Nitrogen (N) and phosphorus (P) are both essential nutrients, indispensable for living species to survive and grow. They are naturally present in the environment and are generally available in small amounts in undisturbed aquatic systems.
However, over the past centuries, humans have increasingly captured N from the atmosphere (through the industrial Haber-Bosch process) and mined P from localized rock formations, mostly for the fertilization of expanding croplands to feed a booming global population. Large amounts of these nutrients end up in surface freshwaters, notably due to leaching or erosion in agricultural lands, and within sewage water effluents from households. The resulting nutrient over-enrichment of these systems can lead to a number of ecological disasters, such as harmful algae proliferation, or oxygen depletion in the water column, potentially causing massive fish kills. These can have important environmental and economic repercussions.
Mitigating such risks has been a major focus of environmental research and policy over the past decades. It requires a thorough understanding of how human activities, climate, and ecosystems interact. Our understanding of these interplays is translated into models to predict the effect of future environmental strategies and assist in policy design. Using such models in a recent study(1), researchers assessed the N and P delivery to global surface freshwaters from both natural and human sources. They showed that N and P inputs have almost doubled over the 20th century, reaching about half of the global production through the Haber-Bosch process and of the global P extraction from mines by the year 2000, respectively.
However, ecological risks do not only depend on the amount of nutrients reaching aquatic systems, but also on their form and the coincidence of their emission with favorable climatic conditions. Dissolved inorganic nutrient forms (noted DIN and DIP for N and P, respectively) are the most readily usable for algae growth. Nutrient forms depend on their origin. In the present study, we define, based on an extensive review of past studies, the forms of N and P from major nutrient sources (atmospheric deposition, aquaculture, sewage water, runoff from agricultural lands, soil loss, groundwater, and detrital vegetation in floodplains). Combining it with results from the aforementioned study and outputs from a water flow model, we assessed their monthly delivery to global surface freshwaters over the 20th century.
In 1900, N and P loads to global rivers mainly originated from natural sources (74% for N and 62% of P inputs), such as decaying vegetation in floodplains or groundwater. They were highest in tropical areas, such as the Amazon and Congo river basins, and large proportions were in the organic form (70% for N, and 44% for P). Natural sources have remained dominant in the Amazon and tropical Africa throughout the 20th century. However, their contribution has dropped to around one-third of the total N and P delivery in the year 2000.
Currently, agricultural activities dominate N and P emissions over all continents, especially areas of intensive agriculture (USA, Europe, Australia, China, India, Southern Africa, and South of Brazil). For N, this leads to increased leaching of fertilizer-DIN. The largest proportion of this agriculture-derived DIN travels through soils and groundwaters before reaching surface freshwaters. Soil loss from agricultural lands is the principal global P source, leading to high particulate-bound inorganic P loads. Its contribution to total P delivery to global surface freshwaters has risen from 40 to 50% between 1900 and 2000. Moreover, due to the growing global population and urbanization, the contribution of DIN and DIP-rich sewage water emissions to nutrient inputs into global rivers has increased from 4-5% to 12%.
Global changes in human activities have not only intensified nutrient inputs to surface freshwaters and modified their form but have also altered the spatial distribution and seasonal variability of nutrient delivery. For example, in areas of high agricultural expansion, such as Brazil and Australia, the proportion of DIN originating from runoff on fertilized lands, a highly seasonally-variable source, has risen. At northern latitudes, DIP delivery to rivers has become less variable over the seasons, due to the increased contribution of sewage inputs from households. These inputs depend mostly on daily consumption and are therefore more stable over the year.
Over the 20th century, global population has risen more than 3-fold, and concentrated in urban areas. Sewage emissions now constitute the dominant nutrient source locally, in many densely populated areas over all continents (e.g., along the US coasts, in Europe, the Middle East, or Southeastern China), exerting a quasi-constant pressure on local freshwater ecosystems. Even in places where sewage does not constitute the major nutrient source on a yearly average, it can become predominant during periods of low runoff, when other sources are lower. For example, in 2000, sewage water constituted the main source of N and P for at least one month over one-third of Western Europe and one-quarter of Central Europe.
In general, over the 20th century, the emergence of significant human sources to surface freshwaters, with an increasing influence of agriculture and sewage water effluents over that of natural sources has led to an increase in the proportion of inorganic N and P forms, more likely to promote excessive algae growth and subsequent disastrous consequences. The DIN leaching through agricultural soils and groundwaters may travel for years or decades before reaching surface freshwaters, temporarily masking the effect of surface water quality protection measures. Similarly, the increasing particle-bound inorganic P eroded from fertilized fields may accumulate by settling in bed sediments, most likely in slow flowing areas, such as lakes or impoundments. This accumulated P may, later on, be released, thereby constituting a long-term source to surface freshwaters.
This work is important for future large-scale studies on nutrient transfers in surface freshwaters. We show that N and P loading estimates could be further refined by improving our current characterization of groundwaters, and refining the spatial information on cropland fertilization methods. Future research on in-stream processes at the global scale, including nutrient accumulation, release from sediments, and uptake for algae growth, is needed. This will allow for the identification of spatial and temporal hotspots of ecological risks, with coinciding periods of high nutrient loading, and adequate climate conditions for biological activity that can drive short-term water quality deterioration events. In the context of a growing population and increasing food and water needs, such research is crucial to identify effective large-scale strategies to sustainably coordinate nutrient and water use without compromising the health of ecosystems.
These findings are described in the article entitled Forms and subannual variability of nitrogen and phosphorus loading to global river networks over the 20th century, recently published in the journal Global and Planetary Change (volume 163, 2018). This work was conducted by Lauriane Vilmin, José M. Mogollón, Arthur H. W. Beusen, and Alexander F. Bouwman from Utrecht University.
- Beusen, A.H.W, Bouwman, A.F., Van Beek, L.P.H., Mogollón, J.M., and Middelburg, J.J (2016). Global riverine N and P transport to ocean increased during the 20thcentury despite increased retention along the aquatic continuum. Biogeosciences, 13, p. 2441-2451.