@article{fdi:010087996,
  title = {{R}e{C}u{S}um : a polyvalent method to monitor tropical forest disturbances [+ {C}orrigendum, 1 p.]},
  author = {{Y}gorra, {B}. and {F}rappart, {F}. and {W}igneron, {J}.{P}. and {M}oisy, {C}. and {C}atry, {T}hibault and {P}illot, {B}enjamin and {C}ourtalon, {J}. and {K}harlanova, {A}. and {R}iazanoff, {S}.},
  editor = {},
  language = {{ENG}},
  abstract = {{C}hange detection methods based on {E}arth {O}bservations are increasingly used to monitor rainforest in the intertropical band. {U}ntil recently, deforestation monitoring was mainly based on remotely sensed optical images which often face limitations in humid tropical areas due to frequent cloud coverage. {T}his leads to late detections of disturbance events. {S}ince the launch of {S}entinel-1 acquiring images with a revisit time of 12 days and a spatial resolution of 5 × 20 m in {B}razil, {S}ynthetic {A}perture {R}adar ({SAR}) images have been increasingly used to monitor deforestation. {I}n this study, we propose a multitemporal version of the change detection method we previously applied to timeseries of {S}entinel-1 {SAR} images, to monitor deforestation/degradation in the {C}ongo rainforest. {O}ur approach is based on a {C}umulative {S}um ({C}u{S}um) method combined with a spatial recombination of {C}ritical {T}hresholds ({C}u{S}um cross-{T}c). {T}he newly developed multitemporal {C}u{S}um method ({R}e{C}u{S}um) was applied to a time-series of 82 dual polarization ({VH}, {VV}) {G}round {R}ange {D}etected ({GRD}) {S}entinel-1 images acquired in the {P}ara {S}tate, in the {B}razilian {A}mazonia, from 29/09/2016 to 01/07/2019. {T}he {R}e{C}u{S}um method consists of iteratively applying the {C}u{S}um cross-{T}c to monitor multiple changes in a time-series by splitting the time-series at each date of detected change and by independently iterating over the time periods resulting from the splits. {T}he number of changes in the time-series was then analysed according to the vegetation type ({F}orest, non-forest vegetation) determined by visual inspection of optical {S}entinel-2 image and {P}lanet{S}cope monthly mosaic. {T}his showed a difference between non-forest vegetation and forested areas. {A} threshold based on the number of changes ({T}nbc) was then developed to differentiate forest from non-forest disturbances. {T}he ability to monitor non-forest vegetation was analysed: the {C}u{S}um cross-{T}c detected up to 90% of the total non-forest vegetation area over the study region in the past period. {A}fter removing past disturbances and past non-forest vegetation, then removing the pixels covered with non-forest vegetations based on {T}nbc, the application of the {R}e{C}u{S}um led to a precision of 81%, a recall of 68%, a kappa coefficient of 0.72 and a {F}1- score of 0.74 in forest disturbance monitoring. {A}ccording to these results, {R}e{C}u{S}um applied to {S}entinel-1 time-series of images can be used for efficient forest disturbance monitoring and for generating a forest / non-forest map after the application of newly developed post-processing steps. {S}entinel-1 imagery can be used for both {F}orest / {N}on-forest mapping and for forest disturbance detection. {R}e{C}u{S}um was released as an open-source {GIT} project available at: https://forgemia.inra.fr/bertrand.ygorra/cusum-deforestation_monitoring},
  keywords = {{BRESIL} ; {AMAZONIE} ; {PARA} {ETAT}},
  booktitle = {},
  journal = {{ISPRS} {J}ournal of {P}hotogrammetry and {R}emote {S}ensing},
  volume = {203},
  numero = {},
  pages = {358--372 [+ {C}orrigendum, vol. 211,  298, 2024]},
  ISSN = {0924-2716},
  year = {2023},
  DOI = {10.1016/j.isprsjprs.2023.08.006},
  URL = {https://www.documentation.ird.fr/hor/fdi:010087996},
}