@article{fdi:010088918,
  title = {{A}pplying a machine learning-based method for the prediction of suspended sediment concentration in the {R}ed river basin},
  author = {{N}guyen, {S}. {Q}. and {N}guyen, {L}. {C}. and {N}go-{D}uc, {T}. and {O}uillon, {S}ylvain},
  editor = {},
  language = {{ENG}},
  abstract = {{K}nowledge of sediment transport is important to understand the transportation and recycling of elements and matter in the {E}arth system. {U}sually, sediment transport in rivers is characterized by suspended sediment concentration ({SSC}) and river discharge ({Q}). {H}owever, {SSC} measurements are often inadequate in many river systems, such as the {R}ed {R}iver basin in {V}ietnam. {I}n this study, we performed a {T}ributary-based {D}ownstream gauge {E}stimation ({TDE}) machine learning ({ML}) approach to estimate {SSC} at {S}on {T}ay hydrological station based on {Q} and {SSC} monthly data from three upstream stations, one per tributary of the {R}ed {R}iver over a 14-years period (2000-2013). {A} comparative analysis of four {ML} algorithms, including {M}ultiple {L}inear {R}egression ({MLR}), {E}lastic {N}et ({EN}), {R}andom {F}orest ({RF}), and {S}upport {V}ector {M}achines ({SVM}) was conducted. {R}esults showed that when using both {Q} and {SSC} of the three upstream stations, the {SVM} algorithm with linear kernel exhibited the highest accuracy (r2 = 0.87 and {RMSE} = 64.7 g m-3). {T}he performance of the {TDE}-{ML} was seasonally dependent, with higher accuracy in the high-flow period. {T}his approach also revealed that {SSC} measured at {Y}en {B}ai station ({T}hao {R}iver) had the highest contribution to the prediction of {SSC} at {S}on {T}ay station meanwhile {V}u {Q}uang station ({L}o {R}iver) contributed the least to downstream {SSC}. {F}urthermore, new dams have been impounded during the 14-years period. {A}lthough the global performance of the {RF} method was slightly less than {SVM} with linear kernel, it was the only one able to fairly estimate {SSC} in the most recent 6-years period affected by new forcing.},
  keywords = {{M}achine learning ; {R}ed {R}iver ; {T}ributary-based approach ; {S}ediment transport ; {M}odelling ; {VIET} {NAM}},
  booktitle = {},
  journal = {{M}odeling {E}arth {S}ystems and {E}nvironment},
  volume = {[{E}arly access]},
  numero = {},
  pages = {[18 p.]},
  ISSN = {2363-6203},
  year = {2024},
  DOI = {10.1007/s40808-023-01915-y},
  URL = {https://www.documentation.ird.fr/hor/fdi:010088918},
}