Jm. Dupuy et Rl. Chazdon, LONG-TERM EFFECTS OF FOREST REGROWTH AND SELECTIVE LOGGING ON THE SEED BANK OF TROPICAL FORESTS IN NE COSTA-RICA, Biotropica, 30(2), 1998, pp. 223-237
To investigate long-term effects of land use on the soil seed bank, we
compared the abundance/density, species richness, life form distribut
ion, and species composition of seeds stored in the soil of four 15-20
yr-old second-growth stands, two old-growth stands, and two previousl
y selectively-logged stands in the Caribbean lowlands of Costa Rice. S
urface soil (10 cm deep, 4.7 cm diameter) was collected at 10 m interv
als along three 120-160 m long transects in each stand (44-48 soil cor
es, 22-24 combined seed bank samples per site). Seed density was highe
st but variable in second-growth stands (8331-14535 seeds/m(2)), low a
nd homogeneous in old-growth stands (2258-2659 seeds/m(2)), and interm
ediate and highly variable in selectively-logged stands (1165-6854 see
ds/m(2)), which also had contrasting logging intensities. Species rich
ness was strongly dependent on seed density, but showed less variation
. Life form distribution did not differ statistically among or within
land-use categories. In each stand, herbs-forbs, shrubs, and vines dom
inated the seed bank (> 75% of the species richness and abundance), wh
ereas trees were a minor component (< 20% of the species richness and
< 5% of the abundance) and were predominantly early successional. Shru
bs and vines were most abundant in second-growth stands where regrowth
vegetation was repeatedly cut before abandonment, whereas grasses and
sedges were most abundant in the only forest stand that was completel
y surrounded by pastures. In terms of species composition, old-growth
stands were more similar to selectively-logged stands than to second-g
rowth stands, bur across stands, selectively-logged forests were most
distinct from the other two forest types. An inventory of the standing
woody vegetation in each site showed little representation of the woo
dy taxa found in the seed bank. We discuss these results in the contex
t of the main factors that have been postulated to influence the abund
ance, life form, and species composition of tropical forest seed banks
, and explore the role of the latter during intermediate phases of tro
pical forest succession and regeneration.