Inﬂuence of Anthropogenic Loads on Surface Water Status: A Case Study in Lithuania

: Twenty-six water bodies and 10 ponds were selected for this research. Anthropogenic loads were assessed according to pollution sources in individual water catchment basins. It was determined that 50% of the tested water bodies had N total values that did not correspond to the good and very good ecological status classes, and 20% of the tested water bodies had P total values that did not correspond to the good and very good ecological status classes. The lake basins and ponds received the largest amounts of pollution from agricultural sources with total nitrogen at 1554.13 t/year and phosphorus at 1.94 t/year, and from meadows and pastures with total nitrogen at 9.50 t/year and phosphorus at 0.20 t/year. The highest annual load of total nitrogen for lake basins on average per year was from agricultural pollution from arable land (98.85%), and the highest total phosphorus load was also from agricultural pollution from arable land (60%).


Introduction
To protect people and ecosystems, the European Commission has adopted a zero pollution action plan to prevent air, water and soil pollution. The European Commission encourages sustainability in agriculture and rural areas across the EU through the common agricultural policy, by which agriculture is environmentally sustainable, protects natural resources and enhances biodiversity. Despite the introduction of legislation such as the EU Nitrates and Water Framework Directives [1,2], agricultural practices are often still regarded as a major influencing factor on poor water quality across many EU member states [3], and agricultural systems and activities will put pressure on freshwater resources [4]. A scientific evaluation of the suitability and cost-effectiveness of options to reduce nutrient loss from rural areas to surface waters at the catchment scale, including the feasibility of the options under different climatic and geographical conditions, was conducted in different EU countries [5]. Agricultural catchments are where farm and landscape management interact with policy and science, especially with regard to the implementation and evaluation of agro-environmental regulation [6]. The increasing rate of water resource usage results in its contamination by wastewater from domestic, industrial and agricultural sectors. The opinions of different researchers confirm that agricultural pollution poses a considerable challenge to grain security and human health, especially in economically developed areas [7]. Various studies show that the excessive use of chemical fertilizers and pesticides has caused serious nonpoint-source water pollution [8]. Diffuse water pollution and other water pollution problems are a major problem in many agroecosystems, especially in irrigated areas linked to ecosystems of high ecological value [9]. Requirements for high-quality water bodies are different, but identifying the spatiotemporal characteristics of water quality and related driving factors is essential for lake water quality management [10]. The quality of water and nature is determined for ground and surface water by first measuring their N and P contents and, for natural areas, by measuring N deposition [11]. Lakes are one of the lem in many agroecosystems, especially in irrigated areas linked to ecosystems of high ecological value [9]. Pollution abatement policies imply the modification of agricultural measures and are usually rejected by farmers because their impact on farm profitability in the environment is more difficult to control than point-source pollution [32]. Another opportunity in many countries to reduce surface water pollution is organic farming. For example, conventional intensive agricultural practices have led to serious nonpoint-source pollution in China and have caused severe ecological threats to agroecosystems, especially in ecologically sensitive areas [33]. Evaluating the relationship between the use of water resources and investment in environmental protection, as well as establishing a reasonable model of ecological compensation, has become an imperative issue [34]. The benefits of using more environmentally friendly practices outweigh other issues when evaluating the environmental impact [35].
The aim of this research is to evaluate the impact of anthropogenic load on indicators of the ecological status of water bodies.

Study Area
The ecological status of water bodies at risk was assessed in accordance with the Procedure for Assessing the Ecological Status of Surface Water Bodies [36].
To determine the risk of water bodies in good condition not complying with water quality standards, 26 lakes and 10 ponds were studied. The investigated lakes and ponds are shown in Figure 1. ferred from agricultural land to drainage ditches and larger water bodies, which causes significant degradation of water quality in rivers and lakes [31]. Diffuse pollution causes serious problems due to its large scale and diffuse nature. Diffuse water pollution is a major problem in many agroecosystems, especially in irrigated areas linked to ecosystems of high ecological value [9]. Pollution abatement policies imply the modification of agricultural measures and are usually rejected by farmers because their impact on farm profitability in the environment is more difficult to control than point-source pollution [32]. Another opportunity in many countries to reduce surface water pollution is organic farming. For example, conventional intensive agricultural practices have led to serious nonpoint-source pollution in China and have caused severe ecological threats to agroecosystems, especially in ecologically sensitive areas [33]. Evaluating the relationship between the use of water resources and investment in environmental protection, as well as establishing a reasonable model of ecological compensation, has become an imperative issue [34]. The benefits of using more environmentally friendly practices outweigh other issues when evaluating the environmental impact [35]. The aim of this research is to evaluate the impact of anthropogenic load on indicators of the ecological status of water bodies.

Study Area
The ecological status of water bodies at risk was assessed in accordance with the Procedure for Assessing the Ecological Status of Surface Water Bodies [36].
To determine the risk of water bodies in good condition not complying with water quality standards, 26 lakes and 10 ponds were studied. The investigated lakes and ponds are shown in Figure 1.

Sample Preparation (Collection)
Surveys of physicochemical quality indicators and suspended solids were performed in January-December 2015-2019, while benthic macroinvertebrate and fish studies were performed in August-September 2015-2019.

Assessment of Pollution Sources
Sources of diffuse pollution include loads of manure and mineral fertilizers generated in agriculture and loads from the residents whose households are not connected to sewage collection systems.
Sources of concentrated pollution include discharges from cities, settlements, industrial enterprises and surface water, as well as from storm water runoff from urban areas.
Concentrated pollution can enter water bodies directly or through tributaries. The following factors were assessed: • Domestic and industrial wastewater dischargers and their pollution loads and the impacts on the status of the water body and the average total N and total P values of wastewater discharged in 2015-2019. Environmental Protection Agency (EPA) data on wastewater dischargers, measured pollutant concentrations and annual wastewater volumes are assessed by dividing their values into the feeding basins of the water bodies. • Number of people connected to the sewage collection systems and sewage management (i.e., central, individual or no management (statistics)). The contamination loads in the environment released by the residents whose wastewater was not collected were assessed according to HELCOM recommendations, which specified that one resident generates 4.4 kg according to total N t and 0.9 kg according to total P. • Current number of organic farms, percent; current area of organic farms, ha; we used data from the national ecological agricultural production certification body, ECOAGROS.

•
To determine the nutrient loads from residential and commercial areas, SWAT (soil and water assessment tool) model data were used to calculate pollution loads. The SWAT model is a basin-scale, continuous time model that operates on a daily time step and evaluates the impact of management practices on water, sediment and agricultural chemical yields in ungauged basins [37]. The model includes weather, hydrology, erosion, soil temperature, plant growth, nutrients, land management, channels and reservoir routing.

•
To determine the effect of the transformation of biogenic materials in soil and water body contamination, a SWAT model was used to calculate the average total nitrogen (total N) and total phosphorus (total P) washout.

Statistical Analyses
To assess statistically significant impacts on quality factors related to the ecological status of water bodies, the impacts of anthropogenic load indicators (P and N from arable land, t/year; P and N from meadows, t/year; P and N from cities, t/year; P and N from forests, t/year; P and N from households not connected to sewage networks, t/year; P and N from municipal wastewater, t/year; P and N from surface wastewater t/year; P and N from organic farms, t/year; current number of organic farms, percent; current area of organic farms, ha; basin area, ha; conditional number of livestock; agricultural land, ha; arable land area, ha; and number of farms in the basin, units) on water quality indicators (Y) for the water in lakes and ponds were determined. The multiple linear regression model was as follows: The coefficient bj shows how much the value of Y increases (decreases) by one unit as xj increases when the remaining xk are fixed. t is the Student's criterion, according to which we determine whether the coefficients bj differ statistically significantly from zero, and according to this, we decide whether the predicted values depend on xj. The standardized coefficients beta are used to determine the relative influence of independent variables on the predicted Y. In absolute terms, a higher Beta coefficient indicates a greater dependence of Y on xj.
The regression model is appropriate because: • The Levene test was used for endogeneity test; R code was used to generate the analyses in this section, Significance shows t tests p < 0.05. • All variance inflation factors (VIF) ≤ 4 (no multicollinearity problems). • All Cook measure values ≤ 1.

Lake and Pond Condition Evaluation According to Total Nitrogen and Total Phosphorus Values
Nitrogen (N) pollution from agriculture is considered a major issue and will remain an issue for the foreseeable future [38]. Concentrations of total nitrogen (total N) and phosphorus (total P) in lakes and ponds are shown in Figure 2.
It was found that the total N values in 50% of the surveyed water bodies did not correspond to the good and very good ecological status class values: 7.5% corresponded to the moderate ecological status class value, 22.5% to the poor status class value and 17.5% to the very poor status class value. Twenty percent of the tested water bodies had Ptotal values that did not correspond to the good and very good ecological status class values. Of these, 7.5% corresponded to the moderate ecological status class value, 5% to the poor status class value and 7.5% to the very poor status class value.

Assessment of Nutrient Loads in Lakes and Ponds
Nutrient loads in lake and pond basins were calculated using SWAT model data. Calculations were performed in tonnes per year for the inflows into the lakes for total nitrogen and for total phosphorus. Total nitrogen and phosphorus loads in lake basins (t/year) are presented in Figure 3.
It was found that the total N values in 50% of the surveyed water bodies did not correspond to the good and very good ecological status class values: 7.5% corresponded to the moderate ecological status class value, 22.5% to the poor status class value and 17.5% to the very poor status class value. Twenty percent of the tested water bodies had Ptotal values that did not correspond to the good and very good ecological status class values. Of these, 7.5% corresponded to the moderate ecological status class value, 5% to the poor status class value and 7.5% to the very poor status class value.

Assessment of Nutrient Loads in Lakes and Ponds
Nutrient loads in lake and pond basins were calculated using SWAT model data. Calculations were performed in tonnes per year for the inflows into the lakes for total nitrogen and for total phosphorus. Total nitrogen and phosphorus loads in lake basins (t/year) are presented in Figure 3. Lake and pond basins received the largest amounts of pollution from agricultural sources: arable land-total nitrogen 1554.13 t/year and 1.94 t/year phosphorus and grassland pastures-total nitrogen9.50 t/year and 0.20 t/year phosphorus. Residents whose sewage was not discharged into sewage treatment systems generated 2.38 t/year of total nitrogen and 0.55/year of phosphorus.
Percentage distribution of total nitrogen and phosphorus loads in lake basins (t/year) are presented in Figure 4.  Lake and pond basins received the largest amounts of pollution from agricultural sources: arable land-total nitrogen 1554.13 t/year and 1.94 t/year phosphorus and grassland pastures-total nitrogen 9.50 t/year and 0.20 t/year phosphorus. Residents whose sewage was not discharged into sewage treatment systems generated 2.38 t/year of total nitrogen and 0.55/year of phosphorus.
Percentage distribution of total nitrogen and phosphorus loads in lake basins (t/year) are presented in Figure 4.
The highest annual total nitrogen load for lake basins on average per year was from agricultural pollution from arable land (98.85%), background pollution from forests (0.94%), background pollution from urban areas (0.39%), pollution from meadows and pastures (60%), pollution from residents who were not connected to sewage systems (0.15%), surface water (0.05%) and municipal wastewater (0.03%).
The highest annual load of total phosphorus in lake basins was from agricultural pollution from arable land (60%), populations not connected to sewage networks (17.04%), agricultural pollution from meadows and pastures (6.26%), background pollution from urban areas (5.81%), background pollution from forests (5.67%), municipal wastewater (3.78%) and surface wastewater (1.33%).  The highest annual total nitrogen load for lake basins on average per year was from agricultural pollution from arable land (98.85%), background pollution from forests (0.94%), background pollution from urban areas (0.39%), pollution from meadows and pastures (60%), pollution from residents who were not connected to sewage systems (0.15%), surface water (0.05%) and municipal wastewater (0.03%).

Effects of Anthropogenic Loading on Total Phosphorus, Total Nitrogen Concentration, Chlorophyll "a" EQS Values, Taxonomic Composition and the Abundance of Macrophytes, Zoobenthos and Ichthyofauna
The effect of anthropogenic loading on total phosphorus concentration total P (Y) was calculated by multiple regression analysis. The results are presented in Table 1. A multiple regression analysis of the influence of anthropogenic loads in basins on the concentration of total phosphorus in the water showed that the total P value was influenced by the discharge of municipal and surface wastewater and the existing area of organic farms (p < 0.05). The higher the total P concentration in municipal and surface wastewater was, the higher the total P value was in the water (positive function), and the larger the area of organic farms in the lake feeding basins was, the lower the total P concentration was in the water (negative function). No statistically significant effect of other studied factors on total phosphorus concentration in water was found.
The effect of anthropogenic loads on the total nitrogen concentration, total N (Y), was calculated by multiple regression analysis. The results are presented in Table 2.
Multiple regression analysis of the influence of anthropogenic loads on the total nitrogen concentration in the water showed that the total N value was affected by agricultural land, and that the total N amount was generated from arable land and meadows, households that were not connected to sewage networks, discharged municipal wastewater, water body feeding area and the existing area of organic farms (p < 0.05). The higher the concentration of total N from arable land not connected to the sewerage network, households and municipal wastewater and the larger the areas of the water body basin and agricultural land were, the higher the value of total N in the water (positive function), and the higher the total N concentration from grassland pastures and the larger the area of organic farms in feeding basins were, the lower the total N concentration in the water (negative function).
The effect of anthropogenic loads on the EQR (Y) values of chlorophyll "a" was calculated by multiple regression analysis. The results are presented in Table 3.
A multiple regression analysis of the influence of anthropogenic loads in basins on the EQS values of chlorophyll "a" in the water showed that the EQF values were influenced by the relative number of animals in the basin (p < 0.05). The higher the relative number of animals in the basin was, the lower the EQS value of chlorophyll "a" in the water and the worse the water status (negative function).  The effect of anthropogenic loading on macrophyte taxonomic composition and abundance, MEI (Y), was calculated by multiple regression analysis. The results are presented in Table 4.
Multiple regression analysis of the influence of anthropogenic loads in basins on the taxonomic composition and abundance of macrophytes, MEI, in the water showed that MEI values were influenced by the relative number of animals and the area of agricultural land in the basin (p < 0.05). The higher the relative number of animals was, the larger the area of agricultural land in the basin. When the taxonomic composition and abundance of macrophytes in the water were lower, the water status was worse (negative function).
The effect of anthropogenic loads on the taxonomic composition, abundance and age of ichthyofauna, LFI (lake fish index) (Y), was calculated by multiple regression analysis. The results are presented in Table 5.
A multiple regression analysis of the influence of anthropogenic loads in basins on the taxonomic composition, abundance and age of ichthyofauna, LFI, in the water showed that the LFI values were affected by the current percentage of organic farms in the basin (p < 0.05). The higher the percentage of organic farms in the basin was, the higher the taxonomic composition, abundance and age of ichthyofauna in the water and the better the water status (positive function).
The effect of anthropogenic loads on the taxonomic composition and abundance (Y) of zoo benthos was calculated by multiple regression analysis. The results are presented in Table 6.
Multiple regression analysis of the influence of anthropogenic loads in basins on the taxonomic composition and abundance of zoo benthos in the water showed that the zoo benthos values were influenced by the current percentage of organic farms in the basin, basin area and arable land area (p < 0.05). The higher the percentage of organic farms in the basin was, the higher the taxonomic composition and abundance values of zoobenthos in the water and the better the water status (positive function). The greater the basin area and the arable land area in the basin were, the lower the taxonomic composition and abundance values of zoobenthos in the water and the poorer the water status (negative function).

Discussion
With the relatively high discharges of N and P from agriculture, industry and wastewater treatment plants, this study found that 50% of the water bodies surveyed did not meet the good and very good ecological status values for total N, and 20% of the studied water bodies did not meet the good and very good ecological status values for total P [38]. Agricultural nitrogen flows in a reservoir watershed have implications for water pollution mitigation [39]. The study found that the highest total nitrogen load in lake basins per year on average is from agricultural pollution from arable land (98.85%), and the highest total phosphorus load is also from agricultural pollution from arable land (60%). Similar results were obtained during the simulation. It was established that, on average, 90% of all nitrogen got to the Mūša subcatchment from the scattered pollution sources. It included 87% of the arable land and just a little more than 3% of the forest territory and pastures. 10% of all nitrogen of the subcatchment came from the point-pollution sources. The highest amounts of all phosphor of the Mūša subcatchment got to the subcatchment from the point-pollution sources (about 49%), ploughed 18 land -(36%) and just a little more than 15% from the forestall territory and pastures [40].
Pollution loads are divided into groups according to their sources: diffuse pollution from agricultural lands (arable land, meadows and pastures); minimal pollution from organic farms [41]; background pollution from natural habitats (forests) not connected to the sewage system; and pollution from household wastewater, urban wastewater [42], municipal and industrial wastewater [43,44] and surface (rain) wastewater that are not connected to treatment plants. Climate change might have profound effects on nitrogen (N) dynamics in the cultivated landscape as well as on N transport in streams and the eutrophication of lakes [45].
A multiple regression analysis of the influence of anthropogenic loads in basins on the concentration of total phosphorus in water showed that the total P value was influenced by the discharge of municipal and surface wastewater and the existing area of organic farms. This result is consistent with the results obtained by another author [46,47]. The decrease in total phosphorus and nitrogen concentrations in the water was due to the increase in the area of organic farms using nitrogen (N), phosphorus (P) and potassium (K) preparations [48]. Multiple regression analysis of the influence of anthropogenic loads in basins on the taxonomic composition, abundance and age of ichthyofauna, LFI, in the water showed that the LFI values were influenced by the current percentage of organic farms in the basin, and such trends have been evaluated by other researchers [49]. It should be noted that this variable is also influenced by seasonality, water body structure [50] and anaerobic substances [51]. Research shows that organic farming can improve surface water quality [33].

1.
The total N values did not correspond to the good and very good ecological status classes in 50% of the tested water bodies, and the total P values did not correspond to the good and very good ecological status classes in 20% of the tested water bodies.

2.
Lakes and pond basins generate the largest amounts of pollution from agricultural sources: arable land-total nitrogen 1554.13 t/year and 1.94 t/year phosphorus and meadow pastures-total nitrogen 9.50 t/year and 0.20 t/year phosphorus. 3.
The highest total nitrogen load in lake basins per year on average is from agricultural pollution from arable land (98.85%), and the highest total phosphorus load is also from agricultural pollution from arable land (60%).

4.
Multiple regression analysis of the influence of anthropogenic loads in basins on the total phosphorus concentration in the water showed that the higher the total P concentration in municipal and surface wastewater was, the higher the total P value in the water, and the larger the area of organic farms in lake feeding basins was, the lower the total P value in the water (p < 0.05). The higher the concentrations of total N from arable land not connected to the sewerage network, households and municipal wastewater and the larger the area of the water body basin and area of agricultural land were, the higher the value of the total N in the water (p < 0.05). The higher the total N concentration from meadow pastures and the larger the area of organic farms in the lakes feeding the basins were, the lower the total N concentration in the water (p < 0.05). The higher the relative number of animals in the basin and the larger the area of agricultural land in the basin were, the higher the chlorophyll "a" LFI value (p < 0.05). The lower the macrophyte taxonomic composition and abundance MEI values and the lower the taxonomic composition and abundance values of zoobenthos in the water were, the poorer the water status (p < 0.05). The higher the percentage of organic farms in the basin; the higher the taxonomic composition, abundance and age of the ichthyofauna in the water; and the higher the taxonomic composition and abundance values of the zoobenthos in the water were, the better the water status (p < 0.05).