Estimation of Potato Yield Using Satellite Data at a Municipal Level: A Machine Learning Approach

Crop growth modeling and yield forecasting are essential to improve food security policies worldwide. To estimate potato (Solanum tubersum L.) yield over Mexico at a municipal level, we used meteorological data provided by the ERA5 (ECMWF Re-Analysis) dataset developed by the Copernicus Climate Change Service, satellite imagery from the TERRA platform, and field information. Five different machine learning algorithms were used to build the models: random forest (rf), support vector machine linear (svmL), support vector machine polynomial (svmP), support vector machine radial (svmR), and general linear model (glm). The optimized models were tested using independent data (2017 and 2018) not used in the training and optimization phase (2004–2016). In terms of percent root mean squared error (%RMSE), the best results were obtained by the rf algorithm in the winter cycle using variables from the first three months of the cycle (R2 = 0.757 and %RMSE = 18.9). For the summer cycle, the best performing model was the svmP which used the first five months of the cycle as variables (R2 = 0.858 and %RMSE = 14.9). Our results indicated that adding predictor variables of the last two months before the harvest did not significantly improved model performances. These results demonstrate that our models can predict potato yield by analyzing the yield of the previous year, the general conditions of NDVI, meteorology, and information related to the irrigation system at a municipal level.

[1]  D. Timlin,et al.  Whole Plant Photosynthesis, Development, and Carbon Partitioning in Potato as a Function of Temperature , 2006 .

[2]  B. Wardlow,et al.  Analysis of time-series MODIS 250 m vegetation index data for crop classification in the U.S. Central Great Plains , 2007 .

[3]  Suxia Liu,et al.  Regional crop yield, water consumption and water use efficiency and their responses to climate change in the North China Plain , 2009 .

[4]  G. Asrar,et al.  Estimating Absorbed Photosynthetic Radiation and Leaf Area Index from Spectral Reflectance in Wheat1 , 1984 .

[5]  Gregory P. Asner,et al.  SCALE DEPENDENCE OF ABSORPTION OF PHOTOSYNTHETICALLY ACTIVE RADIATION IN TERRESTRIAL ECOSYSTEMS , 1998 .

[6]  T. Iizumi,et al.  Global crop yield forecasting using seasonal climate information from a multi-model ensemble , 2018, Climate Services.

[7]  Shu Tao,et al.  Relations between AVHRR NDVI and ecoclimatic parameters in China , 2002 .

[8]  Max Kuhn,et al.  Building Predictive Models in R Using the caret Package , 2008 .

[9]  P. Struik,et al.  Yield levels of potato crops: Recent achievements and future prospects , 2015 .

[10]  D. Spooner,et al.  Ecogeography of ploidy variation in cultivated potato (Solanum sect. Petota). , 2010, American journal of botany.

[11]  Frédéric Baret,et al.  Spectral estimates of the absorbed photosynthetically active radiation and light-use efficiency of a winter wheat crop subjected to nitrogen and water deficiencies† , 1990 .

[12]  Sujit Kumar Bala,et al.  Correlation between potato yield and MODIS‐derived vegetation indices , 2009 .

[13]  J. Peñuelas,et al.  Remote sensing of biomass and yield of winter wheat under different nitrogen supplies , 2000 .

[14]  N. Silleos,et al.  The use of multi-temporal NDVI measurements from AVHRR data for crop yield estimation and prediction , 1993 .

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

[16]  Yanhong Zhang,et al.  Coincidence of variation in potato yield and climate in northern China. , 2016, The Science of the total environment.

[17]  P. Sellers Canopy reflectance, photosynthesis and transpiration , 1985 .

[18]  B. Wardlow,et al.  A comparison of MODIS 250-m EVI and NDVI data for crop mapping: a case study for southwest Kansas , 2010 .

[19]  Dimitrios Moshou,et al.  Contribution of Remote Sensing on Crop Models: A Review , 2018, J. Imaging.

[20]  P. McCullagh,et al.  Generalized Linear Models , 1972, Predictive Analytics.

[21]  J. Monteith SOLAR RADIATION AND PRODUCTIVITY IN TROPICAL ECOSYSTEMS , 1972 .

[22]  P. Teillet Effects of spectral, spatial, and radiometric characteristics on remote sensing vegetation indices of forested regions , 1997 .

[23]  Chang Xu,et al.  Predicting Soybean Yield with NDVI Using a Flexible Fourier Transform Model , 2018, Journal of Agricultural and Applied Economics.

[24]  J. Wolf,et al.  How good is good enough? Data requirements for reliable crop yield simulations and yield-gap analysis , 2015 .

[25]  David B. Lobell,et al.  Remote sensing of regional crop production in the Yaqui Valley, Mexico: estimates and uncertainties , 2003 .

[26]  M. van der Velde,et al.  Performance of the MARS-crop yield forecasting system for the European Union: Assessing accuracy, in-season, and year-to-year improvements from 1993 to 2015 , 2019, Agricultural systems.

[27]  Anika Tahsin,et al.  Yield Prediction Model for Potato Using Landsat Time Series Images Driven Vegetation Indices , 2018, Remote Sensing in Earth Systems Sciences.

[28]  R. Hijmans,et al.  DIVA-GIS version 1.4 : A geographic information system for the analysis of biodiversity data, manual. , 2001 .

[29]  F. Hall,et al.  Global Crop Forecasting , 1980, Science.

[30]  Roberto Benedetti,et al.  On the use of NDVI profiles as a tool for agricultural statistics: The case study of wheat yield estimate and forecast in Emilia Romagna , 1993 .

[31]  H. Dolman,et al.  Terrestrial essential climate variables for climate change assessment, mitigation and adaptation (GTOS 52) , 2008 .

[32]  B. Rudolf,et al.  World Map of the Köppen-Geiger climate classification updated , 2006 .

[33]  Bernhard Schölkopf,et al.  Comparing support vector machines with Gaussian kernels to radial basis function classifiers , 1997, IEEE Trans. Signal Process..

[34]  Salah Sukkarieh,et al.  Machine learning approaches for crop yield prediction and nitrogen status estimation in precision agriculture: A review , 2018, Comput. Electron. Agric..

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

[36]  J. Monteith Climate and the efficiency of crop production in Britain , 1977 .

[37]  C. Hutchinson,et al.  Uses of satellite data for famine early warning in sub-Saharan Africa , 1991 .

[38]  James W. Merchant,et al.  An assessment of AVHRR/NDVI-ecoclimatological relations in Nebraska, U.S.A. , 1997 .

[39]  Michele Meroni,et al.  Investigating the Relationship between the Inter-Annual Variability of Satellite-Derived Vegetation Phenology and a Proxy of Biomass Production in the Sahel , 2014, Remote. Sens..

[40]  F. Maselli,et al.  Enrichment of land‐cover polygons with eco‐climatic information derived from MODIS NDVI imagery , 2009 .

[41]  J. Lichtenberger,et al.  Yield estimation for corn and wheat in the Hungarian Great Plain using Landsat MSS data , 1996 .

[42]  A. Devaux,et al.  Potatoes for Sustainable Global Food Security , 2014, Potato Research.

[43]  M. Bindi,et al.  A simple model of regional wheat yield based on NDVI data , 2007 .

[44]  Craig S. T. Daughtry,et al.  Errors in measuring absorbed radiation and computing crop radiation use efficiency , 1993 .