Prediction of Winter Wheat Parameters with Planet SuperDove Imagery and Explainable Artificial Intelligence
This study investigated the application of high-resolution satellite imagery from SuperDove satellites combined with machine learning algorithms to estimate the spatiotemporal variability of some winter wheat parameters, including the relative leaf chlorophyll content (RCC), relative water content (...
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Agronomy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4395/15/1/241 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589373042851840 |
|---|---|
| author | Gabriele De Carolis Vincenzo Giannico Leonardo Costanza Francesca Ardito Anna Maria Stellacci Afwa Thameur Sergio Ruggieri Sabina Tangaro Marcello Mastrorilli Nicola Sanitate Simone Pietro Garofalo |
| author_facet | Gabriele De Carolis Vincenzo Giannico Leonardo Costanza Francesca Ardito Anna Maria Stellacci Afwa Thameur Sergio Ruggieri Sabina Tangaro Marcello Mastrorilli Nicola Sanitate Simone Pietro Garofalo |
| author_sort | Gabriele De Carolis |
| collection | DOAJ |
| description | This study investigated the application of high-resolution satellite imagery from SuperDove satellites combined with machine learning algorithms to estimate the spatiotemporal variability of some winter wheat parameters, including the relative leaf chlorophyll content (RCC), relative water content (RWC), and aboveground dry matter (DM). The research was carried out within an experimental field in Southern Italy during the 2024 growing season. Different machine learning (ML) algorithms were trained and compared using spectral band data and calculated vegetation indices (VIs) as predictors. Model performance was assessed using R<sup>2</sup> and RMSE. The ML models tested were random forest (RF), support vector regressor (SVR), and extreme gradient boosting (XGB). RF outperformed the other ML algorithms in the prediction of RCC when using VIs as predictors (R<sup>2</sup> = 0.81) and in the prediction of the RWC and DM when using spectral bands data as predictors (R<sup>2</sup> = 0.71 and 0.87, respectively). Model explainability was assessed with the SHAP method. A SHAP analysis highlighted that GNDVI, Cl1, and NDRE were the most important VIs for predicting RCC, while yellow and red bands were the most important for DM prediction, and yellow and nir bands for RWC prediction. The best model found for each target was used to model its seasonal trend and produce a variability map. This approach highlights the potential of integrating ML and high-resolution satellite imagery for the remote monitoring of wheat, which can support sustainable farming practices. |
| format | Article |
| id | doaj-art-010a55552aee4d9a9d7b81ef550562fe |
| institution | Kabale University |
| issn | 2073-4395 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj-art-010a55552aee4d9a9d7b81ef550562fe2025-01-24T13:17:18ZengMDPI AGAgronomy2073-43952025-01-0115124110.3390/agronomy15010241Prediction of Winter Wheat Parameters with Planet SuperDove Imagery and Explainable Artificial IntelligenceGabriele De Carolis0Vincenzo Giannico1Leonardo Costanza2Francesca Ardito3Anna Maria Stellacci4Afwa Thameur5Sergio Ruggieri6Sabina Tangaro7Marcello Mastrorilli8Nicola Sanitate9Simone Pietro Garofalo10Council for Agricultural Research and Economics, Research Center for Agriculture and Environment, Via Celso Ulpiani, 5, 70125 Bari, ItalyDepartment of Soil, Plant and Food Science, University of Bari Aldo Moro, Via Amendola 165/A, 70126 Bari, ItalyDepartment of Soil, Plant and Food Science, University of Bari Aldo Moro, Via Amendola 165/A, 70126 Bari, ItalyDepartment of Soil, Plant and Food Science, University of Bari Aldo Moro, Via Amendola 165/A, 70126 Bari, ItalyDepartment of Soil, Plant and Food Science, University of Bari Aldo Moro, Via Amendola 165/A, 70126 Bari, ItalyLaboratory of Biodiversity, Molecules, Application (BMA), Higher Institute of Applied Biology Medenine, University of Gabes, 4119 Medenine, TunisiaCouncil for Agricultural Research and Economics, Research Center for Agriculture and Environment, Via Celso Ulpiani, 5, 70125 Bari, ItalyDepartment of Soil, Plant and Food Science, University of Bari Aldo Moro, Via Amendola 165/A, 70126 Bari, ItalyCouncil for Agricultural Research and Economics, Research Center for Agriculture and Environment, Via Celso Ulpiani, 5, 70125 Bari, ItalyCouncil for Agricultural Research and Economics, Research Center for Agriculture and Environment, Via Celso Ulpiani, 5, 70125 Bari, ItalyCouncil for Agricultural Research and Economics, Research Center for Agriculture and Environment, Via Celso Ulpiani, 5, 70125 Bari, ItalyThis study investigated the application of high-resolution satellite imagery from SuperDove satellites combined with machine learning algorithms to estimate the spatiotemporal variability of some winter wheat parameters, including the relative leaf chlorophyll content (RCC), relative water content (RWC), and aboveground dry matter (DM). The research was carried out within an experimental field in Southern Italy during the 2024 growing season. Different machine learning (ML) algorithms were trained and compared using spectral band data and calculated vegetation indices (VIs) as predictors. Model performance was assessed using R<sup>2</sup> and RMSE. The ML models tested were random forest (RF), support vector regressor (SVR), and extreme gradient boosting (XGB). RF outperformed the other ML algorithms in the prediction of RCC when using VIs as predictors (R<sup>2</sup> = 0.81) and in the prediction of the RWC and DM when using spectral bands data as predictors (R<sup>2</sup> = 0.71 and 0.87, respectively). Model explainability was assessed with the SHAP method. A SHAP analysis highlighted that GNDVI, Cl1, and NDRE were the most important VIs for predicting RCC, while yellow and red bands were the most important for DM prediction, and yellow and nir bands for RWC prediction. The best model found for each target was used to model its seasonal trend and produce a variability map. This approach highlights the potential of integrating ML and high-resolution satellite imagery for the remote monitoring of wheat, which can support sustainable farming practices.https://www.mdpi.com/2073-4395/15/1/241chlorophyll contentrelative water contentdry matterremote sensingplanetmachine learning |
| spellingShingle | Gabriele De Carolis Vincenzo Giannico Leonardo Costanza Francesca Ardito Anna Maria Stellacci Afwa Thameur Sergio Ruggieri Sabina Tangaro Marcello Mastrorilli Nicola Sanitate Simone Pietro Garofalo Prediction of Winter Wheat Parameters with Planet SuperDove Imagery and Explainable Artificial Intelligence Agronomy chlorophyll content relative water content dry matter remote sensing planet machine learning |
| title | Prediction of Winter Wheat Parameters with Planet SuperDove Imagery and Explainable Artificial Intelligence |
| title_full | Prediction of Winter Wheat Parameters with Planet SuperDove Imagery and Explainable Artificial Intelligence |
| title_fullStr | Prediction of Winter Wheat Parameters with Planet SuperDove Imagery and Explainable Artificial Intelligence |
| title_full_unstemmed | Prediction of Winter Wheat Parameters with Planet SuperDove Imagery and Explainable Artificial Intelligence |
| title_short | Prediction of Winter Wheat Parameters with Planet SuperDove Imagery and Explainable Artificial Intelligence |
| title_sort | prediction of winter wheat parameters with planet superdove imagery and explainable artificial intelligence |
| topic | chlorophyll content relative water content dry matter remote sensing planet machine learning |
| url | https://www.mdpi.com/2073-4395/15/1/241 |
| work_keys_str_mv | AT gabrieledecarolis predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence AT vincenzogiannico predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence AT leonardocostanza predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence AT francescaardito predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence AT annamariastellacci predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence AT afwathameur predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence AT sergioruggieri predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence AT sabinatangaro predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence AT marcellomastrorilli predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence AT nicolasanitate predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence AT simonepietrogarofalo predictionofwinterwheatparameterswithplanetsuperdoveimageryandexplainableartificialintelligence |