Abstract
Understanding the underlying processes and extracting detailed characteristics of rivers is critical and has not yet been fully developed. The purpose of this study was to examine the performance of non-linear time series methods on environmental data. Specifically, we performed an analysis of water level measurements, extracted from sensors, located on specified stations along the Nestos River (Greece), with Recurrence Plots (RP) and Recurrence Quantification Analysis (RQA) methods. A more detailed inspection with the sliding windows (epoqs) method was applied on the Recurrence Rate, Average Diagonal Line and Trapping Time parameters, with results showing phase transitions providing useful information about the dynamics of the system. The suggested method seems to be promising for the detection of the dynamical transitions that can characterize distinct time windows of the time series and reveals information about the changes in state within the whole time series. The results will be useful for designing the energy policy investments of producers and also will be helpful for dam management assessment as well as government energy policy.
Highlights
During recent years, the dynamic evolution of river flows has become a very interesting area since it is influenced by many changes in the climate system
For the Recurrence Plot analysis, we proceeded to the estimation of time delay, τ (Average Mutual Information) and the embedding dimension m (False Nearest Neighbor) [39], and the residuals method was used on each time series to eliminate trends from the time series (Table 2)
Quantification Analysis with epoqs were employed to analyze daily time series of the water level of the Nestos River, with Recurrence Quantification Analysis (RQA) parameters locating transitions of the system. Parameters such as Recurrence Rate and Average Line Length in combination with Trapping Time were used in order to extract useful information about the dynamics of the system, as with the case of trends and monthly and annual variations
Summary
During recent years, the dynamic evolution of river flows has become a very interesting area since it is influenced by many changes in the climate system. There are methods developed by solving stochastic differential equations describing the river dynamics in state space form [7], studies based on fractal geometry detecting correlations [8] or using Fourier spectral analysis of daily inflows, outflows and water levels in river dam reservoirs [9]. The evolution of such dynamic systems as river flows is a complex research field where the use of non-linear methods is necessary. Wendi et al [15,16], with the Cross Recurrence Plots method (an extension of the RPs), performed a successful analysis and quantified the event runoff dynamics by studying non-linear correlations
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
