Rice height and biomass estimations using multitemporal SAR Sentinel-1: Camargue case study

The research and improvement of methods to be used for crop monitoring are currently major challenges, especially for radar images due to their speckle noise nature. The European Space Agency’s (ESA) Sentinel1 constellation provides synthetic aperture radar (SAR) images coverage with a 6 days revisit period at a high spatial resolution of pixel spacing 20 m. Sentinel-1 data are considerable useful, as they provide valuable information of the vegetation cover. The objective of this paper is to provide a better understanding of the capabilities of Sentinel-1 radar images for rice height and dry biomass retrievals. To do this, we train Sentinel1 data against ground measurements with classical machine learning techniques (Multiple Linear Regression (MLR), Support Vector Regression (SVR) and Random Forest (RF)) to estimate rice height and dry biomass. The study is carried out on a multi-temporal Sentinel-1 dataset acquired from May 2017 to September 2017 over the Camargue region, southern France. The ground in-situ measurements were made in the same period to collect rice height and dry biomass over 11 rice fields. The images were processed in order to produce an intensity radar data stack in C-band including dual-polarization VV (Vertical receive and Vertical transmit) and VH (Vertical receive and Horizontal transmit) data. We found that non-parametric methods (SVR and RF) had a better performance over the parametric MLR method for rice biophysical parameter retrievals. The accuracy of rice height estimation showed that rice height retrieval was strongly correlated to the in-situ rice height from dual-polarization, in which Random Forest yielded the best performance with correlation coefficient R2 = 0.92 and the root mean square error (RMSE) 16% (7.9 cm). In addition, we demonstrated that the correlation of Sentinel-1 signal to the biomass was also very high in VH polarization with R2 = 0.9 and RMSE = 18% (162 g.m−2 ) (with Random Forest method). Such results indicate that the highly qualified Sentinel-1 radar data could be well exploited for rice biomass and height retrieval and they could be used for operational tasks

[1]  Alexandre Bouvet,et al.  Detection of Irrigated Crops from Sentinel-1 and Sentinel-2 Data to Estimate Seasonal Groundwater Use in South India , 2017, Remote. Sens..

[2]  Hui Lin,et al.  Application of ENVISAT ASAR Data in Mapping Rice Crop Growth in Southern China , 2007, IEEE Geoscience and Remote Sensing Letters.

[3]  Andreas Reigber,et al.  TOPS Interferometry With TerraSAR-X , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[4]  Yan Chen,et al.  Rice Biomass Estimation Using Radar Backscattering Data at S-band , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[5]  Masaharu Fujita,et al.  Monitoring of rice crop growth from space using the ERS-1 C-band SAR , 1995, IEEE Trans. Geosci. Remote. Sens..

[6]  Jiaguo Qi,et al.  Monitoring Rice Agriculture across Myanmar Using Time Series Sentinel-1 Assisted by Landsat-8 and PALSAR-2 , 2017, Remote. Sens..

[7]  Georgia Fotopoulos,et al.  Fusion of SAR, Optical Imagery and Airborne LiDAR for Surface Water Detection , 2017, Remote. Sens..

[8]  Ian T. Jolliffe,et al.  Principal Component Analysis , 2002, International Encyclopedia of Statistical Science.

[9]  Irena Hajnsek,et al.  First Results of Rice Monitoring Practices in Spain by Means of Time Series of TerraSAR-X Dual-Pol Images , 2011, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[10]  Changsheng Li,et al.  Mapping paddy rice agriculture in southern China using multi-temporal MODIS images , 2005 .

[11]  H. Liu,et al.  Estimation of rice biophysical parameters using multitemporal RADARSAT-2 images , 2016 .

[12]  W.G.N.N. Jayawardhana,et al.  Extraction of Agricultural Phenological Parameters of Sri Lanka Using MODIS, NDVI Time Series Data , 2016 .

[13]  Emile Ndikumana,et al.  Deep Recurrent Neural Network for Agricultural Classification using multitemporal SAR Sentinel-1 for Camargue, France , 2018, Remote. Sens..

[14]  Yuan Zhang,et al.  Retrieving canopy height and density of paddy rice from Radarsat-2 images with a canopy scattering model , 2014, Int. J. Appl. Earth Obs. Geoinformation.

[15]  R. Harris,et al.  Extracting biophysical parameters from remotely sensed radar data: a review of the water cloud model , 2003 .

[16]  B. Brisco,et al.  Rice monitoring and production estimation using multitemporal RADARSAT , 2001 .

[17]  Malcolm Davidson,et al.  GMES Sentinel-1 mission , 2012 .

[18]  Dieu Tien Bui,et al.  Improving Accuracy Estimation of Forest Aboveground Biomass Based on Incorporation of ALOS-2 PALSAR-2 and Sentinel-2A Imagery and Machine Learning: A Case Study of the Hyrcanian Forest Area (Iran) , 2018, Remote. Sens..

[19]  Emile Ndikumana,et al.  Potential value of combining ALOS PALSAR and Landsat-derived tree cover data for forest biomass retrieval in Madagascar , 2018, Remote Sensing of Environment.

[20]  Maxim Neumann,et al.  Assessing Performance of L- and P-Band Polarimetric Interferometric SAR Data in Estimating Boreal Forest Above-Ground Biomass , 2012, IEEE Transactions on Geoscience and Remote Sensing.

[21]  Jinshui Wu,et al.  Rice rhizodeposition and its utilization by microbial groups depends on N fertilization , 2016, Biology and Fertility of Soils.

[22]  J. B. Dent,et al.  Applications of Systems Approaches at the Farm and Regional Levels Volume 1 , 1997, Systems Approaches for Sustainable Agricultural Development.

[23]  Thuy Le Toan,et al.  Relating P-Band Synthetic Aperture Radar Tomography to Tropical Forest Biomass , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[24]  Vojislav Kecman,et al.  Kernel Based Algorithms for Mining Huge Data Sets: Supervised, Semi-supervised, and Unsupervised Learning , 2006, Studies in Computational Intelligence.

[25]  Fumin Wang,et al.  Identification of optimal hyperspectral bands for estimation of rice biophysical parameters. , 2008, Journal of integrative plant biology.

[26]  Shaun Quegan,et al.  Filtering of multichannel SAR images , 2001, IEEE Trans. Geosci. Remote. Sens..

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

[28]  Bo Zhang,et al.  Rice Crop Monitoring in South China With RADARSAT-2 Quad-Polarization SAR Data , 2011, IEEE Geoscience and Remote Sensing Letters.

[29]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[30]  D. L. Hall,et al.  Mathematical Techniques in Multisensor Data Fusion , 1992 .

[31]  N. T. Meetei,et al.  QTL Mapping for Cold Tolerance at the Seedling Stage in Rice , 2016 .

[32]  Ho Tong Minh Dinh,et al.  Tomosar platform supports for Sentinel-1 tops persistent scatterers interferometry , 2017, 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).

[33]  Dennis S. Bernstein,et al.  Matrix Mathematics: Theory, Facts, and Formulas with Application to Linear Systems Theory , 2005 .

[34]  Matthias Drusch,et al.  Sentinel-2: ESA's Optical High-Resolution Mission for GMES Operational Services , 2012 .

[35]  Runhe Shi,et al.  Retrieving rice yield and biomass from Radarsat-2 SAR data with Artificial Neural Network (ANN) , 2013, Optics & Photonics - Optical Engineering + Applications.

[36]  Pablo Tittonell,et al.  On farm assessment of rice yield variability and productivity gaps between organic and conventional cropping systems under Mediterranean climate , 2011 .

[37]  M. Hossain,et al.  Rice supply and demand in Asia: a socioeconomic and biophysical analysis , 1997 .

[38]  Sotiris B. Kotsiantis,et al.  Machine learning: a review of classification and combining techniques , 2006, Artificial Intelligence Review.

[39]  Stefano Pignatti,et al.  Sensitivity analysis of the Aquacrop and SAFYE crop models for the assessment of water limited winter wheat yield in regional scale applications , 2017, PloS one.

[40]  R. Touzi,et al.  Polarimetric decomposition with RADARSAT-2 for rice mapping and monitoring , 2012 .

[41]  Simonetta Paloscia,et al.  The potential of multifrequency polarimetric SAR in assessing agricultural and arboreous biomass , 1997, IEEE Trans. Geosci. Remote. Sens..

[42]  M. Wopereis,et al.  Crops that feed the world 7: Rice , 2012, Food Security.

[43]  Paule Moustier,et al.  Organisation en réseau et durabilité systémique de deux filières alimentaires (riz biologique et petit épeautre en France) , 2014 .

[44]  J. Wu,et al.  Mapping Paddy Rice Biomass Using ALOS/PALSAR Imagery , 2008, 2008 International Workshop on Education Technology and Training & 2008 International Workshop on Geoscience and Remote Sensing.

[45]  F. Rocca,et al.  SAR tomography for the retrieval of forest biomass and height: cross-validation at two tropical forest sites in French Guiana , 2016 .

[46]  Cristian Rossi,et al.  Paddy-Rice Monitoring Using TanDEM-X , 2015, IEEE Transactions on Geoscience and Remote Sensing.

[47]  Jean-Claude Mouret,et al.  Étude de l'agrosystème rizicole en Camargue dans ses relations avec le milieu et le système cultural : aspects particuliers de la fertilité , 1988 .

[48]  Aldo Ferrero,et al.  Meeting the challenges of global rice production , 2006, Paddy and Water Environment.

[49]  Daniela Stroppiana,et al.  Exploitation of SAR and Optical Sentinel Data to Detect Rice Crop and Estimate Seasonal Dynamics of Leaf Area Index , 2017, Remote. Sens..

[50]  P Snoeij,et al.  Sentinel-1 radar mission: Status and performance , 2009, IEEE Aerospace and Electronic Systems Magazine.

[51]  Chih-Jen Lin,et al.  Working Set Selection Using Second Order Information for Training Support Vector Machines , 2005, J. Mach. Learn. Res..

[52]  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.

[53]  Thuy Le Toan,et al.  Rice crop mapping and monitoring using ERS-1 data based on experiment and modeling results , 1997, IEEE Trans. Geosci. Remote. Sens..

[54]  A. Fung,et al.  Electromagnetic wave scattering from some vegetation samples , 1988 .

[55]  Irena Hajnsek,et al.  Estimation of Rice Crop Height From X- and C-Band PolSAR by Metamodel-Based Optimization , 2017, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[56]  Changsheng Li,et al.  Mapping paddy rice agriculture in South and Southeast Asia using multi-temporal MODIS images , 2006 .

[57]  Cuizhen Wang,et al.  Capability of C-band backscattering coefficients from high-resolution satellite SAR sensors to assess biophysical variables in paddy rice , 2014 .