A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Downscaling future projections of monthly precipitation in a catchment with varying physiography
Projections of monthly precipitation data in the Chaliyar river basin for the future scenarios are downscaled. Observed precipitation data at two rain gauge stations located in mid land and high land were compared and found to be highly varying and hence downscaling with respect to the individual stations was adopted. Based on the values of the correlation coefficient between the large-scale atmospheric predictors and observed precipitation data, potential predictors were selected. Predictor-predictand correlation was not similar at the two stations and hence separate sets of potential predictors were selected for both the stations and for three different seasons. Artificial Neural Network-based transfer functions were derived between the potential predictors and the observed precipitation data for each station and for each season and these were subsequently validated. The validated models were used to downscale the data projected by three GCMs, viz. CGCM3, BCM2 and FUB-EGMAM, for the scenarios SRES A1B, A2 and B1 at two rain gauge stations for the period from 2001 to 2100. Results show an increase in precipitation at both the stations for the scenarios SRES A1B, A2 and B1 during the south-west monsoon period. In the north-east monsoon and pre-monsoon, more number of extreme events was present.
Downscaling future projections of monthly precipitation in a catchment with varying physiography
Projections of monthly precipitation data in the Chaliyar river basin for the future scenarios are downscaled. Observed precipitation data at two rain gauge stations located in mid land and high land were compared and found to be highly varying and hence downscaling with respect to the individual stations was adopted. Based on the values of the correlation coefficient between the large-scale atmospheric predictors and observed precipitation data, potential predictors were selected. Predictor-predictand correlation was not similar at the two stations and hence separate sets of potential predictors were selected for both the stations and for three different seasons. Artificial Neural Network-based transfer functions were derived between the potential predictors and the observed precipitation data for each station and for each season and these were subsequently validated. The validated models were used to downscale the data projected by three GCMs, viz. CGCM3, BCM2 and FUB-EGMAM, for the scenarios SRES A1B, A2 and B1 at two rain gauge stations for the period from 2001 to 2100. Results show an increase in precipitation at both the stations for the scenarios SRES A1B, A2 and B1 during the south-west monsoon period. In the north-east monsoon and pre-monsoon, more number of extreme events was present.
Downscaling future projections of monthly precipitation in a catchment with varying physiography
Chithra, N. R. (author) / Thampi, Santosh G. (author)
ISH Journal of Hydraulic Engineering ; 23 ; 144-156
2017-05-04
13 pages
Article (Journal)
Electronic Resource
English
Bayesian Learning and Relevance Vector Machines Approach for Downscaling of Monthly Precipitation
| Online Contents | 2015
Bayesian Learning and Relevance Vector Machines Approach for Downscaling of Monthly Precipitation
| British Library Online Contents | 2015