Identifying sediment critical source areas in subtropical Brazil using SWAT model
Visualizar/ Abrir
Data
2020-10-29Primeiro coorientador
Kalin, Latif
Primeiro membro da banca
Tiecher, Tales
Segundo membro da banca
Rodrigues, Miriam Fernanda
Metadata
Mostrar registro completoResumo
Agriculture without the use of conservation practices has caused intense soil and water
degradation. As a result, erosion and sediment production increase, which, in turn,
increases the load of agrochemicals and fertilizers that contribute to river
eutrophication and siltation of canals. In addition to the impact on water resources,
erosion decreases the productive capacity of the soil, impacting agriculture itself in a
vicious cycle. Identifying areas where diffuse pollution originates is complex, costly and
time consuming, requiring specific screening techniques, especially when applied to
large watersheds. In this sense, the objective of this study is to evaluate the ability of
the Soil and Water Assessment Tool (SWAT) model in identifying the critical areas of
sediment production, providing information to mitigate the problem. The study area
was the Guaporé river basin (2,031 km²), in Rio Grande do Sul, the southernmost
Brazilian State. The Guaporé River basin is characterized by elevations ranging from
136 to 847 m, with subtropical climate and annual average rainfall of 1,800 mm, evenly
distributed throughout the year. Land use consists of areas with native vegetation of
mixed forest and grasslands and agricultural areas cultivated in the gentle hillslopes
with soybean (Glycine max) and maize (Zea mays) in the summer and wheat (Triticum
aestivum) in winter under the no-tillage system; while steep areas are cultivated with
tobacco (Nicotiana tabacum). The parameters that characterize the vegetation (forests
and agricultural crops) in SWAT, as well as the physical properties of the basin soil,
were modified to correctly represent the conditions of the Brazilian subtropical basins.
Flow was calibrated in the period 2009 - 2011 and validated in 2012 - 2014, using a 5-
year warm-up period. For sediment, the calibration period was from 2011 to 2013,
using 2014 data for validation. The Nash-Sutcliffe (NS) coefficient and the Percent Bias
(PBIAS) were used as objective function using the SUFI-2 algorithm. Calibration and
validation were performed at daily time scale and resulted in a coefficient of NS 0.71
and 0.73 for flow and 0.54 and 0.50 for sediment yield, respectively. The total sediment
yield estimated at the basin outlet was approximately 2.4 ton ha-1 y-1, being the tobacco
cultivation responsible for the largest contribution to sediment yield (~ 77.2% avg.),
followed by soybean and bare soil (~ 12.2%) and corn and soybean (~ 6.2%), in
agreement with results by conventional sediment tracing techniques previously applied
in the basin. Results demonstrates the efficiency of SWAT modeling in the proper
identification of critical areas of sediment yield.
Coleções
Os arquivos de licença a seguir estão associados a este item: