@article{fdi:010074952,
  title = {{N}ear real-time agriculture monitoring at national scale at parcel resolution : performance assessment of the {S}en2-{A}gri automated system in various cropping systems around the world},
  author = {{D}efourny, {P}. and {B}ontemps, {S}. and {B}ellemans, {N}. and {C}ara, {C}. and {D}edieu, {G}. and {G}uzzonato, {E}. and {H}agolle, {O}. and {I}nglada, {J}. and {N}icola, {L}. and {R}abaute, {T}. and {S}avinaud, {M}. and {U}droiu, {C}. and {V}alero, {S}. and {B}egue, {A}. and {D}ejoux, {J}. {F}. and {E}l {H}arti, {A}. and {E}zzahar, {J}. and {K}ussul, {N}. and {L}abbassi, {K}. and {L}ebourgeois, {V}. and {M}iao, {Z}. and {N}ewby, {T}. and {N}yamugama, {A}. and {S}alh, {N}. and {S}helestov, {A}. and {S}imonneaux, {V}incent and {T}raore, {P}. {S}. and {T}raore, {S}. {S}. and {K}oetz, {B}.},
  editor = {},
  language = {{ENG}},
  abstract = {{T}he convergence of new {EO} data flows, new methodological developments and cloud computing infrastructure calls for a paradigm shift in operational agriculture monitoring. {T}he {C}opernicus {S}entinel-2 mission providing a systematic 5-day revisit cycle and free data access opens a completely new avenue for near real-time crop specific monitoring at parcel level over large countries. {T}his research investigated the feasibility to propose methods and to develop an open source system able to generate, at national scale, cloud-free composites, dynamic cropland masks, crop type maps and vegetation status indicators suitable for most cropping systems. {T}he so-called {S}en2-{A}gri system automatically ingests and processes {S}entinel-2 and {L}andsat 8 time series in a seamless way to derive these four products, thanks to streamlined processes based on machine learning algorithms and quality controlled in situ data. {I}t embeds a set of key principles proposed to address the new challenges arising from countrywide 10 m resolution agriculture monitoring. {T}he full-scale demonstration of this system for three entire countries ({U}kraine, {M}ali, {S}outh {A}frica) and five local sites distributed across the world was a major challenge met successfully despite the availability of only one {S}entinel-2 satellite in orbit. {I}n situ data were collected for calibration and validation in a timely manner allowing the production of the four {S}en2-{A}gri products over all the demonstration sites. {T}he independent validation of the monthly cropland masks provided for most sites overall accuracy values higher than 90%, and already higher than 80% as early as the mid-season. {T}he crop type maps depicting the 5 main crops for the considered study sites were also successfully validated: overall accuracy values higher than 80% and {F}1 {S}cores of the different crop type classes were most often higher than 0.65. {T}hese respective results pave the way for countrywide crop specific monitoring system at parcel level bridging the gap between parcel visits and national scale assessment. {T}hese full-scale demonstration results clearly highlight the operational agriculture monitoring capacity of the {S}en2-{A}gri system to exploit in near real-time the observation acquired by the {S}entinel-2 mission over very large areas. {S}caling this open source system on cloud computing infrastructure becomes instrumental to support market transparency while building national monitoring capacity as requested by the {AMIS} and {GEOGLAM} {G}-20 initiatives.},
  keywords = {{A}griculture monitoring ; {C}loud computing ; {M}achine learning ; {S}entinel-2 ; {C}rop type mapping ; {C}ropland ; {UKRAINE} ; {MALI} ; {AFRIQUE} {DU} {SUD}},
  booktitle = {},
  journal = {{R}emote {S}ensing of {E}nvironment},
  volume = {221},
  numero = {},
  pages = {551--568},
  ISSN = {0034-4257},
  year = {2019},
  DOI = {10.1016/j.rse.2018.11.007},
  URL = {https://www.documentation.ird.fr/hor/fdi:010074952},
}