The capability of very high-resolution satellite imagery for plot-level early wheat stem rust disease detection, monitoring, and phenotyping in Ethiopia
Abstract
Very high‑resolution satellites (VHRS) have potential for early crop disease detection and enhanced food security. The capability of multispectral SkySat and Pleiades‑Neo imagery for early-stage detection and monitoring was assessed for wheat stem rust (SR) at the plot level. In randomized trials with contrasting fungicide and irrigation treatments, six bread wheat varieties with differing SR susceptibility were monitored using VHRS. 113 multispectral features were evaluated for their association with SR progression and associated yield loss. Several features demonstrated moderate to strong correlations with SR disease levels. Across early stages (healthy, mildly diseased, and moderately diseased), spectral sensitivity was dominated by Blue (B)‑based features, with Red-Blue (R-B) and Green-Blue (G-B) two-band features at mildly and moderately diseased levels, respectively. During late stages (severely diseased and fully diseased), spectral sensitivity was driven by R-based features (e.g., R-B, R-G on both sensors) and by Deep Blue (DB), Red-Edge-DB (RE-DB), and R-DB combinations on Pleiades‑Neo. Early-stage SR detection under moderate disease pressure was possible using ratio (RSI) and normalized difference (NDSI) spectral indices with B-G combinations. SkySat features such as RSI(NIR,B), RSI(R,B), and RSI(G,B) were consistently associated with SR across multiple disease scenarios, trials, and disease metrics. This work delivers the first VHRS-based SR detection at the plot scale and provides a foundation for future investigations into SR monitoring across larger spatial scales.
Related
- Detection of Red Crown Rot of Soybean in Illinois Fields Using High-Resolution Satellite Imagery and Machine Learning
- Reliable detection and continuous monitoring of memory dysfunction in mild cognitive impairment and healthy aging through adaptive computational phenotyping
- Monitoring Carbon Hotspots in Patchy Landscapes from Sparse Lidar and Dense Satellite Time Series
- From Plant Detection to Satellite Mapping: A Multi-Scale AI Toolkit for Ragweed Surveillance under Climate Change
- High-Resolution and Multimodal Structural Phenotyping of Intermediate and Advanced Non-Neovascular Age-Related Macular Degeneration
BibTeX
@article{blasch2026capability,
title = {{The capability of very high-resolution satellite imagery for plot-level early wheat stem rust disease detection, monitoring, and phenotyping in Ethiopia}},
author = {Gerald Blasch and Ashenafi Gemechu and Gizachew Hirpa Regasa and Yoseph Alemayehu and Adama Ndour and Pablo Olivera Firpo and Andries Potgieter and David P. Hodson},
journal = {Scientific Reports},
year = {2026},
month = oct,
doi = {10.1038/s41598-026-77529-y},
url = {https://www.nature.com/articles/s41598-026-77529-y},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - The capability of very high-resolution satellite imagery for plot-level early wheat stem rust disease detection, monitoring, and phenotyping in Ethiopia
AU - Gerald Blasch
AU - Ashenafi Gemechu
AU - Gizachew Hirpa Regasa
AU - Yoseph Alemayehu
AU - Adama Ndour
AU - Pablo Olivera Firpo
AU - Andries Potgieter
AU - David P. Hodson
PY - 2026
DA - 2026/10/28
JO - Scientific Reports
DO - 10.1038/s41598-026-77529-y
UR - https://www.nature.com/articles/s41598-026-77529-y
AB - Very high‑resolution satellites (VHRS) have potential for early crop disease detection and enhanced food security. The capability of multispectral SkySat and Pleiades‑Neo imagery for early-stage detection and monitoring was assessed for wheat stem rust (SR) at the plot level. In randomized trials with contrasting fungicide and irrigation treatments, six bread wheat varieties with differing SR susceptibility were monitored using VHRS. 113 multispectral features were evaluated for their association with SR progression and associated yield loss. Several features demonstrated moderate to strong correlations with SR disease levels. Across early stages (healthy, mildly diseased, and moderately diseased), spectral sensitivity was dominated by Blue (B)‑based features, with Red-Blue (R-B) and Green-Blue (G-B) two-band features at mildly and moderately diseased levels, respectively. During late stages (severely diseased and fully diseased), spectral sensitivity was driven by R-based features (e.g., R-B, R-G on both sensors) and by Deep Blue (DB), Red-Edge-DB (RE-DB), and R-DB combinations on Pleiades‑Neo. Early-stage SR detection under moderate disease pressure was possible using ratio (RSI) and normalized difference (NDSI) spectral indices with B-G combinations. SkySat features such as RSI(NIR,B), RSI(R,B), and RSI(G,B) were consistently associated with SR across multiple disease scenarios, trials, and disease metrics. This work delivers the first VHRS-based SR detection at the plot scale and provides a foundation for future investigations into SR monitoring across larger spatial scales.
SN - 2045-2322
ER -