Modeling phosphorus in rivers at the global scale: recent successes, remaining challenges, and near-term opportunities

https://doi.org/10.1016/j.cosust.2018.10.010Get rights and content

Highlights

  • Multiple global phosphorus (P) models have generated useful insight into P loading hotspots, sources, and drivers.

  • Current models lack some important P processes, providing an opportunity for model enhancement.

  • However, there is still much to learn by applying existing models in new ways.

Understanding and mitigating the effects of phosphorus (P) overenrichment of waters globally, including the evaluation of the global Sustainability Development Goals, requires the use of global models. Such models quantitatively link land use, global population growth and climate to aquatic nutrient loading and biogeochemical cycling. Here we describe, compare, and contrast the existing global models capable of predicting P transport by rivers at a global scale. We highlight important insights gained from the development and application of these models, and identify important near-term opportunities for model improvements as well as additional insight to be gained through new model analysis.

Introduction

Phosphorus (P) is an essential, often limiting, macronutrient in freshwater systems [1] that, under certain conditions, can also limit primary production in terrestrial and coastal ecosystems [2, 3, 4]. By mining phosphorus and using it primarily as a fertilizer, humans have more than doubled the rate at which bioavailable P is supplied to the terrestrial biosphere [5, 6, 7, 8]. Widespread mobilization of geologic P in agriculture has been necessary to feed a large and burgeoning global population. However, P runoff and leaching from agricultural fields and animal production facilities [9], and flowing through inadequately treated sewage systems [8], has incurred substantial environmental costs. These include increased frequency and severity of hypoxic events, harmful algal blooms, changes in primary productivity and ecosystem function, often leading to decreased biodiversity, impaired water quality, and increased greenhouse gas emissions [10, 11, 12, 13]. Understanding and mitigating the effects of P overenrichment of waters globally, including the evaluation of the global Sustainability Development Goals (SDGs), requires the use of global models that quantitatively link land use, global population growth and climate to aquatic nutrient loading and biogeochemical cycling. Such global P models are useful instruments to evaluate global hotspots and future trends of aquatic P loading under global climate and socioeconomic changes, and can therefore help to provide insight in where and to what extent better management and mitigation measures are needed. A variety of models have been developed for use at comparatively small (0.01–1000 km2) scales in regions where model input and evaluation data are available, and these are reviewed elsewhere [14, 15, 16]. Here we focus exclusively on global P loading and transport models, which have emerged just within the past two decades, following early pioneering work [17, 18, 19]. Such global models are also useful at local and regional scales for several reasons. They provide context for local observations and modeling efforts, helping people and governments understand where regions fall in terms of the severity of their present-day and potential future P loading problems. Global P models can also provide reasonable estimates of P loading and sources in data-poor regions where applying complex, locally or regionally calibrated models is not possible. Also, when evaluated at the local-regional scale, Global P models provide a mechanism to test, evaluate, and, ultimately, improve understanding of how P sources and transformation control P delivery to and through rivers and watersheds. Finally, global models, when applied in a regional context, can facilitate transboundary analyses that are often not possible using local water quality models.

Below, we: first, briefly describe, compare, and contrast three ‘families’ of published global total phosphorus (TP) transport models, second, highlight major insights attributable to the development and application of these models, third, discuss important areas for future model enhancement, and fourth, identify important research avenues deserving near-term attention that can be addressed with either existing or somewhat enhanced global P transport models.

Section snippets

Global P models: key characteristics, similarities and differences

There are currently three peer-reviewed ‘families’ of global models capable of predicting river total P (TP) loading and transport: IMAGE-GNM, WaterGAP, and Global NEWS. The specific version of each model reviewed here is the most recent one capable of predicting river TP loading and transport: IMAGE-GNM-TP, WaterGAP3.2, and Global NEWS-2-TP for each modeling family, respectively. NEWS-DIP-HD, a high-resolution (half-degree) version of NEWS-2-DIP [20] is also discussed. Each of these models is

Understanding where river P fluxes are high (and rapidly changing)

Global NEWS-2-TP and IMAGE-GNM-TP models provided the first global, spatially explicit estimates of global P delivery to coastal zones [21,29, 30, 31,32••]; Global NEWS-DIP-HD and IMAGE-GNM-TP provided the first-ever estimates of DIP and TP delivery to (and through) the surface freshwater component of watersheds [20,32••]. WaterGAP3.2 and IMAGE-GNM provided the first-ever estimate of TP loading to the world’s largest lakes [22••]. Understanding spatial patterns of P loading to freshwaters and

Challenges & opportunities

Despite progress over the past two decades in modeling P transport to and through aquatic systems, a number of challenges and opportunities remain. One puzzling outcome (and research opportunity) resulting from the comparison of several P models is that there is substantial uncertainty regarding even the total global rate of P delivery to the coastal zones. Global NEWS-2-TP and IMAGE-GNM-TP produce estimates of global P export that differ by more than a factor of two (9 Tg P year−1 for Global

Conclusion

Within the past two decades, global P loading and transport models have granted useful insight into: P loading hotspots, P sources, and drivers of P delivery to surface waters historically, currently, and under future scenarios. However, there is still much room for improvement in model performance and representativeness, and key questions remain to be answered. Opportunities to achieve model improvements include: first, increased temporal resolution, including representation of ‘event’ fluxes,

References and recommended reading

Papers of particular interest, published within the period of review, have been highlighted as:

  • • of special interest

  • •• of outstanding interest

Acknowledgements

This paper evolved from a workshop titled ‘Water Quality: a new challenge for global scale modelling’ held at Wageningen University 18–21 September 2017. Funding for this workshop came from the OECD-CRP, the support of which is gratefully acknowledged. Harrison received support from an NSF INFEWS grant (NSF EAR1639458). A.F.B. and A.H.W.B. received support from PBL Netherlands Environmental Assessment Agency through in-kind contributions to The New Delta 2014 ALW projects no. 869.15.015 and

References (65)

  • P.N. Froelich

    Kinetic control of dissolved phosphate in natural rivers and estuaries - a primer on the phosphate buffer mechanism

    Limnol Oceanogr

    (1988)
  • H.P. Jarvie et al.

    Water quality remediation faces unprecedented challenges from “legacy phosphorus”

    Environ Sci Technol

    (2013)
  • S.Z. Sattari et al.

    Residual soil phosphorus as the missing piece in the global phosphorus crisis puzzle

    Proc Natl Acad Sci U S A

    (2012)
  • I.M. Gren et al.

    Credit stacking in nutrient trading markets for the Baltic Sea

    Mar Policy

    (2017)
  • J.J. Elser et al.

    Global analysis of nitrogen and phosphorus limitation of primary producers in freshwater, marine and terrestrial ecosystems

    Ecol Lett

    (2007)
  • P.M. Vitousek et al.

    Terrestrial phosphorus limitation: mechanisms, implications, and nitrogen-phosphorus interactions

    Ecol Appl

    (2010)
  • J.B. Sylvan et al.

    Phosphorus limits phytoplankton growth on the Louisiana shelf during the period of hypoxia formation

    Environ Sci Technol

    (2006)
  • L.E. Street et al.

    Phosphorus availability determines the response of tundra ecosystem carbon stocks to nitrogen enrichment

    Ecosystems

    (2018)
  • E.M. Bennett et al.

    Human impact on erodable phosphorus and eutrophication: a global perspective

    Bioscience

    (2001)
  • S.R. Carpenter et al.

    Reconsideration of the planetary boundary for phosphorus

    Environ Res Lett

    (2011)
  • G. Van Drecht et al.

    Global nitrogen and phosphate in urban wastewater for the period 1970 to 2050

    Glob Biogeochem Cycles

    (2009)
  • A.F. Bouwman et al.

    Human alteration of the global nitrogen and phosphorus soil balances for the period 1970-2050

    Glob Biogeochem Cycles

    (2009)
  • B.R. Deemer et al.

    Greenhouse gas emissions from reservoir water surfaces: a new global synthesis

    BioScience

    (2016)
  • D.W. Schindler

    Eutrophication and recovery in experimental lakes implications for lake management

    Science (Washington D C)

    (1974)
  • G.K. Macdonald et al.

    Guiding phosphorus stewardship for multiple ecosystem services

    Ecosyst Health Sustain

    (2016)
  • D.K. Borah et al.

    Watershed-scale hydrologic and nonpoint-source pollution models: review of mathematical bases

    Trans ASAE

    (2003)
  • D.K. Borah et al.

    Watershed scale hydrologic and nonpoint source pollution models for long-term continuous and storm event simulations

    Total Maximum Daily Load (Tmdl): Environmental Regulations II, Proceedings

    (2003)
  • Programme U.N.E

    A Snapshot of the World’s Water Quality: Towards a global Assessment

    (2016)
  • S.V. Smith et al.

    Humans, hydrology, and the distribution of inorganic nutrient loading to the ocean

    Bioscience

    (2003)
  • N.F. Caraco

    Influence of human populations on P transfers to aquatic systems: a regional scale study using large rivers

  • J.D. Newbold et al.

    Measuring nutrient spiralling in streams

    Can J Fish Aquat Sci

    (1981)
  • J.A. Harrison et al.

    Magnitudes and sources of dissolved inorganic phosphorus inputs to surface fresh waters and the coastal zone: a new global model

    Glob Biogeochem Cycles

    (2010)
  • Cited by (18)

    • Advancing understanding of in-river phosphorus dynamics using an advection–dispersion model (ADModel-P)

      2022, Journal of Hydrology
      Citation Excerpt :

      The mathematical modelling of water quality facilitates both understanding of phenomena and its prediction, even in cases where the availability of experimental data is limited (Ani et al., 2009). Such modelling can be carried out at different levels of magnitude and detail, starting from the small scale (hundreds of meters of river stretch) to the global scale (see the models discussed by Harrison et al., 2019b), depending on the area of interest, its particularities and the needs. With respect to the resolution in terms of space and time, the spectrum of approaches extends between: (A) catchment scale modelling (describing a catchment or multiple catchments, not necessarily high-resolution in terms of space and time), mainly based on usual long-term monitoring data (Keraga et al., 2019; Srinivas and Singh, 2018); and (B) modelling of river stretches based on high resolution data (multiple monitoring points along the stretch, where measurements of channel characteristics, hydrodynamic data and water quality data are needed), such as ADModel for nitrogen (Ani et al., 2011).

    • Recent advances in the determination of phosphate in environmental water samples: Insights from practical perspectives

      2020, TrAC - Trends in Analytical Chemistry
      Citation Excerpt :

      With rapid industrial development and rising global populations, humans have exerted a profound influence on the global P cycle, characterized by greatly enhanced P mobility [3,4]. Large amounts of P are released during anthropogenic processes such as phosphate rock mining, fertilizer application, livestock and food production, human consumption and the generation of excreta [4–8]. As global P consumption is projected to grow continuously in the 21st century [9], a recent study has estimated that the rate of P loss from cropland to freshwater (10.4 ± 5.7 Tg P yr−1) outpaced loss from natural erosion (7.6 ± 3.3 Tg P yr−1) [4].

    View all citing articles on Scopus

    The opinions expressed and arguments employed in this publication are the sole responsibility of the authors and do not necessarily reflect those of the OECD or of the governments of its Member countries.

    View full text