Non linear dynamic analysis of Koyna reservoir evaporation

Abstract

The aim of the present study is to characterize the behaviour of the important variable in a reservoir namely the volume of evaporation loss. Many water resource investigations require an estimate of evaporation from reservoirs in order to efficiently manage the available resources. However, evaporation is difficult to measure experimentally over water surfaces such as reservoir and several techniques and models have been used in the past for its determination. Recently researchers are applying data driven techniques to predict this loss. In such cases it is mostly assumed that this reservoir evaporation has a non-linear relationship internally in the observed time series. In the present study the daily evaporation loss from Koyna reservoir in Maharashtra, India was analyzed to characterize its behaviour. Forty nine years of daily estimated reservoir evaporation values are employed to examine its behaviour by applying correlation dimension method. The reservoir evaporation is estimated from a deterministic equation which is entirely based on water spread area and is considered to be a deterministic series; however it may happen that the self generation of a simple deterministic equation can be chaotic, thus the simple deterministic series may behave like a chaotic deterministic series. Hence there are chances that the process of reservoir evaporation may also behave chaotic nature. The results arrived from correlation dimension method turned out that the reservoir evaporation is showing a chaotic behaviour.