Near-infrared reflectance spectroscopy is a rapid, cost-effective predictor of seagrass nutrients.

Near-infrared reflectance spectroscopy was used to analyze nutrient composition of tropical and subtropical seagrasses in Queensland, Australia, as part of a broader study of impacts of grazing by dugongs on seagrass. Seagrass samples of 10 species were collected, transported to the laboratory, and separated into leaf and root/rhizome fractions. They were dried, ground, and near-infrared spectra (400-2500 nm) were collected. We used partial least-squares regression to develop calibration equations relating spectral data to standard compositional analyses performed in the laboratory. These compositional analyses focused on attributes believed to be important determinants of nutritional quality of marine vertebrate herbivores (nitrogen, organic matter, neutral detergent fiber, acid detergent fiber, lignin, neutral starch, water-soluble carbohydrates, and in vitro dry matter digestibility). Calibration equations for each attribute were developed separately for (1) roots/rhizomes and (2) leaves, irrespective of plant species. An equation that combined both plant parts was equally robust. These studies demonstrated the utility of near-infrared spectroscopy in providing rapid and cost-effective analysis of marine plants, which, in turn, permits a rigorous statistical approach to be applied to studies of foraging by marine herbivores.