The use of the vegetative index NDVI to predict grain crop yields

• Published in: Bulletin of NSAU (Novosibirsk State Agrarian University)
• Main Authors: S. A. Rodimtsev, N. E. Pavlovskaya, S. V. Vershinin, I. V. Gorkova, I. N. Gagarina, С. А. Родимцев, Н. Е. Павловская, С. В. Вершинин, И. В. Горькова, И. Н. Гагарина
• Other Authors: Работа выполнена в рамках тематического планазадания на выполнение ФГБОУ ВО Орловским ГАУ НИР по заказу Минсельхоза России за счет средств федерального бюджета в 2021 г. (регистрационный номер НИОКТР № 121091400023-3 от 14.09.2021.).
• Format: Article in Journal/Newspaper
• Language: Russian
• Published: Publishing Centre “Zolotoy Kolos” of Novosibirsk State Agrarian University 2023
• Subjects: прогностическая модель; vegetative index; NDVI time series; grain crops; vegetation phases; UAV; mathematical data treatment; predictive model; вегетативный индекс; временные ряды NDVI; зерновые культуры; фазы вегетации; БПЛА; математическая обработка данных; Arctic
• Online Access: https://vestngau.elpub.ru/jour/article/view/1781; https://doi.org/10.31677/2072-6724-2022-65-4-56-67

The need for unified approaches to determining the phenological phase of a reliable indicator of the vegetative index is one of the critical problems of early forecasting of crop yields using satellite monitoring. Several works of domestic and foreign researchers formulate different estimates of the correlation relationship between NDVI and yield. This study aimed to obtain predictive models for the product of winter wheat and spring barley using indicators that are adequate for formalizing the tasks of predicting the trend section of the vegetative index NDVI of crops within the experimental farm of the Oryol State Agrarian University. Based on the analysis of the dynamics of the vegetation index NDVI, based on multi-year studies, the maximum mean annual values of the vegetation index, 0.72 for winter wheat and 0.56 for spring barley, were determined.The maximum NDVI values of the 2021 season for these crops are 0.78 and 0.58. It was found that the peaks of NDVI values correspond to the earing phase of crops with possible variation from 1 to 13 days. The correlation coefficients between the maximum values of NDVI and productivity of crops were 0.79 and 0.75 for winter wheat and spring barley, respectively, which suggests the possibility of reliable prediction of crop yield based on the data of their peak NDVI values. The authors obtained predictive crop yield models based on polynomial (second-degree) functions. A reliable yield forecast expands the scope of reasonable estimates and the implementation of plans aimed at the progressive development of the individual farm. Furthermore, it contributes to the food security of Russia as a whole. Одна из ключевых проблем раннего прогнозирования урожайности сельхозкультур с использованием спутникового мониторинга заключается в отсутствии единых подходов к определению фенофазы достоверного показателя вегетативного индекса. Рядом работ отечественных и зарубежных исследователей формулируются различные оценки корреляционной зависимости между показателями NDVI и ...