Document details

Avaliação da composição e estrutura ripária Mediterrânica baseada em SIG e detecção remota

Author(s): Fernandes, Maria do Rosário Pereira

Date: 2013

Persistent ID: http://hdl.handle.net/10400.5/6450

Origin: Repositório da UTL

Subject(s): riparian vegetation; remote detection; ecological condition; optical traits; spectral and geometric attributes


Description

Riparian forests are responsible for many functions considered essential to the preservation of the ecological condition of fluvial corridors. The aim of this thesis is to characterize the structural and compositional patterns of the riparian vegetation in relation to its ecological quality using remote detection and geographic information systems. Separability analyses allowed to characterize and distinguish the spectral patterns and divergent optical behavior between the main riparian forests of Portugal. Spectroradiometry analyses enable the identification of the optimal bands for the remote detection of the alien invasive species Arundo donax, giant reed, from the surrounding vegetation, taking into account its seasonal spectral variability. The Geostatistical techniques combined with the application of landscape metrics, in high spatial resolution images, allowed the remote identification of the structural patterns for riparian forests and for the riparian areas invaded by the giant reed. It was obtained a relation between the observed degradation patterns and a gradient of human disturbance in the surrounding areas of fluvial corridors. The combination of the spectral and geometric attributes allowed to increasing giant reed mapping accuracy in riparian habitats, using a semi-automatic technique.

Doutoramento em Engenharia Florestal e dos Recursos Naturais - Instituto Superior de Agronomia

Document Type Doctoral thesis
Language Portuguese
Advisor(s) Cardoso, Maria Teresa Ferreira da Cunha; Pereira, José Miguel Cardoso; Aguiar, Francisca Constança Frutuoso
Contributor(s) Fernandes, Maria do Rosário Pereira
facebook logo  linkedin logo  twitter logo 
mendeley logo