Articles | Volume 10, issue 6
Hydrol. Earth Syst. Sci., 10, 797–806, 2006
https://doi.org/10.5194/hess-10-797-2006
Hydrol. Earth Syst. Sci., 10, 797–806, 2006
https://doi.org/10.5194/hess-10-797-2006

  30 Oct 2006

30 Oct 2006

Modelling of monsoon rainfall for a mesoscale catchment in North-West India I: assessment of objective circulation patterns

E. Zehe1, A. K. Singh2, and A. Bárdossy3 E. Zehe et al.
  • 1Institute of Geoecology, University of Potsdam, Germany
  • 2Civil Engineering Department, Nirma University of Science & Technology (NU) Ahmedabad – 382 481, India
  • 3Institute of Hydraulic Engineering, University of Stuttgart, Germany

Abstract. Within the present study we shed light on the question whether objective circulation patterns (CP) classified from either the 500 HPa or the 700 HPa level may serve as predictors to explain the spatio-temporal variability of monsoon rainfall in the Anas catchment in North West India. To this end we employ a fuzzy ruled based classification approach in combination with a novel objective function as originally proposed by (Stehlik and BᲤossy, 2002). After the optimisation we compare the obtained circulation classification schemes for the two pressure levels with respect to their conditional rainfall probabilities and amounts. The classification scheme for the 500 HPa level turns out to be much more suitable to separate dry from wet meteorological conditions during the monsoon season. As is shown during a bootstrap test, the CP conditional rainfall probabilities for the wet and the dry CPs for both pressure levels are highly significant at levels ranging from 95 to 99%. Furthermore, the monthly CP frequencies of the wettest CPs show a significant positive correlation with the variation of the total number of rainy days at the monthly scale. Consistently, the monthly frequencies of the dry CPs exhibit a negative correlation with the number of rainy days at the monthly scale. The present results give clear evidence that the circulation patterns from the 500 HPa level are suitable predictors for explaining spatio- temporal Monsoon variability. A companion paper shows that the CP time series obtained within this study are suitable input into a stochastical rainfall model.