Bibliographic citations
Zamora, A., (2016). Estudio metodológico para el monitoreo de alertas tempranas de deforestacion basado en focos de calor en la Amazonía peruana [Trabajo de suficiencia profesional, Universidad Nacional Agraria La Molina]. https://hdl.handle.net/20.500.12996/2601
Zamora, A., Estudio metodológico para el monitoreo de alertas tempranas de deforestacion basado en focos de calor en la Amazonía peruana [Trabajo de suficiencia profesional]. PE: Universidad Nacional Agraria La Molina; 2016. https://hdl.handle.net/20.500.12996/2601
@misc{renati/248087,
title = "Estudio metodológico para el monitoreo de alertas tempranas de deforestacion basado en focos de calor en la Amazonía peruana",
author = "Zamora Ríos, Alejandra",
publisher = "Universidad Nacional Agraria La Molina",
year = "2016"
}
This research was conducted throughout 2015 in the framework of the activities of the “Sala de Observación Peru” of the Amazon Cooperation Treaty Organization with the aim of establishing a methodology for reporting near-real-time deforestation alerts based on the active fire data. Due to this, it was planned to perform validation studies where the effectiveness of the active fire data were measured, contrasting fire scars left on the soil surface viewed from a satellite image. Previously, an historical analysis of active fire data in Peru (2000 - 2015) was performed to identify areas, months and sensors where hot spots for the validations would be taken. The first validation was performed with MODIS active fire data, contrasting with Rapid Eye satellite images from 2011-2012. The second validation was performed with VIIRS (NPP-750m and NPP-375m) and MODIS (Terra and Aqua) active fire data, contrasting with Landsat satellite scenes of 2015. The results showed a 93.2% of effectiveness for MODIS active fire data, 98.4% for NPP-750m and 98.8% for NPP-375m. 48% of frequent errors occurred at rivers’ beaches, 38% at forest and 14% at urban areas. Other observations indicated that areas of 0.5ha were detected by active fire data from both sensors and the probability of detection of events increases when the four satellites are used simultaneously. Finally, a protocol was established to consolidate the methodology, which includes a masking of the areas of frequent error and provides a methodology to reduce fire data to events, so as to avoid repetition of alerting.
This item is licensed under a Creative Commons License