Temporal Transferability of Calibrated Hydrological Model Parameters: A Case Study of Gin Catchment, Sri Lanka

Lack of observed data for model calibration hinders the application hydrological models in many poorly-gauged catchments, particularly humid tropical region. Despite much less attention given, it is vital to assess transferability calibrated parameters order apply such catchments assist their water resources planning and management activities. Thus, this study investigated temporal a lumped conceptual model’s (MIKE 11 NAM) rainfall-runoff simulations two different time periods beyond its period. Study area was selected as Gin catchment located MIKE NAM period 1995-1998 [Nash Sutcliffe coefficient (NSE) = 0.73, percent bias (PBIAS) 3.9%, ratio root mean square error standard deviation measured (RSR) 0.52] validated 1999-2002 (NSE=0.66, PBIAS 8.7%, RSR 0.59) catchment. The tested using scenarios which formulated based on lag between transfer period: scenario A having 4-year B 8-year lag. Scenario evaluated performance 2003-2006 streamflow indicated only marginal loss comparison calibration. It showed an overall ‘good’ (NSE=0.64, 8.6%, including promising capability reproduce peak flows (<10th percentile) with Pearson’s correlation 0.6. However, 2007-2010 ‘unsatisfactory’ (NSE=0.42, 13.6%, 0.76). Therefore, suggests that can be temporally transferred within from implying applicability modelling framework especially where sparse.