Seagrass beds form an important part of the coastal ecosystem in many parts of the world but are very sensitive to anthropogenic nutrient increases. In the last decades, stable isotopes have been used as tracers of anthropogenic nutrient sources and to distinguish these impacts from natural environmental change, as well as in the identification of food sources in isotopic food web reconstruction. Thus, it is important to establish the extent of natural variations on the stable isotope composition of seagrass, validating their ability to act as both tracers of nutrients and food sources. Around the world, depending on the seagrass species and ecosystem, values of seagrass N normally vary from 0 to 8 ‰ δ15N. In this study, highly unusual seagrass N isotope values were observed on the east coast of Qatar, with significant spatial variation over a scale of a few metres, and with δ15N values ranging from +2.95 to −12.39 ‰ within a single bay during March 2012. This pattern of variation was consistent over a period of a year although there was a seasonal effect on the seagrass δ15N values. Seagrass, water column and sediment nutrient profiles were not correlated with seagrass δ15N values and neither were longer-term indicators of nutrient limitation such as seagrass biomass and height. Sediment δ15N values were correlated with Halodule uninervis δ15N values and this, together with the small spatial scale of variation, suggest that localised sediment processes may be responsible for the extreme isotopic values. Consistent differences in sediment to plant 15N discrimination between seagrass species also suggest that species-specific nutrient uptake mechanisms contribute to the observed δ15N values. This study reports some of the most extreme, negative δ15N values ever noted for seagrass (as low as −12.4 ‰) and some of the most highly spatially variable (values varied over 15.4 ‰ in a relatively small area of only 655 ha). These results are widely relevant, as they demonstrate the need for adequate spatial and temporal sampling when working with N stable isotopes to identify food sources in food web studies or as tracers of anthropogenic nutrients.