Unmanned aerial vehicle imagery prediction of sorghum leaf area index under water stress, seeding density, and nitrogen fertilization conditions in the Sahel

Abstract

Sahelian Africa must meet the challenge of providing enough food to meet its growing population. Therefore, novel breeding and intensive production methods are needed to mitigate this challenge. The objective of this study was to calibrate and validate sorghum varieties leaf area index (LAI) values estimated from Unmanned Aerial Vehicle (UAV) at different growing seasons in Senegal and Mali. To achieve this objective, four experiments were conducted with 14 sorghum (sorghum bicolor) varieties between 2017 and 2019. At the study sites, LAI was measured and crop reflectance was measured with a multispectral camera mounted on a UAV. The study showed that normalized difference vegetation index (NDVI) and simple ratio (SR) were highly correlated to the area index. The results of validation model revealed a better prediction of measured LAI from NDVI (R² = 0.92) and SR (R² = 0.89) vegetation indices in 2019 dry season in Senegal. In addition, the LAI predictions for Mali from NDVI (p < 0.01) and SR (p < 0.01) were highly correlated. Findings showed that vegetation indices can be used to estimate LAI in Mali and Sahel.

Plant phenotyping relevance

UAVマルチスペクトル画像と植生指数からソルガムのLAIを推定する手法を較正・検証しており、植物形質取得が研究の中心です。

Code and data availability

The supplied blocks describe UAV image processing (Agisoft PhotoScan, QGIS, R) and LAI/NDVI/SR regression analyses, but contain no data availability statement, no deposited phenotype dataset, no author code repository, and no public URL for paper-specific assets. All URLs in the text are citations, the article's own下载/

No evidence-backed public reproduction asset is currently recorded.