scholarly journals Forecasts of Tropical Pacific Sea Surface Temperatures by Neural Networks and Support Vector Regression

2009 ◽  
Vol 2009 ◽  
pp. 1-13 ◽  
Author(s):  
Silvestre Aguilar-Martinez ◽  
William W. Hsieh
Keyword(s):  
Neural Network ◽  
Support Vector Regression ◽  
Tropical Pacific ◽  
Sea Surface ◽  
Water Volume ◽  
Support Vector ◽  
Neural Network Models ◽  
Error Norm ◽  
Sst Anomalies ◽  
The Tropical Pacific

Two nonlinear regression methods, Bayesian neural network (BNN) and support vector regression (SVR), and linear regression (LR), were used to forecast the tropical Pacific sea surface temperature (SST) anomalies at lead times ranging from 3 to 15 months, using sea level pressure (SLP) and SST as predictors. Datasets for 1950–2005 and 1980–2005 were studied, with the latter period having the warm water volume (WWV) above the 20∘C isotherm integrated across the equatorial Pacific available as an extra predictor. The forecasts indicated that the nonlinear structure is mainly present in the second PCA (principal component analysis) mode of the SST field. Overall, improvements in forecast skills by the nonlinear models over LR were modest. Although SVR has two structural advantages over neural network models, namely (a) no multiple minima in the optimization process and (b) an error norm robust to outliers in the data, it did not give better overall forecasts than BNN. Addition of WWV as an extra predictor generally increased the forecast skills slightly; however, the influence of WWV on SST anomalies in the tropical Pacific appears to be linear.

Comparisons of Support Vector Regression and Neural Network in Modelling the Hydraulic Damper

2011 ◽  
Vol 403-408 ◽  
pp. 3805-3812 ◽  
Author(s):  
Kong Hui Guo ◽  
Xian Yun Wang
Keyword(s):  
Neural Network ◽  
Experimental Data ◽  
Support Vector Regression ◽  
Network Models ◽  
Support Vector ◽  
Neural Network Models ◽  
Hydraulic Damper ◽  
Nonparametric Models ◽  
Force Velocity ◽  
Bpnn Model

Nonparametric models of hydraulic damper based on support vector regression (SVR) are developed. Then these models are compared with two kinds neural network models. One is backpropagation neural network (BPNN) model; another is radial basis function neural network (RBFNN) model. Comparisons are carried out both on virtual damper and actual damper. The force-velocity relation of a virtual damper is obtained based on a rheological model. Then these data are used to identify the characteristics of the virtual damper. The dynamometer measurements of an actual displacement-dependent damper are obtained by experiment. And these data are used to identify the characteristics of this actual damper. The comparisons show that BPNN model is best at identifying the characteristics of the virtual damper, but SVR model is best at identifying the characteristics of the actual damper. The reason is that all experimental data include noise more or less. When the amplitude of the noise is smaller than the parameter of SVR, the noise can not affect the construction of the resulting model. So when training a model based on the experimental data, SVR is superior to other neural networks methods.

scholarly journals Impact of Regional SST Anomalies on the Indian Monsoon Response to Global Warming in the CNRM Climate Model

2006 ◽  
Vol 19 (10) ◽  
pp. 2008-2024 ◽  
Author(s):  
H. Douville
Keyword(s):  
Climate Change ◽  
Climate Model ◽  
Indian Monsoon ◽  
Regional Climate ◽  
Tropical Pacific ◽  
Sea Surface ◽  
Time Slice ◽  
High Latitudes ◽  
Sst Anomalies ◽  
The Tropical Pacific

Abstract While transient climate change experiments with coupled atmosphere–ocean general circulation models undoubtedly represent the most comprehensive tool for studying the climate response to increasing concentrations of greenhouse gases (GHGs), less computationally expensive time-slice experiments with atmospheric GCMs are still useful to test the robustness of the projected climate change. In the present study, three sets of time-slice experiments with prescribed sea surface temperature (SST) are compared to a reference climate scenario obtained with the Centre National de Recherches Météorologiques Coupled Climate Model (CCM). The main objective is to assess the sensitivity of the monsoon response to the magnitude or pattern of SST anomalies in two regions where such anomalies are highly model dependent, namely, the circumpolar Southern Ocean and the tropical Pacific Ocean. On the one hand, it is shown that the regional climate anomalies predicted by the CCM can be reproduced at least qualitatively by a pair of time-slice experiments in which the present-day SST biases of the CCM are removed. On the other hand, the results indicate that the Indian monsoon response to increasing amounts of GHG is sensitive to regional uncertainties in the prescribed SST warming. Increasing the sea surface warming in the southern high latitudes to compensate for the weak sea ice feedback simulated by the CCM around the Antarctic has a significant influence on the regional climate change simulated over India, through a perturbation of the regional Hadley circulation. Prescribing zonal mean rather than El Niño–like SST anomalies in the tropical Pacific has an even stronger impact on the monsoon response, through a modification of the Walker circulation. These results suggest that both deficiencies in simulating present-day climate (even at high latitudes) and uncertainties in the SST patterns caused by enhanced GHG concentrations (especially in the tropical Pacific) are major obstacles for predicting climate change at the regional scale.

Sign in / Sign up

Export Citation Format

Share Document