Hydroclimatological approach for Monthly Streamflow Prediction using Genetic programming

Created by W.Langdon from gp-bibliography.bib Revision:1.4192

  author =       "Rajib Maity and S. S. Kashid",
  title =        "Hydroclimatological approach for Monthly Streamflow
                 Prediction using Genetic programming",
  journal =      "The Indian Society for Hydraulics journal of Hydraulic
  year =         "2009",
  volume =       "15",
  number =       "2",
  pages =        "89--107",
  keywords =     "genetic algorithms, genetic programming",
  ISSN =         "0971-5010",
  URL =          "http://www.tandfonline.com/doi/abs/10.1080/09715010.2009.10514943",
  DOI =          "doi:10.1080/09715010.2009.10514943",
  size =         "19 pages",
  abstract =     "An approach for monthly streamflow prediction is
                 illustrated in this paper using the concept of
                 hydroclimatological association. Rainfall-runoff
                 relationship over a catchment is very complex, which
                 may not be revealed very easily. This is due to the
                 fact that streamflow is significantly influenced by
                 catchment characteristics, land-use pattern, spatial
                 distribution of rainfall, evapotranspiration over the
                 catchment, water retention over the basin, etc. Keeping
                 the other factors more or less constant over a
                 sufficiently small temporal span (say monthly),
                 intensity and spatial distribution of rainfall plays a
                 major role behind the streamflow variation. Oceans
                 happen to be the major source of moisture for the
                 precipitation and the rainfall distribution over the
                 continents is proved to be linked with Sea Surface
                 Temperature (SST) and various large-scale atmospheric
                 circulation patterns across the globe. Thus, the
                 variation of basin-scale streamflow is expected to be
                 influenced by these large-scale climatological factors,
                 which is investigated in this paper for the Narmada
                 River basin. The information of El Nino-Southern
                 Oscillation (ENSO) from the tropical Pacific Ocean and
                 Equatorial Indian Ocean Oscillation (EQUINOO) from the
                 tropical Indian Ocean is investigated 1) for their
                 possible influence behind the monthly streamflow
                 variation of Narmada River at central India and 2) the
                 efficacy of genetic programming (GP), which is an
                 artificial intelligence technique, for the prediction
                 of monthly streamflow through the concept of
                 hydroclimatological approach. The results of the study
                 indicate that GP-derived streamflow forecasting models
                 that use historical average of monthly streamflow and
                 the large-scale atmospheric circulation information,
                 for basin-scale streamflow prediction are quite
                 satisfactory. The coefficient of determination for
                 monthly streamflow in case of Narmada River was found
                 to be 0.921 for training and 0.836 for testing, which
                 is quite promising for such a complex system",
  notes =        "rajib@civil.iitb.ac.in, now at Indian Institute of
                 Technology, Kharagpur (rajib@civil.iitkgp.emet.in) 2.
                 Department of Civil Engg., liT Bombay,",

Genetic Programming entries for Rajib Maity Sathishkumar Shahajirao Kashid