COS 28-5
Structure and composition of intact and fragmented riparian forests in an agricultural Amazonian landscape

Tuesday, August 12, 2014: 9:20 AM
Regency Blrm E, Hyatt Regency Hotel
R. Chelsea Nagy, Earth Lab, University of Colorado Boulder, Boulder, CO
Stephen Porder, Institute at Brown for Environment & Society, Brown University, Providence, RI
Christopher Neill, Ecosystems Center, Marine Biological Laboratory, Woods Hole, MA
Paulo M. Brando, Instituto de Pesquisa Ambiental da Amazônia (IPAM), Brasília, Brazil
Raimundo Mota Quintino, Instituto de Pesquisa Ambiental da Amazônia (IPAM), Canarana, Brazil
Sebastiâo Aviz do Nascimento, Instituto de Pesquisa Ambiental da Amazônia (IPAM), Canarana, Brazil
Background/Question/Methods

Conversion of forests and savannas to soybean agriculture and pasture has created a fragmented landscape in the southern Brazilian Amazon.  By law, cleared lands must contain a forested riparian buffer strip, but the fate of these elongate forest fragments in a matrix of pasture and soybeans has not been documented.   We studied riparian fragments surrounded by soybean fields in the southern Amazonian state of Mato Grosso, and compared their structure, biomass, and species composition to nearby riparian forests in intact forested areas.  We predicted that riparian fragments would have higher tree mortality, lower biomass, reduced seedling and sapling regeneration, and more lianas compared with their intact counterparts and that these effects would be most pronounced in the narrowest riparian fragments.  To test these hypotheses, we selected four intact riparian forests (surrounded by forest) and four riparian fragments (surrounded by soy) including three different widths in each riparian fragment (mean width= 89, 201, 264 m).  Along transects running perpendicular to the stream, we compared physical (temperature, relative humidity, light, depth to water table) and biological (number of live and dead trees, seedlings, saplings, lianas) characteristics among these sites. 

Results/Conclusions

The edges of soy riparian forests are hotter and drier compared to similar locations in intact riparian forests during the dry season, but not during the wet season.  Additionally, the water table at any given distance from the stream was higher in riparian forests surrounded by soy than in those surrounded by intact forest.  Nevertheless, tree size distributions, aboveground carbon storage, and the number of live and dead trees did not differ between riparian forest type (intact vs. soy) or width.  However, riparian forests surrounded by soy had fewer seedlings and saplings (13-16 vs. 21 seedlings m-2 and 2-5 vs. 6 saplings m-2 in soy and intact, respectively).  Several diversity metrics indicated that species diversity of soy riparian forests was reduced compared to intact riparian forests.  Furthermore, the composition of soy riparian forests is dissimilar from intact riparian forests, with more lianas and pioneer species found in riparian forests surrounded by soy.  Thus while these riparian forest fragments are not dying off, their regeneration potential and species diversity and composition have been altered by their isolation.  The consequences of these alterations are not yet understood.