Multi-temporal NDVI analysis using UAV images of tree crowns in a northern Mexican pine-oak forest

The use of unmanned aerial vehicles (UAV) for forest monitoring has grown significantly in recent years, providing information with high spatial resolution and temporal versatility. UAV with multispectral sensors allow the use of indexes such as the normalized difference vegetation index (NDVI), which determines the vigor, physiological stress and photosynthetic activity of vegetation. This study aimed to analyze the spectral responses and variations of NDVI in tree crowns, as well as their correlation with climatic factors over the course of one year. The study area encompassed a 1.6-ha site in Durango, Mexico, where Pinus cembroides, Pinus engelmannii, and Quercus grisea coexist. Multispectral images were acquired with UAV and information on meteorological variables was obtained from NASA/POWER database. An ANOVA explored possible differences in NDVI among the three species. Pearson correlation was performed to identify the linear relationship between NDVI and meteorological variables. Significant differences in NDVI values were found at the genus level (Pinus and Quercus), possibly related to the physiological features of the species and their phenology. Quercus grisea had the lowest NDVI values throughout the year which may be attributed to its sensitivity to relative humidity and temperatures. Although the use of UAV with a multispectral sensor for NDVI monitoring allowed genera differentiation, in more complex forest analyses hyperspectral and LiDAR sensors should be integrated, as well other vegetation indexes be considered.

Automated summary

The use of unmanned aerial vehicles (UAV) with multispectral sensors for forest monitoring has increased in recent years. This study analyzed the spectral responses and variations of the normalized difference vegetation index (NDVI) in tree crowns, as well as their correlation with climatic factors. Significant differences in NDVI values were found among different tree species, possibly due to their physiological features and phenology. Quercus grisea had the lowest NDVI values throughout the year, which may be attributed to its sensitivity to relative humidity and temperatures. In more complex forest analyses, the integration of hyperspectral and LiDAR sensors, as well as considering other vegetation indexes, should be considered.