Moisture Stress Detection in Soybean Crops Using Sentinel-2 Time-Series NDVI and Machine Learning Techniques

Authors

  • Rahul B. Mannade

Keywords:

NDVI, NDWI, Time Series NDVI, ML

Abstract

The measurement of moisture stress in crops is necessary in enhancing agricultural productivity as well as making water resource management sustainable especially in rainfed agricultural systems. Conventional field-based approaches tend to be restricted in spatial and temporal terms, and remote sensing could be considered as an option to monitor crops on a large scale. This paper introduces a machine learning-based model of moisture stress detection in soybean plants with the help of time-series satellite data. Sentinel-2 multispectral imagery was used in the growing season. Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI) were calculated using monthly composite images to calculate vegetation and moisture indices respectively. The dataset was in pixel format with moisture stress labels determined based on NDWI values that were determined by a dynamic threshold method. To prevent leaking of data and provide model reliability, NDVI time-series characteristics were taken as input variables, whereas NDWI was applied only to label generation. A Random Forest model was used to estimate the connection between the vegetation dynamics and the moisture stress situation. The overall accuracy of the model was 93.33, which means that the model has a high predictive power. The analysis of the importance of features showed that the values of NDVI during the later growth stages and especially in September and October were the most significant to detect the stress, and crop monitoring over time is important. The findings indicate that vegetation indices are useful to monitor the patterns of moisture stress without necessarily using water-related inputs. This research establishes the possibility of combining remote sensing and machine-learning methods in an efficient and scalable crop stress monitoring. The suggested solution offers an economical solution in making agricultural decisions and can be generalized to other crops and area to apply precision farming.

Downloads

Download data is not yet available.

References

Pandey, G. (2018). Challenges and future prospects of agri-nanotechnology for sustainable agriculture in India. Environmental Technology & Innovation, 11, 299-307.

Wijewardana, C., Reddy, K. R., & Bellaloui, N. (2019). Soybean seed physiology, quality, and chemical composition under soil moisture stress. Food chemistry, 278, 92-100.

Kobayashi, N., Tani, H., Wang, X., & Sonobe, R. (2020). Crop classification using spectral indices derived from Sentinel-2A imagery. Journal of Information and Telecommunication, 4(1), 67-90.

Zheng, Y., Tang, L., & Wang, H. (2021). An improved approach for monitoring urban built-up areas by combining NPP-VIIRS nighttime light, NDVI, NDWI, and NDBI. Journal of cleaner production, 328, 129488.

Gao, J., Nuyttens, D., Lootens, P., He, Y., & Pieters, J. G. (2018). Recognising weeds in a maize crop using a random forest machine-learning algorithm and near-infrared snapshot mosaic hyperspectral imagery. Biosystems engineering, 170, 39-50.

Elavarasan, D., & Vincent, P. D. R. (2021). A reinforced random forest model for enhanced crop yield prediction by integrating agrarian parameters. Journal of Ambient Intelligence and Humanized Computing, 12(11), 10009-10022.

Jha, P. K., Kumar, S. N., & Ines, A. V. (2018). Responses of soybean to water stress and supplemental irrigation in upper Indo-Gangetic plain: Field experiment and modeling approach. Field crops research, 219, 76-86.

Liu, Y., Kumar, M., Katul, G. G., Feng, X., & Konings, A. G. (2020). Plant hydraulics accentuates the effect of atmospheric moisture stress on transpiration. Nature Climate Change, 10(7), 691-695.

Chen, J., et al. (2019). Monitoring Crop Condition Using Multi-Temporal Sentinel-2 Data. Remote Sensing, MDPI.

Zhang, X., et al. (2020). Crop Water Stress Detection Using Spectral Indices and Machine Learning. Remote Sensing, MDPI.

Xu, Y., et al. (2020). Crop Classification Using Random Forest and Multispectral Satellite Data. Sensors, MDPI.

Mandal, B., Dolui, G., & Satpathy, S. (2018). Land suitability assessment for potential surface irrigation of river catchment for irrigation development in Kansai watershed, Purulia, West Bengal, India. Sustainable Water Resources Management, 4(4), 699-714.

Ashok, A., Rani, H. P., & Jayakumar, K. V. (2021). Monitoring of dynamic wetland changes using NDVI and NDWI based landsat imagery. Remote Sensing Applications: Society and Environment, 23, 100547.

Jothimani, M., Gunalan, J., Duraisamy, R., & Abebe, A. (2021, September). Study the Relationship Between LULC, LST, NDVI, NDWI and NDBI in Greater Arba Minch Area, Rift Valley, Ethiopia. In 3rd International conference on integrated intelligent computing communication & security (ICIIC 2021) (pp. 183-193). Atlantis Press.

Chafik, H., Berrada, M., Legdou, A., Amine, A., & Lahssini, S. (2020, May). Exploitation of spectral indices NDVI, NDWI & SAVI in random forest classifier model for mapping weak rosemary cover: Application on Gourrama region, Morocco. In 2020 IEEE international conference of Moroccan geomatics (Morgeo) (pp. 1-6). IEEE.

Yacouby, R., & Axman, D. (2020, November). Probabilistic extension of precision, recall, and f1 score for more thorough evaluation of classification models. In Proceedings of the first workshop on evaluation and comparison of NLP systems (pp. 79-91).

Downloads

Published

31.12.2021

How to Cite

Rahul B. Mannade. (2021). Moisture Stress Detection in Soybean Crops Using Sentinel-2 Time-Series NDVI and Machine Learning Techniques. International Journal of Intelligent Systems and Applications in Engineering, 9(4), 493–499. Retrieved from https://ijisae.org/index.php/IJISAE/article/view/8169

Issue

Vol. 9 No. 4 (2021)

Section

Research Article

License

Creative Commons License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.

IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.