|
Article information
2019 , Volume 24, ¹ 6, p.125-133
Stepanov A.S.
Forecasting of crop yields based on Earth remote sensing data (using soybeans as an example)
Purpose. Develop and describe a general approach to forecasting crop yields (using soybeans as an example). Methodology. Crop yields were estimated using regression models. Values of the vegetative index (NDVI) were considered with Vega-Science system. The normalized NDVI values were approximated by the Gauss function using the Levenberg—Marquardt algorithm to enable early prediction with Python language. Findings. Values of the normalized index were determined by the preceding fiveyears period. For normalized values, approximating Gaussians were constructed and the parameters of the Gaussian function were calculated. The maximum was predicted for the NDVI values at various calendar weeks of the simulated year. The maximum values of NDVI composites in 2009–2018 were accounted for 30–32 calendar weeks. According to the simulation results, it was found that the average absolute error in predicting the maximum NDVI for 10 years at the weeks 29–32 did not exceed 3%, for weeks 27–28 — 4% and for the weeks 21–26 — 7%. At the next stage, a regression model was built to predict yield, where the calculated NDVI maximum was used as an independent variable, and soybean yield calculated according to the statistics of Rosstat on sown areas and gross soybean harvest in the region acted as an independent variable. Analysis of the error in predicting soybean yield for 2018 was obtained according to the simulation results of 2009–2017. It was shown that the absolute forecast error when using the data of 22–32 calendar weeks of 2018 did not exceed 9.1%. Originality/Value. The proposed approach to determining crop yields demonstrate high accuracy, while the method provides the possibility of early forecasting. The use of Earth remote sensing data and developed software modules of Python contribute to the operational formation of the forecast and, accordingly, the possibility of adjusting the agricultural plans.
[full text] [link to elibrary.ru]
Keywords: vegetation index, Gauss function, arable land, forecasting, yield, soybeans, Python
doi: 10.25743/ICT.2019.24.6.015.
Author(s):
Stepanov Alexey Sergeevich
Dr.
Position: Head of Laboratory
Office: Far Eastern Research Institute of Agriculture
Address: 680521, Russia, Khabarovsk, Clubnaya str., 13
E-mail: stepanxx@mail.ru
SPIN-code: 1936-5529 References:
[1] Yakushev, V.P., Dubenok, N.N., Loupian, E.A. Earth remote sensing technologies for agriculture: application experience and development prospects. Current Problems in Remote Sensing of the Earth from Space. 2019; 16(3):11–23. (In Russ.)
[2] Panesh, A.Kh., Tsalov, G.V. Prediction of winter wheat productivity on the basis of geographic information system services. Vestnik AGU. 2017; (4):175–180. (In Russ.)
[3] Kussul, N.N., Kravchenko, A.N., Skakun, S.V., Adamenko, T.I., Shelestov, A.Yu., Kolotii, A.V., Gripich, Yu.A. Crop yield forecasting regression models based on MODIS data. Current Problems in Remote Sensing of the Earth from Space. 2019; 9(1):95–107. (In Russ.)
[4] Strashnaya, A.I., Tarasova, L.L., Maksimenkova, T.A., Bogomolova, N.A., Bereza, O.V. Forecasting of the yield for grain and leguminous crops in the Central Chernozem Regions based on integration of land and satellite data. Trudy gidrometeorologicheskogo nauchno-issledovatel''skogo tsentra rossiyskoy federatsii. 2015; (353):128–153. (In Russ.)
[5] Savin, I.Yu., Chendev, Yu.G. Reasons for long-term dynamics of NDVI (MODIS) averaged for arable lands of municipalities of Belgorod region // Current Problems in Remote Sensing of the Earth from Space. 2018. Vol. 15, No. 2. P. 137–143. (In Russ.)
[6] Bukhovets, E.G., Semin, E.A., Kostenko, E.I., Iablonovskaia, S.I. Simulation of the dynamics of the NDVI of winter wheat in the conditions of the Central Federal District. Vestnik of Voronezh state agrarian university. 2018; 11(2):186–199. (In Russ.)
[7] Tolpin, V.A., Loupian, E.A., Bartalev, S.A., Plotnikov, D.E., Matveev, A.M. Possibilities of agricultural vegetation condition analysis with the VEGA satellite service. Atmospheric and Oceanic Optics. 2014; 27(7):581–586. (In Russ.)
[8] Sirant, O.V. Use big data technology for sales forecasting. Tekhnicheskie nauki — ot teorii k praktike. Using Big Data Technology to forecast sales. 2016; (12):33–40. (In Russ.)
[9] Sheppard, K. Introduction to Python for econometrics, statistics and Data Analysis. Oxford: University of Oxford; 2018: 427.
[10] Gavin, H.P. The Levenberg–Marquardt method for nonlinear least squares curve-fitting problems. Dep. Civ. Environ. Eng. Duke Univ. 2019. Available at: http://people.duke.edu/ ~hpgavin/ce281/lm.pdf (accessed 23.09.2019).
Bibliography link:
Stepanov A.S. Forecasting of crop yields based on Earth remote sensing data (using soybeans as an example) // Computational technologies. 2019. V. 24. ¹ 6. P. 125-133
|
