Biological productivity in terrestrial ecosystems, including forests and agricultural lands, is supported in part by nitrogen (N), an essential nutrient (e.g., Vitousek et al. 1997). In addition, N runoff from various terrestrial land uses reaches lakes and coastal areas through rivers and becomes an important nutrient for phytoplankton and other aquatic organisms (Howarth et al. 1988). Coastal ecosystems are rich in ecosystem services and are important habitats for a wide variety of phytoplankton, fish and other aquatic biota (Beck et al. 2001). When N export from land to water increases above a certain level, excess nutrients yield negative consequences for aquatic primary producers in coastal and estuary ecosystems (Paterson and Whitfield 2000). Excessive application of chemical fertilizers and manure to agricultural lands can lead to high N runoff into water bodies (Howarth et al. 1996; Boyer et al. 2002), leading to a variety of adverse effects, including water quality degradation in downstream lakes and coastal areas, harmful algal blooms, low dissolved oxygen levels in water, and a loss of aquatic biodiversity (e.g., Carpenter et al. 1998). The relationship between terrestrial N inputs and N concentrations in river water depends on the balance between how much N is retained and how much N is exported to the hydrosphere within each land use and land cover type in the catchment (Bellmore et al. 2018).
Many studies have been conducted on the relationship between land use and river water quality. In most cool-temperate forest ecosystems, N is a limiting factor for primary productivity, resulting in externally derived N (i.e., atmospheric N deposition and biological N fixation) being retained in forests through absorption by plants and soil microorganisms (e.g., Shibata and Fukuzawa 2010). In a clear-cutting experiment in a forested catchment in the northeastern United States of America, the concentration of nitrate N (NO3−) in river water increased after logging (Likens et al. 1970), indicating that a large amount of N is retained under natural conditions through biogeochemical cycling within the forest ecosystem. Okazawa et al. (2009) found that NO3− concentrations in river water are lower in densely forested areas, indicating that forest ecosystems play a major role in maintaining low N concentrations in river water. Shibata et al. (2021) also showed that the presence of forest ecosystems in the upper catchment lowers the NO3− concentration in river water in the lower reaches. Riparian forests and wetlands are also important for N cycling and river water quality. Interface zones between land and water, such as riparian wetlands, are rich in soil moisture and possess anaerobic conditions, promoting denitrification by soil bacteria and archaea (Hill 1996). Since denitrification is the process of removing NO3− from soil and water, denitrification processes in riparian zones have the ability to reduce N pollution from terrestrial areas to rivers (Hill 1996; Yoh 2014). However, certain amounts of chemical fertilizers and compost are commonly applied to agricultural lands to grow crops. The nitrogen that is not absorbed by the crops and microorganisms in soil can be leached into the river. Therefore, it is known that the proportion of agricultural land in a catchment is positively correlated with the N concentration of river water downstream (e.g., Woli et al. 2002). In the coastal watershed of Chesapeake Bay in the United States of America, a strong positive correlation was shown between the percentage of agricultural land in the catchment and NO3− and total N concentrations in water discharged from the catchment to the river, with approximately 80% of the total N in water being NO3− (Jordan et al. 1997). Nitrogen concentrations leaving the catchment via river water are intricately controlled by many complex biotic and abiotic factors, including ecosystem N requirements and fertilizer N inputs in agricultural lands, and vary with land use distribution, soil type, topography, N uptake by plants and microbes, other external N inputs on land systems, and hydrological and biogeochemical N processes in the ground (e.g., Jiang et al. 2015).
Similar to river water quality, the proportion of agricultural land in a catchment has been shown to have a significant impact on N and phosphorus (P) concentrations in downstream lakes and coastal areas (Camargo et al. 2006; Paerl et al. 2011; Jordan et al. 2018; Chambers et al. 2012). However, few studies have been conducted to show how much N fertilizer input is ultimately allowed in terrestrial areas and how this affects water quality based on field observations of terrestrial and aquatic areas. As the impact of recent anthropogenic activities on ecosystems has become an issue, it is necessary to understand the changes in N dynamics patterns and processes in response to anthropogenic impacts to optimize N cycling at the catchment scale (e.g., Kimura et al. 2006) and to clarify the relationship between catchment land composition, including human activities, and the resulting river water quality (Hayakawa 2012). To test the impact of land use patterns on river water quality, it is useful to conduct research in catchments where multiple land uses are involved rather than in watersheds dominated by a single land use type.
The Bekanbeushi River and Lake Akkeshi catchment in eastern Hokkaido of northern Japan are the targets of this study; various studies have been conducted in them on the relationship between forests and agricultural lands and river water quality and on the response of primary producers in aquatic ecosystems to the nutrient environment in the lake (e.g., Woli et al. 2004; Hayakawa et al. 2006; Akabane et al. 2003; Kajiwara 2008; Isada et al. 2021). However, there have not been enough studies evaluating N dynamics in the entire catchment, taking into account the ecosystems of downstream lakes. Therefore, this study aimed to clarify the relationship between the N inflow from terrestrial to aquatic ecosystems of the entire catchment and land use distribution and the impact of terrestrial riverine N on the downstream brackish-estuary lake.
Therefore, in this study, we conducted a comparative analysis of nutrient concentrations in river water and land use distribution based on field observations in the Bekanbeushi River and Lake Akkeshi catchment. To consider the dynamics and effects of N as a nutrient, we also analyzed and compared it to P, which is similarly important as a nutrient.