1 Mangrove forests are structurally complex with canopy gaps generated by p
hysical disturbances such as storms and mounds created on the forest floor
by biotic disturbances such as crabs excavating burrows.
2 I studied how variations in canopy cover and substratum topography influe
nce the recruitment and demography of the mangrove Avicennia marina in a te
mperate forest in New South Wales, Australia.
3 Surveys showed that mounds covered up to 44% of the forest floor and were
larger and covered more of the substratum under the canopy than in gaps. P
ropagules and new seedlings (i.e. the most recently established cohort) wer
e more abundant under the canopy than in gaps and on flat areas than on mou
nds.
4 A field experiment tested whether the observed spatial distribution of ne
w seedlings was due to differences in the secondary dispersal of propagules
or in the establishment or early survivorship of seedlings. Propagules wer
e supplied in equal densities on mounds and flats under the canopy and in g
aps and then monitored for 10 months.
5 Four times as many propagules were dispersed away from mounds as from fla
ts, and this difference was greater under the canopy than in gaps. In contr
ast, seedlings established more successfully on mounds, particularly in gap
s, suggesting that this is an important microhabitat for recruitment. Seedl
ing survival to 9 months was independent of habitat structure.
6 The influence of mounds on dispersal was, however, much greater than on e
stablishment, so that densities of new seedlings 9 months after establishme
nt were greater on flats. Patterns of seedling abundance in the mangrove fo
rest are thus primarily driven by the influence of habitat structure on the
supply of propagules.
7 In contrast, saplings were most abundant on mounds in gaps. Thus, there i
s a shift in the spatial distribution of the population from the early to t
he later stages of life history in relation to habitat structure. Long-term
survival of seedlings and their development into saplings may depend both
on light availability in gaps and on sediment disturbance resulting from th
e creation of mounds. The causative agent, crabs, may therefore be critical
for regeneration of mangrove forests.