The NASA Harvest Program on Agriculture and Food Security

Becker‐Reshef, Inbal; Bandaru, Varaprasad; Barker, B.; Coutu, Sylvain; Deines, Jillian M.; Doorn, Bradley; Eilerts, Gary; Franch, B.; Galvez, Antonio Sanchez; Hosseini, Mehdi; Humber, M. L.; Husak, G. J.; Guan, Kaiyu; Justice, C. O.; Keniston, John; Kerner, Hannah; Mitkish, Mary; Mobley, Kara; Munshell, Blake; Nakalembe, Catherine; Puricelli, Estefania; Sahajpal, Ritvik; Skakun, Sergii; Vermote, Éric; Whitcraft, A. K.; Hansen, Matthew C.; Salas, Bill; Justice, Chris

doi:10.1007/978-3-030-92365-5_3

Cited by 1 publication

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Obtaining similar crop cover data can be challenging in regions like Malawi and South Africa due to resource, infrastructure, and technology limitations. Becker-Reshef et al [2] partly addressed this global data gap by providing crop-type maps that illustrate the spatial distribution of major crops at a 0.05-degree resolution. Each pixel in these rasters is associated with a percentage representing the area covered by a particular crop type.…”

Section: Methodsmentioning

confidence: 99%

Leveraging satellite data with machine and deep learning techniques for corn yield and price forecasting

Teste,

Gangloff,

Chen

et al. 2024

Preprint

View full text Add to dashboard Cite

This study introduces an innovative method for forecasting corn yield and price variations, critical for food security and strategic planning. Unlike traditional methods reliant on pre-harvest production data, which are often difficult to access, our approach utilizes satellite-derived gross primary production (GPP) data and dimension-reduction techniques to predict national corn yield and price changes. We conducted case studies in the US, Malawi, and South Africa to validate this approach, extracting predictors from annual GPP variations during peak growing seasons. We utilized various dimension reduction strategies, including spatial averaging, Empirical Orthogonal Functions (EOFs), and deep learning approaches like Autoencoders (AEs) and Variational Autoencoders (VAEs), to extract meaningful features from the GPP datasets. These extracted features serve as predictors in statistical models such as Generalized Linear Models and the Least Absolute Shrinkage and Selection Operator (LASSO), along with a neural network trained to predict variations directly from GPP-derived latent features. Model performances were evaluated using the Area Under Curve, Brier Skill Score, and Matthew Correlation Coefficient. Our results indicate that neural network models based on both AEs and VAEs exhibit superior predictive capabilities across all three countries, with the VAE excelling in the US and the AE leading in South Africa and Malawi. Similarly, the VAE outperforms other approaches for price prediction in the US and South Africa, while the AE achieves the best results in Malawi.Our study demonstrates the potential of combining satellite data and dimension reduction methods to significantly enhance large-scale corn yield and price predictions months before harvest.

show abstract

Section: Methodsmentioning

confidence: 99%