Já é Inscrito? Acesse sua área restrita: login senha


English
Home
Comissões
Palestrantes
Inscrições
Exposição e Apoio
Local do Evento
Conheça Florianópolis
Divulgue o Evento
Associe-se à SBCS
Perguntas Frequentes
Fale Conosco
Indique o Site
Coloque aqui seu e-mail
Aqui o e-mail de seu amigo


Anais do XXXIV Congresso Brasileiro de Ciência do Solo



CHARACTERIZATION OF POTENTIAL AMAZONIAN DARK EARTH AREAS WITH PROXIMAL SENSING THECNOLOGIES


SUZANA ROMEIRO ARAÚJO(1); JOSÉ ALEXANDRE MELO DEMATTE(2); SIMONE VICENTE(2); MARSTON HÉRCULES FRANCESCHINI(2); MATS SORDESTROM(3); JAN ERICKSSON(3);
1 - UNIVERSIDADE DE SAO PAULO; 2 - USP; 3 - SLU;

Understand the Amazonian Dark Earths (ADE) areas is important for choosing appropriate management strategies to maintain their fertility under current agricultural regimes. We investigated if spectroscopy data collection could be used to assess the spatial distribution of soil properties based on their spectral characteristics using Partial Least Square (PLS) regressions. The study was carried out in the Belterra, PA, Brazil. In total more than 300 soil samples were collected. As a reference, 225 of the samples were analyzed chemically (pH, P, K, Ca, CEC, and soil organic carbon (SOC)). Models of soil properties using the mid-IR (4000-400 cm-1) sensor data outperformed the vis-NIR (400-2500 nm). The predictions of the PLSR of CEC and SOC were able to reproduce the spatial pattern of the properties evaluated. The mid-IR method worked well. Proximal soil sensing can be useful for assessing the spatial distribution of ADE soils and quantifying some of the most important soil properties.

[baixar arquivo pdf]

Promoção: Realização:
Apoio:
Patrocínio Bronze:

Apoio Institucional:

Agência Oficial de Turismo: Organização:
Desenvolvido por Zanda Multimeios da Informação