We observed consistently lower live sponge cover along cable transects and at cable index sites compared to control transects and sites. Live sponge cover was the lowest (decreased by 55%) at cable index sites one and a half years after cable installation. In other words, the magnitude of impact on glass sponges was not apparent immediately after cable installation: maximal effect was observed over a year after the physical disturbance. Sponges recovered to 85 ± 30.6% of the original percent cover over the following two years, suggesting that full recovery would take several years but is potentially possible. Recovery appeared to have occurred through a combination of regrowth and recruitment. The lack of statistically significant differences in regression slopes at cable and control sites during this growth/recovery period suggests that recovery rate at sites affected by the cables was comparable to natural rates of sponge reef expansion.