||Experimental trampling was conducted in a coastal sand dune community located in a State Park in Southeastern, Brazil. To evaluate the effects it was used 5 permanent plots. The trampling intensities were 25, 75, 200, 500 and 1,000 passes and each plot had a control area. Response to trampling was assessed by determining species composition, vegetation cover and height evaluation, right after trampling, 2 weeks, 3 months, 6 months, 9 months and 1 year after trampling. The same parameters were evaluated just before trampling. Although there was a substantial loss of vegetation cover on 500 and 1,000 treatment plots, the study area was recovered in a few months. Reductions in height occurred with less impact: 200 passes. The 500 and 1,000 pass interference didn’t show statistical difference for cover and 200, 500 or 1,000 passes were statistically similar for the community, suggesting weak linearity between impact and amount of use. The results show that this community has a good resilience; probably in response to the natural stress the vegetation suffers continuously. The vegetation’s changes in species composition seem to be more accurate; especially the extinction of rare species and the introduction of alien ones.