2025/11/19 by Connell S. Miller, Fahim Jessa, Gregory A. Kopp · 1 voice
Earth and Planetary Sciences · Environmental Science · #Atmospheric aerosols and clouds #Meteorological Phenomena and Simulations #Remote Sensing in Agriculture
paper · doi:10.1175/jamc-d-24-0233.1
openalex created_date 2025/11/19 · openalex publication_date 2025/11/19 · openalex updated_date 2026/07/30
Abstract Tornadoes in Canada frequently impact rural regions, particularly croplands, where traditional enhanced Fujita scale (EF scale) assessments are limited due to the absence of standard damage indicators. Not knowing the true intensities of these tornadoes results in an inaccurate tornado climatology for these regions. This study evaluates the potential of the normalized difference vegetation index (NDVI), derived from high-resolution multispectral satellite imagery, as a proxy for crop damage assessment in the EF-scale framework. According to the results, NDVI percent change analysis reliably detects EF2+ damage but is largely ineffective for EF0 and EF1 tornadoes. The results also indicate that the decrease in plant health is correlated with the EF-scale rating of a tornado. Crop type and seasonal timing significantly influence NDVI detectability, with pasture and forage crops yielding weak signals and peak growing season offering the clearest damage signatures. This study also concludes that it is possible to revisit EF0-Default/EF-Unknown tornadoes that occur in agricultural areas to give them a more definitive rating on the EF scale. While NDVI analysis cannot delineate full tornado paths or replace ground surveys, it offers a scalable, remote sensing–based method to supplement tornado intensity assessments in agricultural regions. This approach may help reclassify underreported tornadoes and improve climatological accuracy in rural areas. Significance Statement Many tornadoes in rural areas go unclassified because they pass through croplands and current rating systems do not account for crop damage. This study explores a method to assess tornado strength in croplands by using satellite imagery to measure changes in plant health, through a remote sensing tool called the normalized difference vegetation index (NDVI). Looking at tornadoes in Canada from 2017 to 2023, the method was able to detect damage from stronger tornadoes but was unable to identify weaker ones. Seasonality and crop type also play a critical role in the overall results. While the results indicate that this approach cannot track the entire tornado path, it offers a promising step toward improving how we assess strong tornadoes in agricultural areas where traditional damage surveys fall short.