02 Aug 2022
 | 02 Aug 2022
Status: this preprint is currently under review for the journal HESS.

Spatiotemporal changes of drought area as input for a machine-learning approach for crop yield prediction

Vitali Diaz, Ahmed A. A. Osman, Gerald A. Corzo Perez, Henny A. J. Van Lanen, Shreedhar Maskey, and Dimitri Solomatine

Abstract. Climate change has increased the possibility of more severe and prolonged droughts worldwide, which requires innovative methods to predict their impacts on different sectors such as agriculture. Crop growth models calculate yield and variables related to plant development and are used for crop yield estimation, a useful variable for monitoring drought impacts. Although used for prediction, these crop models are not explicit forecasting models; they are limited to the physical assumptions reflected in their conceptual model. In addition, the input data availability, the spatial and temporal aggregation, and different sources of uncertainty make the crop yield prediction challenging. Given these limitations, machine learning (ML) models are often utilised following a multivariable forecasting approach, but their use with the spatial characteristics of droughts as input data is limited. This research explored the spatial extent of drought as input data for building an approach for predicting seasonal crop yield. This ML approach is made up of two components. The first includes polynomial regression (PR) models, and the second considers artificial neural network (ANN) models. This approach aimed to evaluate both types of ML models (PR and ANN) and integrate them into one operational tool. The logic is as follows: ANN models determine the most accurate predictions, but in practice, issues regarding data retrieval and processing can make the use of equations, i.e. PR, preferable. The proposed approach provides these PR equations with early and preliminary input to perform such calculations. The estimates can be further improved when the ANN models are run with the final input data. The results indicated that the empirical equations (PR) produced good predictions when using drought area as the input. ANN provides better estimates, in general. Research results show that the spatiotemporal changes of drought area and its temporal aggregation provide an important pre-processing alternative to implement ML models for drought impact prediction.

Vitali Diaz et al.

Status: open (until 28 Mar 2023)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on hess-2022-252', Anonymous Referee #1, 08 Jan 2023 reply

Vitali Diaz et al.

Vitali Diaz et al.


Total article views: 522 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
389 113 20 522 3 13
  • HTML: 389
  • PDF: 113
  • XML: 20
  • Total: 522
  • BibTeX: 3
  • EndNote: 13
Views and downloads (calculated since 02 Aug 2022)
Cumulative views and downloads (calculated since 02 Aug 2022)

Viewed (geographical distribution)

Total article views: 466 (including HTML, PDF, and XML) Thereof 466 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 20 Feb 2023
Short summary
Drought impacts on crops can be assessed in terms of crop yield (CY) variation. The hypothesis is that the spatiotemporal change of drought area is a good input to predict CY. A step-by-step approach for predicting CY is built based on two types of machine learning models. Drought area was found suitable for predicting CY. Since it is currently possible to calculate drought areas within drought monitoring systems, the prediction of drought impacts can be integrated directly into them.