Please login first
Investigation into the relationship between annual precipitation and runoff to the Tsimlyansk reservoir in the context of potential climatic changes
* , *
1  "Land reclamation and water management complex" department, Federal Research Center of Hydraulic Engineering and Land Reclamation named after A.N. Kostyakov, Moscow, 127550 Russia
Academic Editor: Luis Garrote

Abstract:

Many reservoirs employ multi-year regulation operational regimes. This raises the problem of the relationship between annual runoff and the primary runoff-forming factor (precipitation), as well as their long-term variability due to possible climatic changes.

The present study concerns the water balance of the Tsimlyansky reservoir, situated in the lower reaches of the Don River, which is subject to multi-year flow regulation. The objective of this study is to quantify the relationship between precipitation and inflow to the reservoir, expressed as annual values.

Statistical methods, including correlation analysis and data smoothing using moving averages, as well as wavelet transform filtering, were used to address the set tasks.

This study employed multi-year time series observations of precipitation and runoff to the reservoir, comprising a 56-year series for precipitation and a 142-year series for annual runoff. All series were formed as water management years (a 12 month cycle, from April to March of the following calendar year), in accordance with the water users’ requirements.

In the course of this research, the correlation coefficient between the precipitation time series and a part of the annual runoff series of corresponding duration (years 1966/67-2021/2022) was determined. The correlation coefficient was found to be equal to 0.46, with a mean square deviation of 0.13, indicating a significant relationship between annual precipitation and annual runoff to the reservoir.

The studies also revealed cyclic components in both precipitation and runoff time series with a period of approximately 12 years. Cyclic components in the time series were detected using the following two distinct methodologies: a combination of data filtering using the moving average method with calculation of autocorrelation functions of smoothed series, and filtering the data using wavelet analysis methods. The presence of such components requires the results obtained to be taken into account when modelling the long-term use of reservoir water resources.

Keywords: precipitation; river runoff; reservoir; climate change
Top