A large body of work has demonstrated that riparian ecosystems play a criti
cal role regulating interactions between terrestrial and aquatic components
of temperate zone landscapes. However, there have been relatively few stud
ies of the role of riparian forests in tropical landscapes. There is a stro
ng need to evaluate the effects of conversion of these forests to pasture a
nd residential land uses on their ability to prevent the movement of nutrie
nts and other pollutants to coastal water bodies. The objectives of this st
udy were (1) to measure a suite of microbial properties that are indicators
of the ability of soils to attenuate pollutants and cycle nitrogen (N) in
tropical riparian forests under different land use management and (2) to ev
aluate the effects of conversion of two types of tropical forested wetlands
(cativo, Prioria copaifera and yolillo, Raphia taedigera) on this same sui
te of soil microbial variables. To accomplish objective #1, we measured the
se variables in surface soils in intact forest, disturbed forest, pasture a
nd residential land use sites in two locations in the Atlantic Zone of Cost
a Rica. For objective #2 we measured these variables in surface soils of th
ree replicate intact and converted cativo and yolillo forests.
Results suggest that land use conversion does not decrease levels of microb
ial biomass and activity in these soils. Intact forest sites did not have h
igher levels of microbial biomass and activity than 'disturbed sites.' In s
everal cases, levels of these variables were consistently lower in intact f
orest than disturbed sites. There were no statistically significant differe
nces between intact and converted cativo and yolillo sites. The relatively
low level of activity in the intact sites is likely a 'succession effect',
where the disturbed sites have younger, more actively growing vegetation th
at supplies more readily available organic matter to the soil microbial com
munity. High correlations between soil organic matter and microbial biomass
and activity strongly support this 'successional' explanation of our resul
ts. (C) 2001 Elsevier Science Ltd. All rights reserved.