@article{fdi:010083365,
  title = {{P}arsimonious models of precipitation phase derived from random forest knowledge : intercomparing logistic models, neural networks, and random forest models},
  author = {{C}ampozano, {L}. and {R}obaina, {L}. and {G}ualco, {L}. {F}. and {M}aisincho, {L}. and {V}illacis, {M}. and {C}ondom, {T}homas and {B}allari, {D}. and {P}aez, {C}.},
  editor = {},
  language = {{ENG}},
  abstract = {{T}he precipitation phase ({PP}) affects the hydrologic cycle which in turn affects the climate system. {A} lower ratio of snow to rain due to climate change affects timing and duration of the stream flow. {T}hus, more knowledge about the {PP} occurrence and drivers is necessary and especially important in cities dependent on water coming from glaciers, such as {Q}uito, the capital of {E}cuador (2.5 million inhabitants), depending in part on the {A}ntisana glacier. {T}he logistic models ({LM}) of {PP} rely only on air temperature and relative humidity to predict {PP}. {H}owever, the processes related to {PP} are far more complex. {T}he aims of this study were threefold: (i) to compare the performance of random forest ({RF}) and artificial neural networks ({ANN}) to derive {PP} in relation to {LM}; (ii) to identify the main drivers of {PP} occurrence using {RF}; and (iii) to develop {LM} using meteorological drivers derived from {RF}. {T}he results show that {RF} and {ANN} outperformed {LM} in predicting {PP} in 8 out of 10 metrics. {RF} indicated that temperature, dew point temperature, and specific humidity are more important than wind or radiation for {PP} occurrence. {W}ith these predictors, parsimonious and efficient models were developed showing that data mining may help in understanding complex processes and complements expert knowledge.},
  keywords = {precipitation phase ; {A}ndes precipitation ; random forest ; logistic models ; automatic discovery ; {EQUATEUR} ; {ANDES} ; {ANTISANA} {VOLCAN}},
  booktitle = {},
  journal = {{W}ater},
  volume = {13},
  numero = {21},
  pages = {3022 [23 p.]},
  year = {2021},
  DOI = {10.3390/w13213022},
  URL = {https://www.documentation.ird.fr/hor/fdi:010083365},
}