Monitoring of plant phenological traits in a mixed Mediterranean forest species based on colour vegetation indices

Phenology reflect the influences of climate change on individual plant biodiversity. Meteorological factors (temperature, precipitation, sunshine period) are reported to have major influence for determining variation of time phenophases through changes in biological activities.

Functional traits are morphological, physiological, and phenological features that show the plant functional response to a specific environmental condition. Traits are therefore connected to physiological functions and mechanisms that allow species to face the climatic changes.

Phenological traits can help explaining variation in plant growth strategies as it is closely associated with carbon gain and nutrient conservation. The beginning of leaf emergence and senescence can maximize plant survival and productivity, as the duration of stages are closely related to resource acquisition and distribution. Intraspecific trait variation, complementarity, and environmental stress are investigated at several hierarchical levels with the final goal of identify the mechanisms through which individuals and species interact and coexist for the use of resources. Therefore, it is fundamental to understanding how species modulate their phenological traits.

In this work, phenological traits for 12 forest species are monitored and studied. The research was conducted on an experimental site in Sardinia, Italy. A selection of monocultures and mixed communities, with different levels of species richness and functional diversity, were under observation over a 2-years period.

Phenological traits were investigated by using automated monitoring systems for recording high-resolution digital images in the visible spectrum. Start and duration of the growing season were monitored to acquire information regarding the plant status in relation to environmental conditions (temperature/drought). The images were collected at daily interval during the period 2017-2019. Six systems were set-up and installed on a metal pole, at 2.5 m height above ground.

The phenological traits for each individual were determined by analysing images visually. Moreover, a digital image processing, for deriving colour vegetation indices, was performed using a routine specifically developed under the software application MATLAB. RGB color channel information were recorded as digital numbers (DN) in a region of interest (ROI). To reduce the effects of scene illumination, the DNs are converted in to chromatic coordinates. Daily values of vegetation indices (ExG, REI, GRVI, etc) were then calculated.

The pattern of vegetation indices during the growing season were analysed for each species, showing how the color index performance works in capturing vegetation phenological change, and the performance of species growing under different levels of richness and functional diversity.

Information from this study can provide a valid contribution to a more detailed understanding on how plant responds to different environmental conditions, in particular in relation to phenological behaviour in response to climatic changes in Mediterranean regions.