Abstract

MAS, Jean-François et al. Assessment of deforestation rates in Michoacan at detailed scale through a hybrid classification method of SPOT images. Madera bosques [online]. 2017, vol.23, n.2, pp.119-131. ISSN 2448-7597.  https://doi.org/10.21829/myb.2017.2321472.

In the last decades, the state of Michoacán has suffered high deforestation rates. In this work, we implemented a hybrid classification method that combines automated processing, spatial analysis and visual interpretation, which enabled us to produce land use/cover maps for Michoacán, scale 1:50 000 with a minimum mapping unit of one hectare for 2004, 2007 and 2014. Additionally, we assessed the accuracy of the elaborated cartography. Accuracy assessment of the 2007 map shows an overall accuracy of 83.3%, with a confidence interval of 3.1%. Deforestation rates have decreased in Michoacán (i.e., 2004-2007; 0.17%, 2007-2014; 0.07%). However, there are two main deforestation hotspots in which pine and pine-oak forests have been replaced by avocado orchards (center part of the State) and deciduous forests have been transformed into grasslands (sierra-costa region).

Keywords : accuracy assessment; land use/cover maps; cartographic updating; forest inventories; image segmentation.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )