Rice yield prediction model using normalized vegetation and water indices from Sentinel-2A satellite imagery datasets

Aung Myint Htun, Md. Shamsuzzoha and Tofael Ahamed ()
Additional contact information
Aung Myint Htun: University of Tsukuba
Md. Shamsuzzoha: University of Tsukuba
Tofael Ahamed: University of Tsukuba

Asia-Pacific Journal of Regional Science, 2023, vol. 7, issue 2, No 9, 519 pages

Abstract: Abstract Yield predictions prior to harvesting crops is significant for agricultural decision-making. This study aimed to predict rice yield at the stage prior to harvesting using crops and soil phenological properties in the Pathein District of Myanmar. Remote sensing imagery data derived from Sentinel-2A satellite imageries during the month of November at the stage prior to harvest of rice fields were collected and analyzed from 2016 to 2021. Four vegetation indices (VIs): (i) normalized difference vegetation index (NDVI), (ii) normalized difference water index (NDWI), (iii) soil-adjusted vegetation index (SAVI), and (iv) rice growth vegetation index (RGVI) were specified as independent variables for a rice yield prediction model, after which simple and multiple linear regression models were estimated and validated. The accuracy of the estimated models was assessed using observed data from 1790 ground reference points (GRPs) in rice-yielding croplands. The average observed rice yield over 6 years was 1.57 tons per acre, and the average rice yield predictions over 6 years were 1.28, 1.48, 1.28, and 1.17 per acre with simple linear regression models from NDVI, NDWI, SAVI and RGVI, respectively. On the other hand, THE observed rice yield was 1.49 tons per acre with a multiple regression model. This indicates that prediction by the multiple regression model with four vegetation indices is superior to predictions by all other linear regression models. The early predicted yield data is useful for rice-growing farmers to compare expenses against losses after any extreme climatic event.

Keywords: Rice crop; Linear regression; Multiple regression; Myanmar; Satellite remote sensing; Yield prediction (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s41685-023-00299-2 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:apjors:v:7:y:2023:i:2:d:10.1007_s41685-023-00299-2

Ordering information: This journal article can be ordered from
https://www.springer ... cience/journal/41685

DOI: 10.1007/s41685-023-00299-2

Access Statistics for this article

Asia-Pacific Journal of Regional Science is currently edited by Yoshiro Higano

More articles in Asia-Pacific Journal of Regional Science from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().