FLOOD STAGE FORECASTING WITH SUPPORT VECTOR MACHINES1

ABSTRACT: Machine learning techniques are finding more and more applications in the field of forecasting. A novel regression technique, called Support Vector Machine (SVM), based on the statistical learning theory is explored in this study. SVM is based on the principle of Structural Risk Minimization as opposed to the principle of Empirical Risk Minimization espoused by conventional regression techniques. The flood data at Dhaka, Bangladesh, are used in this study to demonstrate the forecasting capabilities of SVM. The result is compared with that of Artificial Neural Network (ANN) based model for one‐lead day to seven‐lead day forecasting. The improvements in maximum predicted water level errors by SVM over ANN for four‐lead day to seven‐lead day are 9.6 cm, 22.6 cm, 4.9 cm and 15.7 cm, respectively. The result shows that the prediction accuracy of SVM is at least as good as and in some cases (particularly at higher lead days) actually better than that of ANN, yet it offers advantages over many of the limitations of ANN, for example in arriving at ANN's optimal network architecture and choosing useful training set. Thus, SVM appears to be a very promising prediction tool.     

Historical trends of polychlorinated dibenzo-p-dioxins and dibenzofurans in three dated sediment cores from Mexico

This paper describes the first report of dioxins and furans (PCDDs/Fs) in sediment cores from Mexico. Sedimentation rates and vertical fluxes were estimated using ²¹⁰Pb dating. Two cores correspond to marine sediments and one to an endorheic lake. Concentrations of PCDDs/Fs found in the three sites are typical of non-impacted areas with low concentrations when compared to reference values. However the PCDDs/Fs sediment profiles show an increasing concentration trend in the upper core sections. This behavior is different from that found at many sites around the globe where diminishing concentrations have been reported. A strong predominance of OCDD was observed, and a comparison to typical composition profiles of industrial and other sources did not result in clear origin assignments for these measured compounds. We suggest that local sources may be responsible for the increase in concentration and, because these undetermined sources have not been curtailed, their importance is still growing.     

Institutions for a sustainable development—experiences from EU-countries

This article is based on an international comparative study analysing innovations in national administrative institutions. Results of that research are presented and discussed here. Against the background of the institutionalising process of political institutions the study focused on innovations initiated or introduced by governments or parliaments by example of policies and politics supporting sustainable development. The countries involved in the original study were Belgium, Denmark, Finland, France, Germany, the Netherlands, Sweden and the United Kingdom. Three general lessons can be drawn: (1) Countries with a long tradition of integrated environmental policy-making are more open to the concept of sustainability. (2) A (institutional) “cure-all” solution does not seem to exist. It seems necessary to diffuse the concept of sustainability into all spheres of politics and society. Reaching this strategic goal requires a specific policy and polity mix. (3) Strong and well-equipped institutions result from specific policies and politics: they play a central role in promoting and intensifying sustainable development. Successful innovations integrating SD into everyday politics and policies are only a very first step though. Further improvements with regard to effective participation and public debate, global orientation of policies and the measurability of achieved progress are necessary.     

EVIDENCE OF CHAOTIC BEHAVIOR IN SINGAPORE RAINFALL1

ABSTRACT: This paper focuses on the investigation of the existence of chaotic behavior in the Singapore rainfall data. The procedure for the determination of the minimum number of variables essential and the number of variables sufficient to model the dynamics of the rainfall process was studied. An analysis of the rainfall behavior of different time periods was also conducted. The correlation dimension was used as a basis for discriminating stochastic and chaotic behaviors. Daily rainfall records for durations of 30, 20, 10, 5, 4, 3, 2, and 1 years from six stations were analyzed. The delay time for the phase-space reconstruction was computed using the autocorrelation function approach. The results provide positive evidence of the existence of chaotic behavior in the daily rainfall data. The minimum number of variables essential to model the dynamics of the rainfall process was identified to be 3 while the number of variables sufficient to model the dynamics of the rainfall process ranges from 11 to 18. The results also suggest that the attractor dimensions of rainfall data of longer time periods are higher than that of shorter time periods. The study suggests a minimum number of 1500 data points required for the computation of the correlation dimension of the rainfall data.     

APPLICATION OF RORB MODEL TO A CATCHMENT IN SINGAPORE1

ABSTRACT: Runoff Routing model (RORB) is a general model applicable to both rural and urban catchments. The performance of the model is illustrated through its simulation of flood runoff hydrographs in an urban catchment in Singapore. The essential feature of the model is the routing of rainfall excesses on subareas through some arrangement of concentrated storage elements, which represent the distribution of temporary storage of flood runoff on the watershed. This nonlinear routing procedure of the storage elements has two common parameters, k_c and m. With the limited data available, these two parameter values were determined through calibration runs. The same set of values of k_c and m were then used in the model to determine the runoff hydrographs of five other storms selected from the rainfall events between 1979 and 1981. It was found that the simulated runoff hydrographs matched reasonably well with the recorded hydrographs.     

GENETIC PROGRAMMING AND ITS APPLICATION IN REAL‐TIME RUNOFF FORECASTING1

ABSTRACT: Genetic programming (GP), a relatively new evolutionary technique, is demonstrated in this study to evolve codes for the solution of problems. First, a simple example in the area of symbolic regression is considered. GP is then applied to real‐time runoff forecasting for the Orgeval catchment in France. In this study, GP functions as an error updating scheme to complement a rainfall‐runoff model, MIKE11/NAM. Hourly runoff forecasts of different updating intervals are performed for forecast horizons of up to nine hours. The results show that the proposed updating scheme is able to predict the runoff quite accurately for all updating intervals considered and particularly for updating intervals not exceeding the time of concentration of the catchment. The results are also compared with those of an earlier study, by the World Meteorological Organization, in which autoregression and Kalman filter were used as the updating methods. Comparisons show that GP is a better updating tool for real‐time flow forecasting. Another important finding from this study is that nondimensionalizing the variables enhances the symbolic regression process significantly.     

DEVELOPMENT OF PREDICTION EQUATIONS FOR A PHREATIC AQUIFER IN RESPONSE TO INFILTRATION: DIMENSIONAL ANALYSIS1

ABSTRACT: The need for a quick means to predict the subsurface flow into a stream caused the authors to take a different approach to the problem. Dimensional analysis was applied to deduce a functional relationship between the dependent and independent variables. This paper lays the ground work for the development of prediction equations for a phreatic aquifer in response to infiltration.     

GENETIC PROGRAMMING: A NEW PARADIGM IN RAINFALL RUNOFF MODELING1

ABSTRACT: Genetic Programming (GP) is a domain‐independent evolutionary programming technique that evolves computer programs to solve, or approximately solve, problems. To verify GP's capability, a simple example with known relation in the area of symbolic regression, is considered first. GP is then utilized as a flow forecasting tool. A catchment in Singapore with a drainage area of about 6 km² is considered in this study. Six storms of different intensities and durations are used to train GP and then verify the trained GP. Analysis of the GP induced rainfall and runoff relationship shows that the cause and effect relationship between rainfall and runoff is consistent with the hydrologic process. The result shows that the runoff prediction accuracy of symbolic regression based models, measured in terms of root mean square error and correlation coefficient, is reasonably high. Thus, GP induced rainfall runoff relationships can be a viable alternative to traditional rainfall runoff models.