To further complicate the matter, the detected disturbance in the forest should ideally be evaluated against a desired
baseline (corresponding to ‘intact’ forest) in order to determine the level of degradation, taking into account the duration
of the detected effects and the natural variation of the forest (Thompson et al., 2013). Available remote sensing data often
do not cover long enough historical time-frame to properly monitor the full duration of the effects. Therefore, although the
aspect of duration is discussed at some points in the following sections, it is important to emphasize that we primarily look
at indicators of forest degradation which can be derived from remote sensing data. In other words, we look at methods
which can identify areas of ‘‘potential recent forest degradation’’, using a single date or time series of satellite datasets. For
practical applications, these potential signs of degradation derived from high to medium resolution satellite data need to be
verified and potentially calibrated with auxiliary information (e.g. field data or very high resolution satellite sampling).