ErikFerrara/Coral_Color_Index: Coral_color_index

RGB color indices as proxy for Symbiodiniaceae cell density and chlorophyll content during coral bleaching Coral bleaching, the breakdown of the symbiosis between the coral host and endosymbiotic microalgae, is the main cause of widespread coral reef degradation. Current visual assessment methods for coral health, including color reference cards, are constrained by subjective human color perception and limited resolution. RGB (red, green, blue) color channel analysis through digital photography provides a rapid, non-invasive, and standardized approach for estimating key physiological parameters. While image analysis is extensively used in plant studies, the relationship between coral color and physiological parameters during bleaching requires further attention. In this study, we analyzed the correlation between Symbiodiniaceae cell density and chlorophyll content across three coral species (Acropora muricata, Pocillopora verrucosa, and Stylophora pistillata) with 19 color indices currently established as predictors of chlorophyll content in plants. We exposed corals to three bleaching treatments (short-term and long-term heat stress, and menthol bleaching) and correlated physiological parameters with color indices to identify the index with the strongest linear correlation. We found that the red channel had the strongest and most consistent linear correlation with Symbiodiniaceae cell density and chlorophyll content across species (R2 values up to 0.97), enabling a direct interpretation of relative changes in channel intensity values as changes in tissue parameters. Additionally, training a model with 10 to 12 samples allows for accurate predictions of Symbiodiniaceae densities, achieving over 95% accuracy, compared to models trained on larger datasets. This research improves image analysis as a reliable, non-invasive tool for monitoring coral health using RGB data. This repository contains all raw data and the code used to extract RGB values from images and to process the data for identifying the most suitable colour index to be used as a proxy for coral health For the statistical assessment and data visualization, we used the statistical R Project for Statistical Computing Software together with the graphical user interface RStudio.