Dugongs and green turtles: grazers in the tropical seagrass ecosystem

This study examined aspects of the interactions between dugongs, green turtles and their tropical seagrass food. In order to examine the effects of herbivory on the community structure, productivity, and nutritional composition of seagrass, experiments simulating intensive and light dugong grazing (uprooting whole plants) and intensive turtle cropping (removal of aboveground biomass) were carried out in intertidal seagrass beds at Cardwell (18°14'S, 146°E) and Ellie Point (16 ° 53 S, 145 ° 46 ' E) on the northeast Queensland coast. Grazing experiments at Cardwell and Ellie Point were monitored monthly for a year before the seagrass samples were harvested. An additional short-term experiment was also carried out at Cardwell only, wherein samples were harvested one month and two months after cropping, while those from the grazing plots were harvested after four months. Seagrasses were harvested opportunistically from eight sites and from four depths at one site to investigate specific and spatial variation in nutrient composition. The effect of artificial nitrogen and phosphorus fertiliser treatments on seagrass nutrients was investigated experimentally at Shelley Beach (19°19'S, 146°50'E). Determinants of the nutritional composition of tropical seagrasses and the nutritional basis of the observed feeding preference of these herbivores were also considered.\ud \ud Two techniques were used in seagrass ecology for the first time:\ud \ud (1) Video recording was used for monitoring temporal changes in the species composition and abundance in tropical seagrass communities;\ud \ud (2) Near infra-red reflectance spectroscopy (NIRS) was used to measure the concentrations of the following: nitrogen, organic matter, neutral detergent fibre, acid detergent fibre, lignin, water soluble carbohydrate, and starch and in vitro digestibility of dry matter.\ud \ud The development of the NIRS technique involved the collection of 10 species of seagrasses: Halophila ovalis, H minor, H spinulosa, H decipiens, H. trichostata, Halodule uninervis, Cymodocea serrulata, C. rotundata, Syringodium isoetifolium, and Zostera capricorni (with H uninervis and Z capricorni exhibiting two varieties). From this collection, a seagrass database consisting of 1,165 samples of leaves (n = 556), roots/rhizomes (n = 552), whole plant (n = 11), seeds (n = 3), and detrital matter (n= 43), including the samples from the grazing experiments, was developed. Then, using NIRS, the spectra of all samples were collected. From this spectral population, some 200 spectra representative of the whole population were selected, using a computer algorithm package (NIRS 3) as the calibration set and prediction equations (multivariate models) developed for the above seagrass nutritional components.\ud \ud The nature and extent of the effects of grazing and cropping were related to:\ud \ud (1) the intensity of the grazing impact; and\ud \ud (2) the nature of the seagrass community, including its species composition and location.\ud \ud In a mixed-species bed at Ellie Point, intensive grazing altered the species composition by promoting the growth of a more opportunistic (short-lived) species, Halophila ovalis in the spaces created by the grazing disturbance at the expense of a long-lived species, Zostera capricorni. Grazing also reduced the amounts of detrital matter. The species composition of a monospecific bed of Halodule uninervis was not affected by grazing. Both light and intensive grazing, and cropping increased the net above-ground biomass productivity of H. ovalis and Halodule uninervis. Recovery times varied from months for H. ovalis and Zostera/Cymodocea at Ellie Point to more than one year for H uninervis at Cardwell. In both cases, grazing improved the seagrass bed as grazing habitat for dugongs and green turtles.\ud \ud Simulated dugong grazing improved the nutritional composition (nitrogen and water soluble carbohydrate) of H. ovalis and H uninervis. This improvement was detectable 10 to 12 months later. In short-term experiments, both grazing and cropping increased the leaf nitrogen concentration of H uninervis. The digestibility (in vitro) of dry matter of H uninervis moderately increased after grazing and cropping. Grazing and cropping had variable effects on the fibre and lignin contents of H. uninervis depending on the plant part, nature and intensity of herbivory and duration of the recovery. Enhanced nutrients in the sediments increased the concentrations of nitrogen, starch, and fibre of H minor and H uninervis.\ud \ud The nutritional composition of seagrasses also varied between plant parts, among species, between varieties, among depths, and among locations (sites). Halophila species, together with Syringodium isoetifolium, were more digestible than Z. capricorni, C. serrulata, and C. rotundata, while Halodule uninervis had the highest nitrogen and starch concentrations of any of the species.\ud \ud Dugongs and green turtles appear to optimise their diet by selecting food species that maximise digestible nutrients. This is achieved by selecting seagrass species that are more digestible and have higher nutrients (e.g. nitrogen and carbohydrates/starch) and/or species which can compensate for grazing.\ud \ud Changes in feeding habitats due to herbivory by dugongs and green turtles affect the functional dynamics of tropical seagrass ecosystems through the alteration of resource availability and sediment redox conditions resulting from grazing disturbance. Consequently, mosaics of patches of varying species and nutritional compositions are produced at a local scale.\ud \ud A major and long-term reduction in the number of dugongs and green turtles in some areas may lead to an irreversible degradation of their habitats as preferred food species are replaced by less-preferred species. In other areas, other forms of natural disturbance and environmental constraints probably maintain the community at a low seral stage