@article{fdi:010076124,
  title = {{A} classification approach to reconstruct local daily drying dynamics at headwater streams},
  author = {{B}eaufort, {A}. and {C}arreau, {J}ulie and {S}auquet, {E}.},
  editor = {},
  language = {{ENG}},
  abstract = {{H}eadwater streams ({HS}s) are generally naturally prone to flow intermittence. {T}hese intermittent rivers and ephemeral streams have recently seen a marked increase in interest, especially to assess the impact of drying on aquatic ecosystems. {T}he two objectives of this work are (a) to identify the main drivers of flow intermittence dynamics in {HS} and (b) to reconstruct local daily drying dynamics. {D}iscrete flow states-"flowing" versus "drying"-are modelled as functions of covariates that include information on climate, hydrology, groundwater, and basin descriptors. {T}hree classifiers to estimate flow states using covariates are tested on four contrasted regions in {F}rance: (a) a linear classifier with regularization ({LASSO} for least absolute shrinkage and selection operator) and two non-linear non-parametric classifiers, (b) a one-hidden-layer feedforward artificial neural network ({ANN}) classifier, and (c) a random forest ({RF}) classifier. {T}he three classifiers are compared with a benchmark classifier ({BC}) that simply estimates dominant flow state for each month based on observations (without using covariates). {T}he performance assessment over the period 2012-2016 carried out by cross-validation shows that the three classifiers for flow state based on covariates outperformed the {BC}. {T}his demonstrates the predictive power of the covariates. {ANN} is the classifier that globally achieves the best performance to predict the daily drying dynamics whereas both {RF} and {LASSO} tend to underestimate the proportion of drying states. {T}he covariates are ranked in terms of relevance for each classifier. {T}he monthly proportion of drying states provided by the discrete observation network has a major importance for the three classifiers {ANN}, {LASSO}, and {RF}. {T}his may reflect the proclivity of a site to flow intermittence. {ANN} gives higher importance to climatic and hydrological covariates and its non-linearity allows a greater degree of freedom.},
  keywords = {artificial neural network ; drying prediction ; flow state ; intermittent rivers ; least absolute shrinkage and selection operator ; random forest ; {FRANCE}},
  booktitle = {},
  journal = {{H}ydrological {P}rocesses},
  volume = {33},
  numero = {13},
  pages = {1896--1912},
  ISSN = {0885-6087},
  year = {2019},
  DOI = {10.1002/hyp.13445},
  URL = {https://www.documentation.ird.fr/hor/fdi:010076124},
}