Land use dynamics in the semi-arid region of Brazil (Quixaba, PE): Characterization by principal component analysis (PCA)
2002; Taylor & Francis; Volume: 23; Issue: 23 Linguagem: Inglês
10.1080/0143116021000013313
ISSN1366-5901
AutoresFrancisco Darío Maldonado, João Roberto dos Santos, Vitor Celso Carvalho,
Tópico(s)Remote Sensing and Land Use
ResumoThe changes caused by the pressure of human occupation can be observed by non-sustainable land use forms, in the Brazilian semi-arid region. The multitemporal analysis of changes in the caatinga land cover provides sufficient information about the dynamics of this typical land use. Within this frame, principal component analysis (PCA) was applied in combination with field survey data, which permitted estimation of point samples of new recovery/degradation. The area selected for this study is located in the central-southern semi-arid section of Pernambuco State, north-east Brazil. It covers an area of about 190 km 2 and is located between geographical coordinates 8° 00' to 8° 07' S and 39° 45' to 39° 53' W. Five classes of changed and unchanged multitemporal effects were discriminated. The results of this study were evaluated using combined Geographical Information System (GIS) and field survey techniques. This work was less time-consuming and economically significant for the time period 1983-96. An alternative PCA method was proposed based on the analysis of the results and errors in a confusion matrix. This alternative method is based on the controlled rotative analogy of the PCA transformation, where a smaller angle of rotation was applied.
Referência(s)