Abstract
Abstract. Deforestation can be defined as the conversion of forest land cover to another type. It is a process that has massively accelerated its rate and extent in the last several decades. Mainly due to human activities related to socio-economic processes as population growth, expansion of agricultural land, wood extraction, etc. In the meantime, there are great efforts by governments and agencies to reduce these deforestation processes by implementing regulations, which cannot always be properly monitored whether are followed or not. In this work is proposed an approach that can provide forest loss estimations for a short period of time, by using Synthetic Aperture Radar imagery for an area in the Brazilian Amazon. SAR are providing data with almost no alteration due to weather conditions, however they may present other limitations. To mitigate the speckle effect, here was applied the dry coefficient, which is the mean of image values under the first quartile while preserving the spatial resolution. While for obtaining land cover maps containing only forest and non-forest areas an object-based machine learning classification on the Google Earth Engine platform was applied. The preliminary tests were carried out in a bitemporal manner between 2015 and 2019, followed by applying the approach monthly for the year of 2020. The outputs yielded very satisfactory and accurate results, allowing to estimate the forest dynamics for the area under consideration for each month.
Highlights
Due to various processes natural and anthropogenic, deforestation is increasing, and it represents a global issue leading to climate change, loss of biodiversity and larger probability of hazards
The related object-based classifications are depicting extremely low level of noised misclassification pixels that is usually very evident in the pixelwise approaches. It comes with its limitation, it can be seen in a misclassification of areas that are relatively small and inside a bigger region from the opposite class. This issue is related to the parameter definition in the segmentation step which for the current case study and its scale was found to be a challenging task to find the balance between high precision and acceptable computational demand
In regard to the fusion of Synthetic Aperture Radar (SAR) and Multispectral Instrument (MSI) datasets it was noted an overall improvement in the final results; on one hand the validation metrics were increased and, on the other, the final map was more consistent, and it was noted that the misclassifications were usually in areas hard to be interpreted even upon a visual inspection
Summary
Due to various processes natural and anthropogenic, deforestation is increasing, and it represents a global issue leading to climate change, loss of biodiversity and larger probability of hazards. Spaceborne Synthetic Aperture Radar (SAR) has on the opposite the advantage that sensors provide data with almost no restrictions from weather conditions. The limitation in this case is that, due to backscatter speckle, SAR imagery can introduce additional difficulty to apply, otherwise, straightforward processing, such as change detection with single thresholding. Scholars applying various approaches to deal with the SARrelated limitation, such as, statistical thresholding of time-series (Canty et al, 2020); adoption of the shadow effect created from the SAR imaging on the border between forested and nonforested areas (Bouvet et al, 2018); usage of different SAR frequency ranges (Rahman and Sumantyo, 2010) or directly combining radar data with optical (Hirschmugl et al, 2020)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
