A Leaf Chlorophyll Content Dataset for Crops: A Comparative Study Using Spectrophotometric and Multispectral Imagery Data

This paper presents a dataset for a comparative analysis of direct (spectrophotometric) and indirect (multispectral imagery-based) methods for quantifying crop leaf chlorophyll content. The dataset originates from a study conducted in the Department of Cauca, Colombia, a region characterized by diverse agricultural production. Data collection focused on seven economically important crops, namely coffee (Coffea arabica), Hass avocado (Persea americana), potato (Solanum tuberosum), tomato (Solanum lycopersicum), sugar cane (Saccharum officinarum), corn (Zea mays) and banana (Musa paradisiaca). Sampling was conducted across various locations and phenological stages (healthy, wilted, senescent), with each leaf subdivided into six sections (A–F) to facilitate the analysis of intra-leaf chlorophyll distribution. Direct measurements of leaf chlorophyll content were obtained by laboratory spectrophotometry following the method of Jeffrey and Humphrey, allowing for the determination of chlorophyll A and B content. Simultaneously, indirect estimates of leaf chlorophyll content were obtained from multispectral images captured at the leaf level using a MicaSense Red-Edge camera under controlled illumination. A set of 32 vegetation indices was then calculated from these multispectral images using MATLAB. Both direct and indirect methods were applied to the same leaf samples to allow for direct comparison. The dataset, provided as an Excel (.xlsx) file, comprises raw data covering laboratory-measured chlorophyll A and B content and calculated values for the 32 vegetation indices. Each row of the tabular dataset represents an individual leaf sample, identified by plant species, leaf identifier, and phenological stage. The resulting dataset, containing 16,660 records, is structured to support research evaluating the direct relationship between spectrophotometric measurements and multispectral image-based vegetation indices for estimating leaf chlorophyll content. Spearman’s correlation coefficient reveals significant positive relationships between leaf chlorophyll content and several vegetation indices, highlighting its potential for a nondestructive assessment of this pigment. Therefore, this dataset offers significant potential for researchers in remote sensing, precision agriculture, and plant physiology to assess the accuracy and reliability of various vegetation indices in diverse crops and conditions, develop and refine chlorophyll estimation models, and execute meta-analyses or comparative studies on leaf chlorophyll quantification methodologies.