Title: Estimating and understanding crop yields with explainable deep learning in the Indian Wheat Belt
Authors: WOLANIN ALEKSANDRAMATEO-GARCIA GONZALOCAMPS-VALLS GUSTAVOGOOMEZ-CHOVA LUISMERONI MICHELEDUVEILLER BOGDAN GRÉGORY HENRY ELIANGZHI YOUGUANTER LUIS
Citation: ENVIRONMENTAL RESEARCH LETTERS vol. 15 no. 2 p. 024019
Publisher: IOP PUBLISHING LTD
Publication Year: 2020
JRC N°: JRC118324
ISSN: 1748-9326 (online)
URI: https://iopscience.iop.org/article/10.1088/1748-9326/ab68ac
https://publications.jrc.ec.europa.eu/repository/handle/JRC118324
DOI: 10.1088/1748-9326/ab68ac
Type: Articles in periodicals and books
Abstract: Forecasting crop yields is becoming increasingly important under the current context in which food security needs to be ensured despite the challenges brought by climate change, an expanding world population accompanied by rising incomes, increasing soil erosion, and decreasing water resources. Temperature, radiation, water availability and other environmental conditions influence crop growth, development, and final grain yield in a complex non-linear manner, which traditional statistical methods based on linear relationships may fail to represent. Machine learning (ML) techniques, and deep learning (DL) methods in particular, can account for such non-linear relations between yield and its covariates. However, they typically lack transparency and interpretability, since the way the predictions are derived is not directly evident. Yet, in the context of yield forecasting, understanding which are the underlying factors behind both a predicted loss or gain is of great use. Here, we explore how to benefit from the increased predictive performance of DL methods without compromising our understanding of the drivers behind. To do so, we applied a deep neural network to multivariate time series of vegetation and meteorological data to estimate the wheat yield in the Indian Wheat Belt. Then, we visualized and analyzed the features and yield drivers learned by the model with the use of regression activation maps. The DL model outperformed other tested models (ridge regression and random forests) and facilitated the interpretation of variables and processes that lead to yield variability. The learned features were mostly related to the length of the growing season, temperature, and light conditions during the growing season. For example, our results showed that high yields in 2012 were associated with low temperatures accompanied by sunny conditions during the growing period. The proposed methodology can be used for other crops and regions in order to facilitate application of DL models in agriculture.
JRC Directorate:Sustainable Resources

Files in This Item:
There are no files associated with this item.


Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.