content in maize , 2009 .

[52]  D. Horler,et al.  The red edge of plant leaf reflectance , 1983 .

[53]  John A. Gamon,et al.  Monitoring seasonal and diurnal changes in photosynthetic pigments with automated PRI and NDVI sensors , 2015 .

[54]  Wan,et al.  Application of artificial neural network in predicting crop yield: a review , 2014 .

[55]  Ismael Moya,et al.  A new instrument for passive remote sensing: 2. Measurement of leaf and canopy reflectance changes at 531 nm and their relationship with photosynthesis and chlorophyll fluorescence , 2004 .

[56]  B. Govaerts,et al.  The normalized difference vegetation index (NDVI) Greenseeker(TM) handheld sensor: toward the integrated evaluation of crop management. Part A - Concepts and case studies , 2010 .

[57]  A. Huete,et al.  A Modified Soil Adjusted Vegetation Index , 1994 .

[58]  Matthew Bardeen,et al.  Artificial Neural Network to Predict Vine Water Status Spatial Variability Using Multispectral Information Obtained from an Unmanned Aerial Vehicle (UAV) , 2017, Sensors.

[59]  Craig S. T. Daughtry,et al.  A visible band index for remote sensing leaf chlorophyll content at the canopy scale , 2013, Int. J. Appl. Earth Obs. Geoinformation.