Classification of hydro-meteorological conditions and multiple artificial neural networks for streamflow forecasting

Self Cite

Abstract. The formulation of objective procedures for the delineation of homogeneous groups of catchments is a fundamental issue in both operational and research hydrology. For assessing catchment similarity, a variety of hydrological information may be considered; in this paper, gauged sites are characterised by a set of streamflow signatures that include a representation, albeit simplified, of the properties of fine time-scale flow series and in particular of the dynamic components of the data, in order to keep into account the sequential order and the stochastic nature of the streamflow process.The streamflow signatures are provided in input to a clustering algorithm based on unsupervised SOM neural networks, obtaining groups of catchments with a clear hydrological distinctiveness, as highlighted by the identification of the main patterns of the input variables in the different classes and the interpretation of their interrelations. In addition, even if no geographical, morphological nor climatological information is provided in input to the SOM network, the clusters exhibit an overall consistency as far as location, altitude and precipitation regime are concerned.In order to assign ungauged sites to such groups, the catchments are represented through a parsimonious set of morphometric and pluviometric variables, including also indexes that attempt to synthesise the variability and correlation properties of the precipitation time series, thus providing information on the type of weather forcing that is specific to each basin. Following a principal components analysis, needed for synthesizing and better understanding the morpho-pluviometric catchment properties, a discriminant analysis finally assigns the ungauged catchments, through a leave-one-out cross validation, to one of the above identified hydrologic response classes. The approach delivers a quite satisfactory identification of the membership of ungauged catchments to the streamflow-based classes, since the comparison of the two cluster sets shows a misclassification rate of around 20 %.Overall results indicate that the inclusion of information on the properties of the fine time-scale streamflow and rainfall time series may be a promising way for better representing the hydrologic and climatic character of the study catchments.

Section: Classification Of Streamflow Signatures With Som Neural Netwmentioning

confidence: 99%

Catchment classification based on characterisation of streamflow and precipitation time series

Toth

2013

Self Cite

“…Concerning the problems of classification and pattern recognition, Self Organising Maps (SOM's, Kohonen, 1982;1997) are an unsupervised learning method to analyze, cluster, and model various types of large databases. The SOM method counts several hydrological applications (see e.g., Kalteh et al, 2008;Céréghino and Park, 2009), such as classification of hydrological and meteorological conditions for streamflow forecasting (Toth, 2009). SOM networks cluster groups of similar input patterns from a high dimensional input space in a non-linear fashion onto a low dimensional (most commonly two-dimensional for representation and visualization purposes) discrete lattice of neurons in an output layer (Kohonen, 2001;Kalteh et al, 2008).…”

Section: Som Network and Catchment Classificationmentioning

confidence: 99%

“…A very interesting and promising approach to classification makes use of an innovative and data-driven classification method based on unsupervised artificial neural networks (ANNs), known as Self Organising Maps (SOM, Kohonen, 1982, Toth, 2009Ley et al, 2011).…”

Section: Introductionmentioning

confidence: 99%

Data-driven catchment classification: application to the pub problem

Prinzio

Castellarin

Toth

2011

Self Cite

Abstract.A promising approach to catchment classification makes use of unsupervised neural networks (Self Organising Maps, SOM's), which organise input data through non-linear techniques depending on the intrinsic similarity of the data themselves. Our study considers ∼300 Italian catchments scattered nationwide, for which several descriptors of the streamflow regime and geomorphoclimatic characteristics are available. We compare a reference classification, identified by using indices of the streamflow regime as input to SOM, with four alternative classifications, which were identified on the basis of catchment descriptors that can be derived for ungauged basins. One alternative classification adopts the available catchment descriptors as input to SOM, the remaining classifications are identified by applying SOM to sets of derived variables obtained by applying Principal Component Analysis (PCA) and Canonical Correlation Analysis (CCA) to the available catchment descriptors. The comparison is performed relative to a PUB problem, that is for predicting several streamflow indices in ungauged basins. We perform an extensive cross-validation to quantify nationwide the accuracy of predictions of mean annual runoff, mean annual flood, and flood quantiles associated with given exceedance probabilities. Results of the study indicate that performing PCA and, in particular, CCA on the available set of catchment descriptors before applying SOM significantly improves the effectiveness of SOM classifications by reducing the uncertainty of hydrological predictions in ungauged sites.

“…Herbst et al (2009a) used SOMs for hydrological model evaluation and optimization. Toth (2009) used SOM to classify hydro-meteorological catchment conditions for streamflow forecasting. Kalteh et al (2008) gives an overview of various applications of SOM in hydrology e.g.…”

Section: Introductionmentioning

confidence: 99%

Catchment classification by runoff behaviour with self-organizing maps (SOM)

Ley

Casper

Hellebrand

et al. 2011

119

Abstract. Catchments show a wide range of response behaviour, even if they are adjacent. For many purposes it is necessary to characterise and classify them, e.g. for regionalisation, prediction in ungauged catchments, model parameterisation.In this study, we investigate hydrological similarity of catchments with respect to their response behaviour. We analyse more than 8200 event runoff coefficients (ERCs) and flow duration curves of 53 gauged catchments in RhinelandPalatinate, Germany, for the period from 1993 to 2008, covering a huge variability of weather and runoff conditions. The spatio-temporal variability of event-runoff coefficients and flow duration curves are assumed to represent how different catchments "transform" rainfall into runoff. From the runoff coefficients and flow duration curves we derive 12 signature indices describing various aspects of catchment response behaviour to characterise each catchment.Hydrological similarity of catchments is defined by high similarities of their indices. We identify, analyse and describe hydrologically similar catchments by cluster analysis using Self-Organizing Maps (SOM). As a result of the cluster analysis we get five clusters of similarly behaving catchments where each cluster represents one differentiated class of catchments.As catchment response behaviour is supposed to be dependent on its physiographic and climatic characteristics, we compare groups of catchments clustered by response behaviour with clusters of catchments based on catchment properties. Results show an overlap of 67 % between these two pools of clustered catchments which can be improved using the topologic correctness of SOMs.