@article{fdi:010068846,
  title = {{L}inear and {N}on-{L}inear approaches for statistical seasonal rainfall forecast in the {S}irba watershed region ({SAHEL})},
  author = {{D}jibo, {A}. {G}. and {K}arambiri, {H}. and {S}eidou, {O}. and {S}ittichok, {K}. and {P}hilippon, {N}. and {P}aturel, {J}ean-{E}mmanuel and {S}aley, {H}. {M}.},
  editor = {},
  language = {{ENG}},
  abstract = {{S}ince the 90s, several studies were conducted to evaluate the predictability of the {S}ahelian rainy season and propose seasonal rainfall forecasts to help stakeholders to take the adequate decisions to adapt with the predicted situation. {U}nfortunately, two decades later, the forecasting skills remains low and forecasts have a limited value for decision making while the population is still suffering from rainfall interannual variability: this shows the limit of commonly used predictors and forecast approaches for this region. {T}hus, this paper developed and tested new predictors and new approaches to predict the upcoming seasonal rainfall amount over the {S}irba watershed. {P}redictors selected through a linear correlation analysis were further processed using combined linear methods to identify those having high predictive power. {S}easonal rainfall was forecasted using a set of linear and non-linear models. {A}n average lag time up to eight months was obtained for all models. {I}t is found that the combined linear methods performed better than non-linear, possibly because non-linear models require larger and better datasets for calibration. {T}he {R}-2, {N}ash and {H}it rate score are respectively 0.53, 0.52, and 68% for the combined linear approach; and 0.46, 0.45, 61% for non-linear principal component analysis.},
  keywords = {rainfall forecasting ; neural network ; non-linear principal component ; analysis ; {S}irba basin ; {W}est {A}frican monsoon ; air temperature ; {NIGER} ; {SAHEL}},
  booktitle = {},
  journal = {{C}limate},
  volume = {3},
  numero = {3},
  pages = {727--752},
  ISSN = {2225-1154},
  year = {2015},
  DOI = {10.3390/cli3030727},
  URL = {https://www.documentation.ird.fr/hor/fdi:010068846},
}