Skip to main content
ARS Home » Midwest Area » West Lafayette, Indiana » National Soil Erosion Research Laboratory » Research » Publications at this Location » Publication #300828
Title: Phytogeographic retrospective in ecotonal areas guided by soil attributes

Author
item MONTEIRO, TALITA - Sao Paulo State University (UNESP)
item ALCARDE ALVARES, CLAYTON - Forestry Science And Research Institute (IPEF) And Forest Productivity Coopeartive (fpc)
item Stott, Diane
item MARCO DA SILVA, ALEXANDRE - Sao Paulo State University (UNESP)

Submitted to: International Journal of Environmental Science and Technology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/7/2014
Publication Date: 10/21/2014
Citation: Monteiro, T., Alcarde Alvares, C., Stott, D.E., Marco Da Silva, A. 2014. Phytogeographic retrospective in ecotonal areas guided by soil attributes. International Journal of Environmental Science and Technology. 12(9):2829-2840. doi:10.1007/s13762-014-0684-y.

Interpretive Summary: Ecotones are transitional areas that lie between two well-defined ecological biomes. It is where the two biomes, or ecological communities meet and integrate. While ecotones can be sharply defined with a distinct line between the two communities, with plant species of each biome migrating into the ecotonal areas. Often these areas have been converted to agriculturally managed lands and have been subjected to soil degradation. There is increasing interest in restoring some of these areas to the native vegetation, however it is difficult to determine which areas within an ecotonal ecosystem was originally populated by plant species of either of the bordering biomes. In the study area located in San Paolo State in Brazil, the two biomes of interest are the Atlantic forest and Savannah. The study area is near the city of Iperó in San Paolo State, Brazil, which primarily consists of three land-cover ecosystems: Atlantic Forest (33.3%), savannah (27.8%) and pasture (29.6%). There is interest in restoring some of the pasture areas to the original ecotonal plantings, however it is visually impossible to determine which areas were originally forest vegetation and which were savannah. Soils were sampled from pristine forest and savannah areas, as well as the pasture areas of interest for restoration, and 14 soil indicators were measured, with 4 being distinctly different between the two pristine biomes. Statistical procedures were used to identify which of the pasture sampling spots were originally forest and which were originally under savannah grassland. The statistical methodologies identified should be useful in restoration projects in other ecotonal regions around the world. The soil indicators that distinguished between the two biomes in this case are likely to be different in other ecoregions due to the presence of different soil orders and vegetation. This research impacts land managers and restoration ecologists interested in restoring degraded ecotonal lands to its natural vegetation.

Technical Abstract: Re-establishing the deforested ecosystems to pre-settlement vegetation is difficult, especially in ecotonal areas, due to the lack of knowledge about the original physiognomy. Our objective was to use a soils database that included chemical and physical parameters to distinguish soil samples from forest sites and savannah sites in a municipality located in the Brazilian southeastern region. Discriminant Analysis (DA) was used to determine the original biome vegetation (forest or savannah) in ecotone regions that have been converted to pasture and are generally degraded. First soils from pristine forest and savannah sites were tested, resulting in a reference database to compare to the degraded soils. Although the data presented in general high level of similarity among the two biomes, some differences occurred and it was sufficient for DA distinguish the sites and classify the soil samples took from grassy areas into forest or savannah. The soils from pastured areas presented quality worse than the soils of the pristine areas. Through DA analysis we observed that, from seven soil samples collected from grassy areas, five were most likely originally forest biome and two were savannah, ratified by a complementary cluster analysis carried out with the database of these samples. The model here proposed is pioneer. However, the users should keep in mind that establish a regional-level database of soil features using soil samples collected both from pristine and degraded areas is critical for success of the project, especially because the ecological and regional particularities of each biome. This research impacts land managers and restoration ecologists interested in restoring degraded lands to its natural vegetation.