i

Call for Papers for the semi-thematic N° 67: (Re)defining rural territories, between the global South and North: actors, processes, scales.

Full papers are invited to be submitted via the journal's official platform by 15 March 2024.

For more information, please check this link

Using object-oriented image analysis to map and monitor land cover change in the Region Costa Maya, México: 1993-2010

Authors

  • Morgan Simbangala Simbangala División de Ciencias e Ingeniería, Universidad de Quintana Roo, Boulevard Bahía esq. Ignacio Comonfort, Colonia del Bosque, C.P. 77019, Chetumal, Quintana Roo, México
  • Bonnie Lucía Campos Cámara División de Ciencias e Ingeniería, Universidad de Quintana Roo, Boulevard Bahía esq. Ignacio Comonfort, Colonia del Bosque, C.P. 77019, Chetumal, Quintana Roo, México
  • Lourdes Castillo Villanueva División de Ciencias e Ingeniería, Universidad de Quintana Roo, Boulevard Bahía esq. Ignacio Comonfort, Colonia del Bosque, C.P. 77019, Chetumal, Quintana Roo, México
  • Óscar Frausto Martínez División de Ciencias e Ingeniería, Universidad de Quintana Roo, Boulevard Bahía esq. Ignacio Comonfort, Colonia del Bosque, C.P. 77019, Chetumal, Quintana Roo, México
  • David Velázquez Torres División de Ciencias e Ingeniería, Universidad de Quintana Roo, Boulevard Bahía esq. Ignacio Comonfort, Colonia del Bosque, C.P. 77019, Chetumal, Quintana Roo, México
  • Rafael Romero Mayo División de Ciencias e Ingeniería, Universidad de Quintana Roo, Boulevard Bahía esq. Ignacio Comonfort, Colonia del Bosque, C.P. 77019, Chetumal, Quintana Roo, México
  • María Estela Orozco Hernández Facultad de Planeación Urbana y Regional, Universidad Autónoma del Estado de México, Toluca, Estado de México, México.

Abstract

Accurate, cost effective and timely multiple spatial-temporal information on the patterns of land cover change is crucial for environmental management and understanding. For this reason, segmentation and object-oriented classification was applied to Landsat TM/ ETM+ imagery to map and monitor land cover dynamics in the Región Costa Maya (RCM) in 1993, 2000 and 2010. Overall mapping accuracy for land-cover map in 2000 was 94.29% (ĸ= 0.9141). Post-classification approach, involving cross tabulation of three generated maps, was used to characterize spatial temporal rates and patterns of land cover change to infer major processes of changes over 17 years. Results revealed rapid urbanization, agricultural land abandonment (forest transition) and destruction of mangrove forests, mediated by socio-economic factors linked to tourism development as the leading drivers of land cover change, with grave implications on environmental sustainability in the Costa Maya area. The study has confirmed the value of segmentation and object-oriented classification for mapping and monitoring land cover change at regional scale.

Keywords:

change detection, object-oriented classification, image segmentation, remote sensing, Landsat images