+Starting from v1.7, we added a `greendown_aware` parameter to account for the summer EVI2 greendown phenomenon when using the threshold-based phenometrics. Specifically, when the greendown phenomenon is substantial (e.g., in PhenoCam data), the `Senescence` metric, which is defined as 90% amplitude in autumn, can be biased early. However, this EVI2 decrease is not necessarily `Senescence` (the exact mechanisms that induced this summer greendown are still unclear. Some previous studies have attributed it to shadow and/or leaf angle). So, by using `FitBLSP(.., opt = list(method = "threshold", greendown_aware = TRUE))`, the `Senescence` metric will be retrieved as the end date of summer greendown (or, the edge of the curve), and the `MidGreendown` will be the date with the mininum first derivative of the autumn EVI2 curve after `Senescence`.
0 commit comments