@article{fdi:010096452,
  title = {{A}ssessing urban flood susceptibility using random forest machine learning and geospatial technologies : application to the {B}onoumin-{P}almeraie watershed, {A}bidjan ({C}{\^o}te d'{I}voire)},
  author = {{D}anumah, {J}. {H}. and {A}taba, {W}. {A}. and {S}okeng, {V}. {C}. {J}. and {A}kpa, {Y}. {L}. and {S}aley, {M}. {B}. and {O}gilvie, {A}ndrew},
  editor = {},
  language = {{ENG}},
  abstract = {{R}ecurrent flooding poses a persistent and growing threat to {W}est {A}frican watersheds facing rapid urbanization and climate change. {D}espite advances in machine learning and geospatial datasets, urban planning and flood prevention often rely on limited datasets and traditional analysis. {T}his study addresses this research gap in the {B}onoumin-{P}almeraie watershed ({A}bidjan, {C}{\^o}te d'{I}voire) by developing an integrated approach leveraging remote sensing, {G}eographic {I}nformation {S}ystems ({GIS}), and the {R}andom {F}orest algorithm to assess and map flood susceptibility. {T}welve conditioning factors related to topography, hydrology, land use, and climate were derived from {S}entinel-1, {ALOS} {PALSAR}, and multi-source earth observation datasets. {H}istorical flood extents were mapped in {G}oogle {E}arth {E}ngine to train the {R}andom {F}orest model in a {G}oogle {C}olab environment. {T}he model demonstrated high discriminatory power, yielding an {A}rea {U}nder the {C}urve of 0.94 and {O}verall {A}ccuracy of 0.83. {D}rainage density, rainfall, and altitude were identified as the primary explanatory drivers. {T}he resulting flood susceptibility map indicates that 39% of the watershed exhibits medium to very high susceptibility, with critical hotspots in the neighborhoods of {P}almeraie, {A}ttoban, {A}kouedo, {D}jorogobit{\'e}, and {R}iviera-{S}ogefiha. {W}hile limited by the exclusion of certain anthropogenic variables and ground truth constraints, the study provides a reproducible, data-driven framework for flood risk assessment in tropical urban environments. {T}hese findings offer essential scientific support for urban planners and decision-makers to enhance territorial planning and sustainable flood management in {A}bidjan.},
  keywords = {random forest ; artificial intelligence ; flood susceptibility ; urban ; hydrology ; {W}est {A}frica ; {COTE} {D}'{IVOIRE} ; {ABIDJAN}},
  booktitle = {},
  journal = {{W}ater},
  volume = {18},
  numero = {3},
  pages = {402 [19 p.]},
  ISSN = {2073-4441},
  year = {2026},
  DOI = {10.3390/w18030402},
  URL = {https://www.documentation.ird.fr/hor/fdi:010096452},
}