Search

Abstract: Nature-inspired metaheuristics found various applications in different fields of science, including the problem of artificial neural networks (ANN) training. However, very versatile opinions regarding the performance of metaheuristics applied to ANN training may be found in the literature. Both nature-inspired metaheuristics and ANNs are widely applied to various geophysical and environmental problems. Among them the water temperature forecasting in a natural river, especially in colder climate zones where the seasonality plays important role, is of great importance, as water temperature has strong impact on aquatic life and chemistry. As the impact of possible future climate change on water temperature is not trivial, models are needed to allow projection of streamwater temperature based on simple hydro-meteorological variables. In this paper the detailed comparison of the performance of nature-inspired optimization methods and Levenberg–Marquardt (LM) algorithm in ANNs training is performed, based on the case study of water temperature forecasting in a natural stream, namely Biala Tarnowska river in southern Poland. Over 50 variants of 22 various metaheuristics, including a large number of Differential Evolution, as well as some Particle Swarm Optimization, Evolution Strategies, multialgorithms and Direct Search methods are compared with LM algorithm on ANN training for the described case study. The impact of population size and some control parameters of particular metaheuristics on the ANN training performance are verified. It is found that despite widely claimed large improvement in nature-inspired methods during last years, the vast majority of them are still outperformed by LM algorithm on the selected problem. The only methods that, based on this case study, seem competitive to LM algorithm in terms of the final performance (but not speed) are Differential Evolution algorithms that benefit from the concept of Global and Local neighborhood-based mutation operators. The streamwater forecasting performance of the neural networks is adequate, the major prediction errors are related to the river freezing and melting processes that occur during winter in the mountainous catchment under study. [Copyright &y& Elsevier]

Abstract: Previous studies have shown that it is very difficult to distinguish human-induced changes from those caused by natural forcing. In this paper we try to quantify the influence of land use and water management on flows of the Upper Narew River in north-east Poland. Apart from climatic and land use changes, the Upper Narew catchment was changed by the construction of a storage reservoir at Siemianówka, near Bondary, on the upstream reach of the river. We apply four different approaches to analysing the changes in flow regime and catchment response for the periods before and after reservoir construction. First we estimate the cumulative distribution functions for low and high flow events. The second approach is a time series analysis of flow variation over the whole length of available data and the derivation of cumulative distribution functions for the flows and 0.25–0.75 quantiles followed by a statistical analysis of the number of events below and above the thresholds and their duration. The third approach consists of the application of the Wittenberg baseflow separation method and tests for changes in baseflow indices. In the fourth approach an analysis of changes in flow regime is performed by studying the changes in transfer function-based flow model parameters. Long-term changes in land use are assessed using previous studies of the catchment and the analysis of Corine land cover data and government yearbooks. The results show that different methods explain different aspects of changes in the catchment and flow regime due to climatic changes and changes in land use and water management practices. The analysis of cumulative distribution functions gave evidence of the influence of Siemianówka reservoir on low flows which was also confirmed by the low flow analysis using the Wittenberg approach. The STF analysis of flows indicates the existence of changes in flow regime that can be attributed to the roughness changes in the channel. [Copyright &y& Elsevier]

Abstract: This paper shows that maximum snow depth and the length of accumulation and ablation periods observed at the local scale of Hornsund, SW Spitsbergen, are partly explained by monthly and seasonal values of the AO and NAO indices in the given and previous hydrological years. This analysis is followed by an application of a statistically efficient lumped parameter time series approach to modelling the dynamics of snow depth, based on daily meteorological and snow depth measurements from the same area. A dynamic Stochastic Transfer Function (STF) model is developed that follows the Data Based Mechanistic approach, where a stochastic data-based identification of model structure and an estimation of its parameters are followed by a physical interpretation. Apart from snow depth estimates, the model provides also the uncertainty limits. An analysis of the variation in parameter estimates over the whole measurement period provides an insight into the possible influence of recent climate change on snow cover dynamics at Hornsund. To help explain the physical meaning of the model parameters, we classified the data into accumulation and ablation periods. The models were run for each period separately. The first order model structure was found to be the most suitable to explain the variability of the snow cover. The cross-validation of models performance on the other years shows that the predictive value of the obtained models is not very consistent, with a mixture of good and bad years. The analysis shows that variability in the NAO and AO indices, reflecting the changes in global circulation patterns, is reproduced by local, physically meaningful, STF model-derived parameters in the form of residence times and temperature and precipitation related gains. [Copyright &y& Elsevier]

• Depending on stream temperature model, different warming of streamwater is projected. • Projection from particular model occasionally differ from the majority of models. • A number of stream temperature models should be used together for future climates. • Stream temperature in analyzed streams is to be warmed by 2–3.5 °C by 2100 (RCP 8.5). • In specific streams marginal cooling in projected in some months until 2050. In the majority of studies aiming at stream temperature warming due to climate change just a single water temperature model is used. Choosing a single model may highly impact the conclusions from the study. In this paper four relatively different empirical or semi-empirical models: perceptron neural networks, product unit networks, extended logistic regression and air2stream were applied to project the impact of climate change on water temperature in rivers located in temperate climatic zones of the USA and Poland. The models were driven by daily air temperature and streamflow projected by the rainfall-runoff model. In the first step, the models were calibrated and validated. Then the projections of water temperature were derived for the historical periods and two future periods taking into account: (a) climate simulations from the CORDEX initiative (NA-CORDEX and EURO-CORDEX), (b) the GR4J rainfall-runoff model and (c) different water temperature models. The obtained results indicate that due to global warming, the stream temperatures are expected to increase by about 1–2 °C for 2021–2050 and by 2–3 °C for 2071–2100 periods. These changes are not uniformly distributed throughout the year. The largest warming in the USA is found in the summertime, in Poland – in spring and autumn. For some months the discrepancies in the projected stream temperature between various stream temperature models are large. Product unit neural network, logistic regression-based model or air2stream occasionally led to projections that differ from those obtained by the majority of models even by 2 °C. We strongly recommend using at least a few stream temperature models for analysing the impact of climate change on water temperatures or the fate of the aquatic ecosystem. [ABSTRACT FROM AUTHOR]

The effect of methods of statistical downscaling of daily precipitation on changes in extreme flow indices under a plausible future climate change scenario was investigated in 11 catchments selected from 9 countries in different parts of Europe. The catchments vary from 67 to 6171 km 2 in size and cover different climate zones. 15 regional climate model outputs and 8 different statistical downscaling methods, which are broadly categorized as change factor and bias correction based methods, were used for the comparative analyses. Different hydrological models were implemented in different catchments to simulate daily runoff. A set of flood indices were derived from daily flows and their changes have been evaluated by comparing their values derived from simulations corresponding to the current and future climate. Most of the implemented downscaling methods project an increase in the extreme flow indices in most of the catchments. The catchments where the extremes are expected to increase have a rainfall-dominated flood regime. In these catchments, the downscaling methods also project an increase in the extreme precipitation in the seasons when the extreme flows occur. In catchments where the flooding is mainly caused by spring/summer snowmelt, the downscaling methods project a decrease in the extreme flows in three of the four catchments considered. A major portion of the variability in the projected changes in the extreme flow indices is attributable to the variability of the climate model ensemble, although the statistical downscaling methods contribute 35–60% of the total variance. [ABSTRACT FROM AUTHOR]

Summary A number of methods have been proposed for the prediction of streamwater temperature based on various meteorological and hydrological variables. The present study shows a comparison of few types of data-driven neural networks (multi-layer perceptron, product-units, adaptive-network-based fuzzy inference systems and wavelet neural networks) and nearest neighbour approach for short time streamwater temperature predictions in two natural catchments (mountainous and lowland) located in temperate climate zone, with snowy winters and hot summers. To allow wide applicability of such models, autoregressive inputs are not used and only easily available measurements are considered. Each neural network type is calibrated independently 100 times and the mean, median and standard deviation of the results are used for the comparison. Finally, the ensemble aggregation approach is tested. The results show that simple and popular multi-layer perceptron neural networks are in most cases not outperformed by more complex and advanced models. The choice of neural network is dependent on the way the models are compared. This may be a warning for anyone who wish to promote own models, that their superiority should be verified in different ways. The best results are obtained when mean, maximum and minimum daily air temperatures from the previous days are used as inputs, together with the current runoff and declination of the Sun from two recent days. The ensemble aggregation approach allows reducing the mean square error up to several percent, depending on the case, and noticeably diminishes differences in modelling performance obtained by various neural network types. [ABSTRACT FROM AUTHOR]