Assessment of nutrient load on the Cheboksary Reservoir using the results of modeling runoff and removal of biogenic elements from the pilot catchments
- Authors: Yasinskiy S.V.1, Kondratyev S.A.2, Shmakova M.V.2, Kashutina E.A.1, Rasulova A.M.2
-
Affiliations:
- Institute of Geography of the Russian Academy of Sciences
- Institute of Limnology of the Russian Academy of Sciences
- Issue: No 3 (2024)
- Pages: 130-141
- Section: Articles
- URL: https://journals.rcsi.science/2658-3518/article/view/282590
- DOI: https://doi.org/10.31951/2658-3518-2024-A-3-130
- ID: 282590
Cite item
Full Text
Abstract
The paper presents the results of an approximate assessment of the nutrient load on the Cheboksary reservoir of the Volga Cascade from the right-bank and left-bank parts of the catchment basin. The main solution tools are the recognition of the underlying surfaces in the catchment and mathematical modeling. The catchment basins of the Kudma (right-bank tributary) and Linda (left-bank tributary) rivers have been identified as pilot sites. The catchment basins of the Kudma (right-bank tributary) and Linda (left-bank tributary) rivers have been identified as pilot sites. The representativeness of the selected pilot sites for the catchment basins of the Cheboksary reservoir is demonstrated. The mathematical basis of the research was the “precipitation-runoff-removal” model describing the formation of runoff and removal of biogenic elements from the catchment basin. To calibrate the mathematical model, the materials of observations of water discharges and the content of chemical agents in the gauging sections of the pilot catchment basins were used. The modeling data provides an approximate estimate of the total nutrient load on the Cheboksary reservoir, as well as the contribution of natural nitrogen and phosphorus removal to the load from the catchment basin.
Full Text
1. Introductioon
Anthropogenic eutrophication is one of the significant issues in the Volga Cascade reservoirs (Mineeva et al., 2020). The reason is the intense anthropogenic nutrient load from the catchment basin. At the final stages of anthropogenic eutrophication in eutrophic and hypereutrophic water bodies, a disbalance may occur in the ratio of productive and destructive processes within the aquatic ecosystem. This leads to the emergence of oxygen-free (anaerobic) zones, fish kills phenomena, a reduction in fish stocks, and water pollution with toxic substances as a result of the development of certain species of phytoplankton, “blooming” the water (Rossolimo, 1977; Anthropogenic eutrophication..., 1982). At the same time, the current monitoring system is unable to provide an accurate assessment of the nitrogen and phosphorus inflow to the reservoirs from tributaries due to the limited number of measurement points for water discharge and hydrochemical characteristics.
The aim of the paper is to provide an approximate assessment of the loading of total nitrogen (Ntotal) and total phosphorus (Ptotal) on the Cheboksary reservoir formed in the catchment area, based on mathematical modeling using available observational data on the formation of runoff and removal of chemicals in pilot catchment basins.
The Cheboksary reservoir is formed on the Volga River by the Cheboksary hydroelectric plant, located in the city of Novocheboksarsk. The length of the reservoir is 341 km, the area is 2190 km2, the total volume of water is 13.9 km3, the area of its own catchment basin without the upstream Volga basin is 131.9 thousand km2. The Linda (left tributary of the Volga, its length is 122 km, its catchment area is 1682 km2) and the Kudma (right tributary of the Volga, its length is 144 km, its catchment area is 3248 km2) catchments in the basin of the Cheboksary reservoir were selected as pilot sites reflecting the main patterns of biogenic element removal on the basis of expert assessment (Fig.1). The catchment basins are representative of the forested left bank and the agriculturally developed right bank of the Cheboksary reservoir. Moreover, karstic phenomena are widespread in the Kudma basin, which also influence the formation of runoff and the removal of nitrogen and phosphorus. There are gauging sections of state hydrological and hydrochemical monitoring of the Russian meteorological service in the selected rivers, which enables using these data for the calibration of mathematical models.
Fig.1. The location of the Kudma and Linda pilot catchments: 1 – hydrographic network, 2 –Linda River catchment, 3 –Kudma River catchment.
2. Research methods
At the initial stage of the study, the hypothesis concerning the suitability of the selected pilot catchments was tested. As is well-known, the removal of nutrients, namely nitrogen and phosphorus, from the catchment area is mainly determined by the structure of the underlying surface (Kondratyev and Shmakova, 2019; Khrisanov and Osipov, 1993). This applies to both natural (background) removal, which is formed in the parts of the catchment not affected by human impact (forest), and to anthropogenic removal (agricultural and urbanized areas). Therefore, the criterion for the correct selection of a pilot catchment is the similarity of their underlying surface structure with the structure of the basin as a whole.
In the present study, the differentiation of the land cover was carried out using global archives of satellite data of the underlying surfaces. The land cover classification of the Cheboksary reservoir basin and pilot catchments was based on the Copernicus Global Land Service Collection 3 (CGLS) (Buchhorn et al., 2021a; Buchhorn et al., 2020b). The CGLS collection is formed from satellite imagery from PROBA-V (PROBA-Vegetation) and Sentinel-2 with spatial resolutions of 100/110/300 m. The depth of the archives of satellite images used to create the CGLS data collection is from 2015-01-01 to 2020-12-31. The UN Land Cover Classification System (LCCS) was used to classify the land surfaces, which are in the CGLS archive. The main data source is PROBA-V multispectral satellite imagery with a temporal resolution of 5 days and a spatial resolution of 100 m of surface reflectivity at the Top-of-Canopy (TOC). The secondary data source is the daily PROBA-V multispectral satellite imagery with a spatial resolution of 300 m of surface reflectivity. Their median composite is made to archive regular 5-day images at 100-meter and 300-meter spatial resolution of the PROBA-V time series. This is necessary because the PROBA-V satellite provides daily global coverage for data with a spatial resolution of 300 m, which corresponds to 5-day coverage for the same data with a spatial resolution of 100 m. The identification of the types of underlying surfaces of the CGLS collection is based on spectral indices (Mousaei Sanjerehei, 2014), other global data archives (Pekel et al., 2016), and the WorldDEM™ digital elevation model. The detailed classification algorithm used to obtain the CGLS data collection is given in (Buchhorn et al., 2020a).
This study area includes 15 land cover classes that are necessary for calculating the load on a water body (Table 1). The Table 1 demonstrates that the difference in the percentage of different surface classes of the pilot sites and the corresponding parts of the whole reservoir basin does not exceed 6%. The above correspondence of the land cover classes of the pilot catchments with the structure of the right- and left-bank parts of the Cheboksary reservoir basin confirms the legitimacy of the chosen objects as pilot ones. Besides, for the main classes of the underlying surface, based on the analysis of literature data (Rossolimo, 1977; Anthropogenic eutrophication..., 1982; Pozdnyakov et al., 2020), the emission characteristics of the intake of nutrients into the runoff are approximately estimated.
Table 1. Comparison of land cover classes of the pilot catchments with the right- and left-bank parts of the Cheboksary reservoir basin
Land cover classes | Kudma River catchment, % | Right Bank, % | Linda River catchment, % | Left Bank, % |
Shrubs | 0.00 | 0.00 | 0.00 | 0.00 |
Herbaceous vegetation | 10.49 | 16.38 | 3.18 | 2.94 |
Cultivated and managed vegetation / agriculture | 37.96 | 37.00 | 11.96 | 8.85 |
Urban / built up | 3.73 | 2.15 | 1.73 | 0.52 |
Bare/sparse vegetation | 0.00 | 0.00 | 0.00 | 0.00 |
Permanent water bodies | 0.24 | 0.72 | 0.07 | 1.19 |
Herbaceous wetland | 0.01 | 0.07 | 0.01 | 0.06 |
Closed forest, evergreen needle leaf | 1.88 | 4.03 | 10.49 | 12.23 |
Closed forest, deciduous broad leaf | 23.00 | 23.28 | 33.07 | 37.78 |
Closed forest, mixed. | 10.04 | 6.96 | 25.22 | 25.32 |
Closed forest that not matching any of the other definitions | 1.45 | 1.63 | 1.71 | 3.61 |
Open forest, evergreen needle leaf | 0.00 | 0.02 | 0.01 | 0.04 |
Open forest, deciduous broad leaf | 0.20 | 0.05 | 0.53 | 0.44 |
Open forest, mixed | 1.76 | 1.56 | 1.69 | 0.30 |
Open forest that does not correspond to any of the other definitions | 9.22 | 6.15 | 10.33 | 6.72 |
Total catchment area, km2 | 3248 | 75687 | 1682 | 56176 |
The main mechanisms for attaining the goal were a mathematical model of runoff formation of ILHM and a model of biogenic elements removal of ILLM developed at the Institute of Limnology RAS and modified with the participation of Federal State Budgetary Scientific Institution “Federal Scientific Agroengineering Center” (Kondratyev and Shmakova, 2019; Yasinskiy et al., 2020).
The runoff model, ILHM (Institute of Limnology Hydrological Model, Certificate of State Registration No. 2015614210) (Kondratyev and Shmakova, 2019), is designed for calculations of hydrographs of snowmelt and rainfall runoff from the catchment area, as well as water levels in the waterbody. The model has a conceptual base and describes the processes of snow accumulation and snowmelt, evaporation and soil moisture in the aeration zone, runoff formation, as well as runoff within a homogeneous catchment, the characteristics of which are assumed to be constant for the entire area. The model can function with a monthly time step and with an annual time step. During the simulation, the catchment is represented as a homogeneous simulated storage, accumulating incoming water and then gradually allowing it to flow way. The values of the basic parameters of the hydrological model, determining the shape of the hydrograph of the runoff, are determined depending on the percentage of water body, that is, as the ratio of the area of the water area to the overall area of the catchment. The model has been verified in a lot of catchments located in Russia (Tigoda, Lizhma, Syanga, Olonka, Sunah, Shuya, Ojat, Sjas, Vuoksa, Svir, Velikaya, Kazanka, Svijaga, and Neva Rivers) and Finland (Mustajoki and Harajoki Rivers) (Kondratyev and Shmakova, 2019).
The model of nutrient removal, ILLM (Institute of Limnology Loading Model, Certificate of State Registration No. 2014612519), was developed based on existing modeling of runoff and the removal of biogenic elements from the catchment areas and nutrient inputs into the water bodies (Kondratyev and Shmakova, 2019; Behrendt and Dannowski, 2007; Behrendt and Opitz, 1999). The recommendations of the HELCOM for assessing the load on water bodies in the Baltic Sea were also built into the model (Guidelines..., 2015). The model is designed to solve problems associated with the quantification of nutrient loading formed by point and nonpoint sources of pollution and the forecast of its changes under the influence of possible anthropogenic and climatic changes. The model incorporates the existing capabilities of data input from the state monitoring system of water bodies as well as data from state statistical reporting on wastewater discharges and agricultural activities in the catchment areas. The model also allows the calculation of the removal of biogenic elements from the catchment under the influence of hydrological factors and retention by the catchment. The final result of the modeling is an evaluation of the nutrient load and its components on the received water body from the catchment. The model of nutrient load has been verified at several catchment basins in Russia (Velikaya, Luga, Mga, Izhora, Slavyanca, Sestra, Shuya, Vodla, Sunah, Kazanka, Svijaga, and Irtysh Rivers) (Kondratyev and Shmakova, 2019; 80 years of limnological research..., 2023). The materials of the Helsinki Commission (Applied methodology..., 2019) describe models that can be used to calculate the external load on the water objects of the Baltic Sea basin. These include the ILLM model.
The combination of ILHM and ILLM models is a “precipitation-runoff-removal” model that transforms meteorological parameters (precipitation and air temperature) into runoff (water discharges) and the removal of biogenic elements from the catchment area, depending on the characteristics of the land cover classes and the intensity of external influences of a natural and anthropogenic nature. At the same time, the ILLM model provides for the calculation of natural (background) load. In accordance with the HELCOM definition (Guidelines..., 2015), the natural load of biogenic elements is formed due to their removal from non-cultivated lands and part of the removal from cultivated lands, which occurs independently of economic activity.
3. Results and Discussion
Inputs to the calculations require information on precipitation and air temperature, areas of various types of underlying surface forming a diffuse removal of nutrients, the intensity of point sources of nutrient loading, the atmospheric load of nitrogen and phosphorus, the number of animals and poultry in the catchment area, as well as applied mineral and organic fertilizers.
As noted earlier, in order to inform the model, a classification of the underlying surface types in the catchment was carried out. The values of phosphorus and nitrogen concentrations in runoff from various types of underlying surfaces were set according to field studies conducted in 2018 and 2019 by employees of the Institute of Limnology RAS (Pozdnyakov et al., 2020). To assess the contribution of point sources to the nutrient load on the lake, data from statistical forms of state reporting 2TP (vodkhoz) were used. Sufficiently high values of emission coefficients and concentrations in runoff from urbanized areas are an expression of the contribution of a dispersed rural population without connection to sewerage networks and treatment facilities (Behrendt and Opitz, 1999). The atmospheric load was set in accordance with the research materials of Kazan Federal University (Minakova et al., 2019), and no separation into natural and anthropogenic components was made. To calibrate the model, data from observations of runoff and water quality at the corresponding posts of the state monitoring of the Russian meteorological service in the closing reaches of the Kudma and Linda Rivers for the period from 2008 to the present were used.
The agricultural nitrogen and phosphorus loading on the catchments was estimated by the methodology presented in (Bryukhanov et al., 2016). According to this method, the following main factors in the formation of nutrient load on agricultural fields were taken into account in the calculations:
- content of nitrogen and phosphorus in the soil, the share of their removal from the total content in the soil;
- amount of nitrogen and phosphorus in the composition of mineral fertilizers and their emission coefficient;
- amount of nitrogen and phosphorus in the composition of organic fertilizers and their emission coefficient;
- distance of the contour of agricultural land from water objects;
- soil type by origin;
- soil type by mechanical composition;
- structure of farmland (ratio of arable land and perennial grasses, meadows, pastures, and deposits).
Calculations of the agricultural nitrogen and phosphorus load were performed only for pilot catchments. For the right- and left-bank parts of the reservoir catchment, the load values were recalculated in proportion to the area ratio.
The calibration of the “precipitation-runoff-removal” model on the pilot catchments of the Kudma and Linda rivers is presented below. Figure 2 shows the observed and calculated runoff from the catchments of the Linda River (gauge Vasilkovo) and Kudma River (gauge Kstovo). The Nash-Sutcliffe criterion is 78% for Linda and 67% for Kudma, which confirms the adequacy of the model for the described processes of runoff in the catchment area.
Fig.2. The observed (1) and calculated (2) average monthly runoff layers from the Kudma (a) and Linda (b) catchments.
Table 2 shows the results of calibration of the ILLM model according to the correspondence of the average long-term values of the removal of biogenic elements with the runoff in the gauge-stations of the Kudma and Linda Rivers, which also confirms the correspondence of the simulation results to the available monitoring data.
Table 2. The results of the model calibration in the Kudma (a) and Linda (b) catchments according to the annual nitrogen and phosphorus removal in the closing gauge-stations
Source of information | Kudma – gauge Kstovo | Linda – gauge Vasilkovo | ||
Ntot, t/year | Ptot, t/year | Ntot, t/year | Ptot, t/year | |
Average annual values (according to monitoring data) | 715.69 | 18.00 | 443.73 | 9.16 |
Calibration results | 717 | 18.00 | 442 | 9.11 |
When calculating the nitrogen and phosphorus removal from the right- and left-bank parts of the Cheboksary reservoir catchment, a model calibrated on the corresponding pilot catchments was used. As input information about the structure of the underlying surface of the reservoir catchment, data from the satellite image recognition mentioned above were used. The missing information on the sources of anthropogenic load was set based on the assumption that the load is proportional to the areas of the considered catchments exposed to anthropogenic impact. The results of an approximate assessment of the nutrient load on the Cheboksary reservoir based on data from modeling the runoff and nitrogen and phosphorus removal from pilot catchments are presented in Table 3.
Table 3. The results of an approximate assessment of the nutrient load on the Cheboksary reservoir, based on nitrogen and phosphorus removal modeling for the pilot catchments
Calculation results | Ntotal | Ptotal |
Right bank of the Cheboksary reservoir (area – 75,687 km2, average long-term runoff – 100 mm/year) | ||
Nutrient load on the reservoir (t/year) | 16729 | 425 |
Natural (background) component (t/year) | 2287 | 80.8 |
Removal module (kg/km2 year) | 222 | 5.6 |
Left bank of the Cheboksary reservoir (area – 56176 km2, average long-term runoff – 188 mm/year) | ||
Nutrient load on the reservoir (t/year) | 14591 | 300 |
Natural (background) component (t/year) | 4111 | 118 |
Removal module (kg/km2 year) | 262 | 5.4 |
According to the calculation results, the average long-term nutrient load on the Cheboksary reservoir is approximately estimated at 31320 tN/year and 725 tP/year. At the same time, the specific load from the right-bank part is 222 kg N/km2 year and 5.6tP/km2 year; from the left-bank part, it is 262 kg N/km2 year and 5.4 tP/km2 year. The presented modeling data do not contradict the results of other studies on nutrient removal by tributaries of the reservoir (Yasinskiy et al., 2020).
4. Conclusion
Thus, an approximate assessment of the nutrient load on a large water body from a catchment area that is not sufficiently sanctified by monitoring observations is possible under the following conditions:
- A reasonably well-studied analogue pilot catchment has been identified that has a point of hydrological and hydrochemical measurements at the trailing gauge-station and a similar land surface structure to the main catchment (e.g. % area of the main land surface classes);
- The selected mathematical model “precipitation - runoff – removal” is provided with information on the main sources of nutrient load on the hydrographic network and is calibrated for the closing gauge of the pilot catchment.
In this case, the lack of information on the sources of anthropogenic load on the whole catchment in the calculations can be compensated by data on the pilot site. In this case, the assumption of proportionality of catchment area parameters is made.
At the same time, it should be remembered that in future studies, in order to increase the degree of reliability, the modeling results should be compared with regular ground-based monitoring of runoff and water quality, which should be carried out on the main tributaries of the reservoir.
Acknowledgements
The paper was supported by the grant of the Russian National Science Foundation No. 22-17-00224 “Formation of hydrological and geochemical processes in the catchments of the cascades of the Upper Volga and Kama reservoirs under various scenarios of land use and climate change in their territories”.
Conflict of interest
The authors declare no conflicts of interest.
About the authors
S. V. Yasinskiy
Institute of Geography of the Russian Academy of Sciences
Email: arasulova@limno.ru
ORCID iD: 0000-0002-2478-8256
Russian Federation, Staromonetny Lane, 29/4, Moscow, 119017
S. A. Kondratyev
Institute of Limnology of the Russian Academy of Sciences
Email: arasulova@limno.ru
ORCID iD: 0000-0003-1451-8428
Russian Federation, Sevastyanova, Str., 9, St. Petersburg, 196105
M. V. Shmakova
Institute of Limnology of the Russian Academy of Sciences
Email: arasulova@limno.ru
Russian Federation, Sevastyanova, Str., 9, St. Petersburg, 196105
E. A. Kashutina
Institute of Geography of the Russian Academy of Sciences
Email: arasulova@limno.ru
ORCID iD: 0000-0003-0181-5036
Russian Federation, Staromonetny Lane, 29/4, Moscow, 119017
A. M. Rasulova
Institute of Limnology of the Russian Academy of Sciences
Author for correspondence.
Email: arasulova@limno.ru
ORCID iD: 0000-0003-4400-2000
Russian Federation, Sevastyanova, Str., 9, St. Petersburg, 196105
References
- years of limnological research at Institute of Limnology of the Russian Academy of Sciences. 2023. In: Kondratyev S.A. (Ed.). Mosсow: Russian Academy of Sciences. doi: 10.12731/978-5-907645-12-7 (in Russian)
- Anthropogenic eutrophication of Lake Ladoga. 1982. In: Petrova N.A. (ed.). Leningrad: Nauka. (in Russian)
- Applied methodology for the PLC-6 assessment, Baltic Marine Environment Protection Commission. 2019. HELCOM, Helsinki.
- Behrendt H., Dannowski R. 2007. Nutrients and heavy metals in the Odra River system. Berlin: Weissensee Verlag Publ.
- Behrendt H., Opitz D. 1999. Retention of nutrients in river systems: dependence on specific runoff and hydraulic load. Hydrobiologia 410: 111–122.
- Bryukhanov A.Yu., Kondratyev S.A., Oblomkova N.S. et al. 2016. Methodology for determining the nutrient load of agricultural production on water bodies. Tekhnologii i tekhnicheskie sredstva mekhanizirovannogo proizvodstva produkcii rastenievodstva i zhivotnovodstva [Technologies and technical means of mechanized production of crop and livestock products] 89: 175–183. (in Russian)
- Buchhorn M., Bertels L., Smets B. et al. 2020. Copernicus Global Land Service: Land Cover 100m: version 3 Globe 2015-2019: Algorithm Theoretical Basis Document. Zenodo. Dataset v3.0. doi: 10.5281/zenodo.3938968
- Buchhorn M., Lesiv M., Tsendbazar N.-E. 2020. Copernicus Global Land Cover Layers – Collection 2. Remote Sensing 12(108): 1044. doi: 10.3390/rs12061044
- Guidelines for the annual and periodical compilation and reporting of waterborne pollution inputs to the Baltic Sea (PLC-Water). 2015. In: Svendsen L.M. (ed.). Helsinki: HELCOM Publ.
- Khrisanov N.I., Osipov G.K. 1993. Management of eutrophication of water bodies. St. Petersburg: Gidrometeoizdat. (in Russian)
- Kondratyev S.A., Shmakova M.V. 2019. Mathematical modeling of mass transfer in the watershed-watercourse-reservoir system. St. Petersburg: Nestor-Istoriya. (in Russian)
- Minakova E.A., Shlychkov A.P., Pozdnyakov Sh.R. et al. 2019. Assessment of the amount of nutrient deposition from the atmosphere in a private catchment of the Kuibyshev reservoir within the Republic of Tatarstan. Problemy regional’noj ekologii [Problems of regional ecology] 1: 34–38. (in Russian)
- Mineeva N.M., Semadeni I.V., Makarova O.S. 2020. Chlorophyll content and current trophic state of the river reservoirs. Volga (2017–2018). Biologiya vnutrennikh vod [Biology of inland waters] 2: 205–208. (in Russian)
- Mousaei Sanjerehei M. 2014. Assessment of spectral vegetation indices for estimating vegetation cover in arid and semiarid shrublands. Range Management and Agroforestry 35(1): 91-100.
- Pekel J., Cottam A., Gorelick N. et al. 2016. High-resolution mapping of global surface water and its long-term changes. Nature 540: 418-422.
- Pozdnyakov Sh.R., Bryukhanov A.Yu., Kondratyev S.A. et al. 2020. Perspectives of the Reduction of Nutrient Export from River Watersheds through the Introduction of Best Available Technologies for Agricultural Production: Based on Modeling Results. Water Resources 47(5): 771-784.
- Rossolimo L.L. 1977. Changes in limnic ecosystems under the influence of anthropogenic factors. Mosсow: Nauka. (in Russian)
- Yasinskiy S.V., Kashutina E.A., Sidorova M.V. et al. 2020. Anthropogenic Load and the Effect of Drainage Area on the Diffuse Runoff of Nutrients into a Large Water Body: Case Study of the Cheboksary Reservoir. Water Resources 47(5): 810–827.
Supplementary files
