In New Zealand, black swans have been observed to graze a moderate to large proportion of the biomass of intertidal seagrass (Zostera muelleri) meadows. To evaluate the resilience of seagrass meadows to swan grazing, we performed a grazing simulation experiment at four sites in a large temperate estuary (Tauranga Harbour). In late summer, when swan grazing pressure typically peaks, we established control (0% biomass removed), low grazing intensity (~ 40% biomass removed) and high grazing intensity (~ 100% biomass removed) plots (0.25 m2) to mimic the grazing scars created by black swans. For one year, percentage seagrass above-ground cover of plots and a wider area surrounding the plots (0.75 m2) was monitored monthly. Seagrass biomass and leaf size were measured at the start and end of the experiment. Results showed no significant grazing treatment effect on seagrass leaf size (length and width). The low grazing intensity treatment had no detectable effect on seagrass cover or biomass relative to controls. For the high grazing intensity plots at all sites seagrass cover was restored within 8 months but biomass was not fully restored (only 25–42% of controls) one year later. This suggests that it may take several years (2.4–4.0, assuming a constant recovery rate) for seagrass below-ground biomass and leaf layering to regenerate completely in the grazing scars created by swans.