Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Environmental science

Nutrients in synergy

A literature meta-analysis of the effects of nitrogen and phosphorus on plant growth prompts a thought-provoking inference — that the supply of, and demand for, these nutrients are usually in close balance.

The nutritional needs of plants start with carbon, hydrogen and oxygen, which they obtain from atmospheric gases and from water. They also require nitrogen (a constituent of all proteins) and phosphorus (not least as a component of nucleotides, including those in DNA and RNA). Writing in Ecology Letters, Elser and colleagues1 provide a timely contribution to our understanding of how nitrogen and phosphorus, both individually and in combination, affect primary producers such as crops, trees and algae.

This is a topic with a long history. In the nineteenth century, Justus von Liebig proposed his 'law of the minimum', which describes crop production as being limited by the nutrient in shortest supply. According to this law, once the nitrogen (N) need of a crop is met by fertilization, another element such as phosphorus (P) becomes limiting (Fig. 1a), and there is no further response to additional N.

Figure 1: Nutrients and primary production.

a, The conventional view of Liebig's law of the minimum. Here, after the need for nitrogen (N) is met, N is not limiting, and increased production successively depends on satisfying limitation by phosphorus (P) and then any other limiting factors. b, A modification of the law of the minimum, based on one interpretation of the synthesis by Elser et al.1. From their meta-analysis, they infer that supply and demand of N and P are usually in close balance, so that an incremental addition of one causes a modest limitation by the other. Alleviation of each incremental limitation in turn produces a synergistic effect when both nutrients are added together.

Since Liebig's day, the use of N and P fertilizers has revolutionized agriculture. The consequences have been both desirable and undesirable. Synthetic fertilizers have fuelled the Green Revolution and greatly lessened world hunger and malnutrition. But they have also led to severe air and water pollution and other adverse effects on the environment and human health2,3. The widespread distribution of excess N and P within natural ecosystems has caused degradation of coastal waters globally, and an increase in hypoxic areas ('dead zones') on nearly every continent3,4. Excess N and/or P lead to algal blooms and so consumption of the oxygen required for productive fisheries and for healthy marine habitats. In the United States, two-thirds of coastal rivers and bays are degraded through the consequences of excess nutrients5. Similar problems occur in fresh waters, and excess fertilization of forests can lead to decline and loss of biodiversity. Management strategies for mitigating the two pollutants can differ2. So the long-standing debate over the relative importance of N and P as agents of excess production in different ecosystems is of immense practical significance.

This is the context in which Elser and colleagues' study1 is set. It is a meta-analysis of more than 300 publications reporting results of nutrient-amendment experiments in marine, freshwater and terrestrial ecosystems, with the effects of the two elements being assessed in terms of increased biomass or production. The study supports some well-established rules of thumb in biogeochemistry. Examples are the greater limitation on P than N in mature forests growing on highly weathered lowland soils in the tropics; the greater responses to P than N addition in freshwater ecosystems; and the greater responses to N addition in marine ecosystems.

More importantly, the analysis demonstrates a surprisingly consistent pattern of a synergistic effect of N and P addition on net primary productivity across all ecosystem types. Adding N and P together seems to give photosynthesis by algae and higher plants more of a boost than adding either one separately. The authors infer from this that the stoichiometry of N and P supply and demand must generally be in close balance in most ecosystems. According to this interpretation, P is rarely available in great excess relative to N, so a modest addition of N quickly provokes a limitation on P. When N and P are added together, N and P limitation may alternate in numerous small incremental steps, ultimately producing a synergistic effect (Fig. 1b).

This is an impressive synthesis1, but several caveats are in order. First, we lack a mechanistic understanding of how the availability of one resource affects the supply of and demand for another resource. At the cellular level, regulation of the relative dynamics of the demand for N for the synthesis of enzymes, and the demand for P for the synthesis of nucleic acids and also of ATP, is poorly known. At the organismal level, some species have adaptations for obtaining a nutrient that would otherwise be difficult to acquire (for example, diverting carbon and nitrogen to the fungal mycorrhizae that form symbioses with plant roots to improve access to P in nutrient-poor soils). But our knowledge of such trade-offs is only qualitative. And at the ecosystem level, the factors that may influence the relative importance of N fixation from the atmosphere, such as the availability of N, P and molybdenum (an essential component of N-fixing enzymes), and competition for light and water, are not well understood6.

Second, the doses of N and P addition — either through natural pulses or in bioassay experiments — may be very important, but were not specifically addressed by Elser and colleagues. If the doses are large enough, adding both nutrients may simply alleviate first one nutrient limitation and then the other, as per Liebig's classic law, and this could look like a synergistic effect. The assertion that the stoichiometric supply of N and P in natural systems is generally close to balanced could be further tested by quantitative dose–response experiments.

Third, there is the issue of the timescale of nutrient-amendment experiments. Experiments that are short in duration relative to the life cycles of the organisms being studied measure only the response of the organisms that are dominant in the ecosystem at the time of the assay5,7. Longer-term ecosystem-scale responses to nutrients can be different, as the dominant species change with changes in biogeochemical processes. A short-term assay in a freshwater lake, for example, would probably indicate P limitation, followed by an apparent synergistic effect with N limitation once enough P were added. But in one experiment8, years of P fertilization led to a predominance of N-fixing cyanobacteria that produced enough reactive nitrogen to keep the lake P limited.

Finally, resource limitation may simultaneously involve several nutritional elements, along with light, water and carbon dioxide. If responses to the addition of N and P are synergistic, we would expect complex synergies among other potential limiting resources as well.

So we are left with plenty of questions. Nonetheless, Elser and colleagues' meta-analysis1 provides the most thorough examination of short- and medium-term nutrient-amendment experiments to date. Their synthesis adds to a growing body of evidence that addressing the 'off-farm' environmental consequences of food production will require efforts to reduce losses of both N and P from agricultural systems2,5,7. It also provides an improved basis for formulating testable hypotheses to describe nutrient interactions in other ecosystems.


  1. 1

    Elser, J. J. et al. Ecol. Lett. doi: 10.1111/j.1461-0248.2007.01113.x (2007). | Article |

  2. 2

    Howarth, R. W. et al. in Millennium Ecosystem Assessment. Ecosystems and Human Well-being Vol. 3: Policy Responses 295–311 (Island, Washington DC, 2005).

    Google Scholar 

  3. 3

    UNEP/WHRC. Reactive Nitrogen in the Environment: Too Much or Too Little of a Good Thing (United Nations Environment Programme, Paris, 2007).

  4. 4

    Diaz, R. J., Nestlerode, J. & Diaz, M. L. Proc. 7th Int. Symp. Fish Physiology, Toxicology, and Water Quality EPA-600-R-04-049, 1–33 (US Environmental Protection Agency, 2003).

  5. 5

    National Research Council. Clean Coastal Waters: Understanding and Reducing the Effects of Nutrient Pollution (National Academies Press, Washington DC, 2000).

  6. 6

    Vitousek, P. M. et al. Biogeochemistry 57/58, 1–45 (2002).

    CAS  Article  Google Scholar 

  7. 7

    Howarth, R. W. & Marino, R. Limnol. Oceanogr. 51, 364–376 (2006).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Schindler, D. W. Science 195, 260–262 (1977).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Davidson, E., Howarth, R. Nutrients in synergy. Nature 449, 1000–1001 (2007).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing