Search

Several studies have been devoted to dynamic and statistical downscaling for both climate variability and climate change. This paper introduces an application of temporal neural networks for downscaling global climate model output and autocorrelation functions. This method is proposed for downscaling daily precipitation time series for a region in the Amazon Basin. The downscaling models were developed and validated using IPCC AR4 model output and observed daily precipitation. In this paper, five AOGCMs for the twentieth century (20C3M; 1970–1999) and three SRES scenarios (A2, A1B, and B1) were used. The performance in downscaling of the temporal neural network was compared to that of an autocorrelation statistical downscaling model with emphasis on its ability to reproduce the observed climate variability and tendency for the period 1970–1999. The model test results indicate that the neural network model significantly outperforms the statistical models for the downscaling of daily precipitation variability. [ABSTRACT FROM AUTHOR]

Regression-based statistical downscaling is a method broadly used to resolve the coarse spatial resolution of general circulation models. Nevertheless, the assessment of uncertainties linked with climatic variables is essential to climate impact studies. This study presents a procedure to characterize the uncertainty in regression-based statistical downscaling of daily precipitation and temperature over a highly vulnerable area (semiarid catchment) in the west of Iran, based on two downscaling models: a statistical downscaling model (SDSM) and an artificial neural network (ANN) model. Biases in mean, variance, and wet/dry spells are estimated for downscaled data using vigorous statistical tests for 30 years of observed and downscaled daily precipitation and temperature data taken from the National Center for Environmental Prediction reanalysis predictors for the years of 1961 to 1990. In the case of daily temperature, uncertainty is estimated by comparing monthly mean and variance of downscaled and observed daily data at a 95 % confidence level. In daily precipitation, downscaling uncertainties were evaluated from comparing monthly mean dry and wet spell lengths and their confidence intervals, cumulative frequency distributions of monthly mean of daily precipitation, and the distributions of monthly wet and dry days for observed and modeled daily precipitation. Results showed that uncertainty in downscaled precipitation is high, but simulation of daily temperature can reproduce extreme events accurately. Finally, this study shows that the SDSM is the most proficient model at reproducing various statistical characteristics of observed data at a 95 % confidence level, while the ANN model is the least capable in this respect. This study attempts to test uncertainties of regression-based statistical downscaling techniques in a semiarid area and therefore contributes to an improvement of the quality of predictions of climate change impact assessment in regions of this type. [ABSTRACT FROM AUTHOR]

Summary ¶The climate system can be regarded as a dynamic nonlinear system. Thus, traditional linear statistical methods fail to model the nonlinearities of such a system. These nonlinearities render it necessary to find alternative statistical techniques. Since artificial neural network models (NNM) represent such a nonlinear statistical method their use in analyzing the climate system has been studied for a couple of years now. Most authors use the standard Backpropagation Network (BPN) for their investigations, although this specific model architecture carries a certain risk of over-/ underfitting. Here we use the so called Cauchy Machine (CM) with an implemented Fast Simulated Annealing schedule (FSA) (Szu, 1986) for the purpose of attributing and detecting anthropogenic climate change instead. Under certain conditions the CM-FSA guarantees to find the global minimum of a yet undefined cost function (Geman and Geman, 1986). In addition to potential anthropogenic influences on climate (greenhouse gases (GHG), sulphur dioxide (SO 2)) natural influences on near surface air temperature (variations of solar activity, explosive volcanism and the El Niño/Southern Oscillation phenomenon) serve as model inputs. The simulations are carried out on different spatial scales: global and area weighted averages. In addition, a multiple linear regression analysis serves as a linear reference. It is shown that the adaptive nonlinear CM-FSA algorithm captures the dynamics of the climate system to a great extent. However, free parameters of this specific network architecture have to be optimized subjectively. The quality of the simulations obtained by the CM-FSA algorithm exceeds the results of a multiple linear regression model; the simulation quality on the global scale amounts up to 81% explained variance. Furthermore the combined anthropogenic effect corresponds to the observed increase in temperature Jones et al. (1994), updated by Jones (1999a), for the examined period 1856?1998 on all investigated scales. In accordance to recent findings of physical climate models, the CM-FSA succeeds with the detection of anthropogenic induced climate change on a high significance level. Thus, the CM-FSA algorithm can be regarded as a suitable nonlinear statistical tool for modeling and diagnosing the climate system. [ABSTRACT FROM AUTHOR]