obtained for the following factors: landscape
structure of river basins, basin lateral slope,
precipitation, temperature, arable land area.
Both RSR < 0.60 (RMSE-standard deviation
ratio) and NSE > 0.65 (Nash–Sutcliffe model
efficiency coefficient) estimated for each of the
developed runoff models represent their good
or very good performance. The most probable
water and hydrochemical runoff can be forecasted for 3–4 months ahead with a twice
reduced variance as compared with the similar
forecast by the observed mean runoff. The
elaborated balance models with anew selected
landscapes and the updated values of parameters can be applied to any mountainous area
and allow to estimate and manage the seasonal
and long-term dynamics of water quality.
Keywords
Water quality Á Hydrochemical runoff Á
Mountain rivers Á System-analytical
modeling Á GIS Á Altai-Sayan
7.1 Introduction
The problem of water resources management
under the influence of climate change is constantly growing in importance (Deng et al. 2015).
The present-day water management calls for the
development of adequate mathematical models to
assess quantitatively hydrological and hydrochemical processes in river basins. The models
should take into account both temporal and
spatial effects of natural and anthropogenic factors on these processes (Loucks and Beek 2005;
Singh et al. 2014). The complicated orographic
structure of mountain areas and space–time
composition of climatic fields impede the task
solution greatly (Sevast'yanov 1998).
Water runoff (WR) and hydrochemical runoff
(HCR) are of prime interest for management. For
their assessment, various methods (among them
distributed hydrological modeling) making
allowance for physiographic and hydrographic
features of river catchments were developed
(Jarrett 1990; Koren et al. 2004; Sene 2008). The
application of GIS technologies makes it possible
to automatize the calculations to a greater or a
lesser degree. For instance, they are used to
calculate spatial (areal) distribution of such climatic elements as precipitation, temperature, air
humidity, etc., which are essential for WR estimation but monitored only in some sites of the
territory under study. GIS technologies used for
data interpolation often have errors close to that
of traditional statistical analysis (Konovalova
et al. 2006). The same holds true for the methods
that estimate hydrological characteristics (Brus
1993). In the case of mountains, a computational
error of areal meteorological characteristics is
predominantly responsible for the inaccuracy of
WR/HCR models.
At present, GIS tools are combined with
complex physicomathematical models describing
hydrological and hydrochemical processes.
The GIS is applied here for processing the spatially distributed input data and visualization of
the derived spatial distributions. In turn, the
models consist of differential equations of
hydrodynamics, hydrochemistry, and mathematical physics supplemented by experimentally
found relationships between elements of the
modeled system (Vinogradov and Vinogradova
2010; Rumynin 2013). The differential equations
require detailed spatially distributed data on
precipitation, air temperature, evaporation and
plant transpiration, subsurface aquifers, basin
morphometry, properties of water-saturated soils
and rocks, flow resistance on slopes and in the
river network, etc. For example, in different parts
of the mountain river basin, the following processes can occur at the same time: rain, snow or
absence of any precipitation, snow melting, and
substantial change in evaporation with altitude.
All this important information is practically
absent for mountain countries, and complex
physically grounded models become unavailing
(Vinogradov and Vinogradova 2010). The simplest hydrological and hydrochemical models,
which ignore this information, also become
useless due to an unacceptable increase in calculation error. Instead of such models, it is reasonable to practice just the long-term mean value
of observed river WR and HCR.
80
Y. Kirsta and A. Puzanov
structure of river basins, basin lateral slope,
precipitation, temperature, arable land area.
Both RSR < 0.60 (RMSE-standard deviation
ratio) and NSE > 0.65 (Nash–Sutcliffe model
efficiency coefficient) estimated for each of the
developed runoff models represent their good
or very good performance. The most probable
water and hydrochemical runoff can be forecasted for 3–4 months ahead with a twice
reduced variance as compared with the similar
forecast by the observed mean runoff. The
elaborated balance models with anew selected
landscapes and the updated values of parameters can be applied to any mountainous area
and allow to estimate and manage the seasonal
and long-term dynamics of water quality.
Keywords
Water quality Á Hydrochemical runoff Á
Mountain rivers Á System-analytical
modeling Á GIS Á Altai-Sayan
7.1 Introduction
The problem of water resources management
under the influence of climate change is constantly growing in importance (Deng et al. 2015).
The present-day water management calls for the
development of adequate mathematical models to
assess quantitatively hydrological and hydrochemical processes in river basins. The models
should take into account both temporal and
spatial effects of natural and anthropogenic factors on these processes (Loucks and Beek 2005;
Singh et al. 2014). The complicated orographic
structure of mountain areas and space–time
composition of climatic fields impede the task
solution greatly (Sevast'yanov 1998).
Water runoff (WR) and hydrochemical runoff
(HCR) are of prime interest for management. For
their assessment, various methods (among them
distributed hydrological modeling) making
allowance for physiographic and hydrographic
features of river catchments were developed
(Jarrett 1990; Koren et al. 2004; Sene 2008). The
application of GIS technologies makes it possible
to automatize the calculations to a greater or a
lesser degree. For instance, they are used to
calculate spatial (areal) distribution of such climatic elements as precipitation, temperature, air
humidity, etc., which are essential for WR estimation but monitored only in some sites of the
territory under study. GIS technologies used for
data interpolation often have errors close to that
of traditional statistical analysis (Konovalova
et al. 2006). The same holds true for the methods
that estimate hydrological characteristics (Brus
1993). In the case of mountains, a computational
error of areal meteorological characteristics is
predominantly responsible for the inaccuracy of
WR/HCR models.
At present, GIS tools are combined with
complex physicomathematical models describing
hydrological and hydrochemical processes.
The GIS is applied here for processing the spatially distributed input data and visualization of
the derived spatial distributions. In turn, the
models consist of differential equations of
hydrodynamics, hydrochemistry, and mathematical physics supplemented by experimentally
found relationships between elements of the
modeled system (Vinogradov and Vinogradova
2010; Rumynin 2013). The differential equations
require detailed spatially distributed data on
precipitation, air temperature, evaporation and
plant transpiration, subsurface aquifers, basin
morphometry, properties of water-saturated soils
and rocks, flow resistance on slopes and in the
river network, etc. For example, in different parts
of the mountain river basin, the following processes can occur at the same time: rain, snow or
absence of any precipitation, snow melting, and
substantial change in evaporation with altitude.
All this important information is practically
absent for mountain countries, and complex
physically grounded models become unavailing
(Vinogradov and Vinogradova 2010). The simplest hydrological and hydrochemical models,
which ignore this information, also become
useless due to an unacceptable increase in calculation error. Instead of such models, it is reasonable to practice just the long-term mean value
of observed river WR and HCR.
80
Y. Kirsta and A. Puzanov