FOCUS | progress article

Published online: 18 aPril2010|doi: 10.1038/ngeo838

the impact of agricultural soil erosion on

biogeochemical cycling
John n. Quinton1*, gerard govers2, Kristof Van oost3 and richard d. bardgett1
Soils are the main terrestrial reservoir of nutrients, such as nitrogen and phosphorus, and of organic carbon. Synthesizing
earlier studies, we find that the mobilization and deposition of agricultural soils can significantly alter nutrient and carbon
cycling. Specifically, erosion can result in lateral fluxes of nitrogen and phosphorus that are similar in magnitude to those induced
by fertilizer application and crop removal. Furthermore, the translocation and burial of soil reduces decomposition of soil organic
carbon, and could lead to long-term carbon storage. The cycling of carbon, nitrogen and phosphorus are strongly interrelated. For
example, erosion-induced burial of soils stabilizes soil nutrient and carbon pools, thereby increasing primary productivity and
carbon uptake, and potentially reducing erosion. Our analysis shows soils as dynamic systems in time and space.

he potential impact of soil processes on the biogeochemical

cycling of carbon, and nutrients such as nitrogen and phosphorus, remains one of the great uncertainties in our knowledge of global climate change1,2. Beginning with the pioneering work
of Stallard3, scientists have become increasingly aware that lateral
fluxes induced by soil erosion are of key importance in the global
carbon cycle. So far, work on the influence of erosion on elemental
cycles has focused on nutrient losses and effects on primary productivity; the influence on global nutrient and carbon cycles has yet to
be studied in detail. In this Progress Article we estimate the impact
of soil erosion on the carbon, nitrogen and phosphorus cycles using
global datasets, and identify interactions between erosion and elemental cycles.

nutrient mobilization

Several attempts have been made to estimate soil and carbon erosion rates associated with agriculture in recent years. From a critical
analysis of these estimates, we calculate that sediment flux due to
water erosion is about 28 Pg yr1, and that a further ~5 Pg yr1 and
~2 Pg yr1 of sediment are mobilized by tillage and wind erosion,
respectively, leading to a total sediment flux of about 3510 Pg yr1
(see Supplementary Information S1). This corresponds to an agricultural carbon erosion flux of 0.50.15 Pg C yr1. Furthermore, we
estimate that 0.080.02 Pg C is delivered to river systems by water
erosion each year.
To estimate the flux of nitrogen associated with erosion, we
combine spatial estimates of soil erosion with global soil nitrogen
data4 (see Supplementary Information S1). We estimate that around
2342 Tg of nitrogen is moved by erosion each year. Lateral fluxes
of nitrogen due to erosion are on the same order of magnitude as the
112 Tg of nitrogen applied to agricultural land in the form of chemical fertilizers each year 5, the 75 Tg of nitrogen removed in harvested
crops each year 6, and the estimated riverine fluxes of particulate
nitrogen, which range between 23 and 30 Tg N yr1 (refs 7,8).
We estimate that soil erosion is responsible for a flux of 2.13.9 Tg
of organic phosphorus per year, and 12.522.5 Tg of inorganic phosphorus per year (see Supplementary Information S1). However,
owing to the limited availability of global soil phosphorus data,
these estimates are uncertain. Global mean phosphorus fluxes are
considerably lower than the 40 Pg of phosphorus stored in soils globally 9, but are similar in magnitude to crop uptake6 (14 Tg yr1) and
fertilizer phosphorus additions to agricultural land (~18 Tg yr1).

However, in some parts of the world, erosion-induced fluxes of

phosphorus exceed phosphorus additions (Fig. 1), placing soil fertility and food production under increased strain.

consequences for the carbon cycle

Soil erosion encompasses soil mobilization, transport and deposition. Understanding erosional effects on the carbon cycle requires
consideration of all three phases. When soil material is mobilized,
soil structure is at least partially disrupted. Laboratory experiments
indicate that sediment mobilization could result in a significant
increase in the rate of soil organic carbon (SOC) mineralization,
during, or shortly after, mobilization; this could lead to the loss of
over 20% of the total SOC as carbon dioxide10.
When considering the potential effect of transport on SOC
mineralization, a distinction should be made between SOC deposited in a local depositional store after being transported over a
relatively small distance (<500 m) by water or tillage over a short
time (<1 day), and the fate of SOC that is delivered to rivers. Field
observations indicate that the additional SOC mineralization that
occurs during transport of soil over land is relatively unimportant:
erosiondeposition simulations based on 137Cs inventories show
that the carbon inventory found at depositional sites is inconsistent
with significant mineralization during the transport phase11. Also,
recent observations under field conditions suggest that SOC losses
from soil that is re-deposited after a short transport phase are relatively low (<2.5% of eroded SOC), and therefore not very significant
for the global carbon budget 12. On the other hand, a large amount of
SOC that is delivered to rivers may be mineralized within the river
system13.
Understanding the impact of erosion on the carbon cycle also
requires consideration of the longer-term effects. Recent work suggests that erosion can increase both the emission and sequestration
of carbon. The disruption of soil structure during erosion may lead to
the immediate release of carbon dioxide. Enhanced emissions over
longer time frames are associated with a reduction in the capacity
of eroded soils to support plant growth14, resulting in lower carbon
inputs through plant and root matter 15. Erosion could also result in
carbon sequestration16. Erosion leads to the mixing of carbon-poor
subsoil into the plough layer. If the newly exposed mineral surfaces
bind organic matter, soil carbon inventories may increase.
The promotion of carbon sequestration by erosion relies on
reduced rates of SOC decomposition, owing to the burial of

Lancaster Environment Centre, Lancaster University, Lancaster LA1 4YQ, UK, 2 KU Leuven, Department of Earth and Environmental Sciences,
Celestijnenlaan 200E, 3001 Leuven, Belgium, 3Georges Lematre Centre for Earth and Climate Research (TECLIM), Universit catholique de Louvain,
Place Louis Pasteur 3, 1348 Louvain-la-Neuve, Belgium. *e-mail: J.Quinton@Lancaster.ac.uk
1

nature geoscience | VOL 3 | MAY 2010 | www.nature.com/naturegeoscience

2010 Macmillan Publishers Limited. All rights reserved

311

progress article
article | FOCUS
progress

NaTure geOScieNce doi: 10.1038/ngeo838

Erosion/fertilization

Erosion/fertilization

2.5

Sediment flux
(Mg ha1 yr1)

16

Phosphorus
(Tg yr1)

10

Nitrogen
(Tg yr1)
0

30
112

75

18

Fertilizer input
17.5

Erosion

14
Crop uptake

Figure 1 | global fluxes of sediment, nitrogen and phosphorus. a, Shaded areas show the global distribution of sediment fluxes derived using methods
described in Supplementary Information S2. Bars show the continental fluxes of nitrogen and phosphorus by water and tillage erosion compared with
fertilizer use6. b, Global fluxes of nitrogen and phosphorus (Tg yr1) due to fertilizer input, erosion and crop uptake.

sediment in depositional environments. Although the mechanisms

that contribute to the reduction in decomposition at depth17 have
only recently received attention18, the burial of pedogenic carbon at
sites of deposition has repeatedly been shown to stabilize soil carbon over timescales of several decades, leading to reduced emissions
of carbon dioxide11. Furthermore, mineralization can be actively
suppressed in depositional environments where net primary production is greater than that in the source areas. For example, the
influx of low-carbon sediments into wetlands and lowland valley
bottoms may stimulate net carbon sequestration by diluting the
concentration of soil carbon3. Overall, the extent to which mobilization and deposition lead to carbon storage is critically dependent on
how much of the depositional accumulation is replaced by newly
produced plant-derived soil carbon at eroding sites16 dynamic
replacement of soil (Fig. 2).
The removal of soil by erosion also brings the subsoil and parent material closer to the surface. There is increasing empirical and
theoretical evidence that erosion is associated with increased rates
of chemical weathering of silicate-rich parent material under steadystate conditions19,20. As the weathering of silicate minerals consumes
carbon dioxide, it seems likely that there may also be a link between
erosion-induced weathering and the consumption of carbon dioxide,
312

although the flux is likely to be small. In contrast, where parent

materials are calcareous, accelerated weathering may result in carbon
dioxide release to the atmosphere. For example, in the Canadian prairies it has been estimated that 10% of the carbonates acidified during
erosion may be released as carbon dioxide, resulting in an estimated
carbon loss of 0.12 to 1.2 Mg ha1 yr1 (D. A. Lobb, D. L. Burton,
M. J. Lindstrom & D. C. Reicosky, unpublished observation).

impact on nutrient cycles

Research into the impact of erosion on nitrogen and phosphorus

cycling has hitherto focused on the assessment of nutrient mobilization and the delivery of nutrients to aquatic ecosystems. Less is
known about the influence of erosion on nitrogen and phosphorus
cycling within terrestrial environments.
Soil organic matter contains large quantities of nitrogen and
phosporus. Thus enhanced mineralization of soil carbon owing to
soil mobilization21 will lead to a relative increase in dissolved nitrogen and phosphorus. These dissolved forms will be more accessible
to biota than particulate or organic forms. On the other hand, burial
and preservation of deposited carbon will lead to the stabilization of
organic nitrogen. The stability of nitrogen in depositional environments may be high, and primarily determined by the rate of carbon
nature geoscience | VOL 3 | MAY 2010 | www.nature.com/naturegeoscience

2010 Macmillan Publishers Limited. All rights reserved

FOCUS | progress article

NaTure geOScieNce doi: 10.1038/ngeo838

implications for soil function

Erosion-induced changes in the cycling of carbon, nitrogen and

phosphorus may influence a range of soil processes31,32. For example, changes in the relative availability of carbon and nitrogen in
soil organic matter is a primary regulator of microbial nitrogen
mineralizationimmobilization dynamics, and hence plant nitrogen supply 33,34. Changes in the N:P ratio of soil organic matter are
also known to have significant consequences for decomposition,
nutrient cycling and plant production35,36. These feedbacks are

500
Carbon residence time (yr)

mineralization. This could explain why C:N ratios in topsoils are

remarkably constant within a given ecological context, and why
palaeosol investigations report similar C:N values to those seen in
present-day soils22. At eroding sites, dynamic replacement of carbon
will also lead to the stabilization of nitrogen. However, nitrogen may
also control carbon cycling: in some environments, biomass production and hence dynamic replacement may be directly limited by
nitrogen availability 23.
Data from the Chinese loess indicate that buried soil phosphorus also remains relatively stable over long (>10 kyr) periods
of time24. However, C:P ratios in soil organic matter show a larger
variation than C:N ratios. Furthermore, a significant fraction of
the phosphorus reservoir in soils is stored in inorganic form. Thus,
the erosional effects on phosphorus cycling will be less tightly coupled to carbon cycling. Given that phosphorus is strongly bound
to the mineral and organic soil fractions, we may tentatively
assume that the evolution of phosphorus inventories in soils will be
directly proportional to the amount of soil that is either mobilized
or deposited.
Indeed, erosion is an important mechanism for the decline in soil
phosphorus levels over longer time periods25. This is not only due to
the physical removal of phosphorus, and to the exposure of subsoil
with lower phosphorus contents, but also to the interaction between
erosion rates and chemical weathering (see above). As phosphorus
levels decline over time, the phosphorus in the soil profile changes
from a mix of mineral, occluded, non-occluded and organic forms
to a mix dominated by organic and occluded forms25. In depositional sites, sediment can be an important source of phosphorus. In
Hawaii, for instance, phosphorus limitation of forest growth on old
soils is partially alleviated by dust deposition26.
At sites where erosion dominates and inputs of nitrogen and
phosphorus are low, primary production declines exponentially as
erosion increases27, thereby reducing the potential for dynamic SOC
replacement by vegetative material. The reduction in primary productivity is not only due to the removal of nutrients, but also to the
degradation of soil structure and, critically, to a reduction in the
availability of water as soil thickness declines.
More subtle interactions may also take place. Evidence suggests
that water erosion causes sediment carbon, nitrogen and phosphorus concentrations to be enriched, relative to the parent soil, to
different extents during sediment mobilization and deposition28.
Consequently, it is likely that the relative abundance of carbon,
nitrogen and phosphorus in soils will change depending on the
relative selectivity of mobilization and deposition processes: enrichment associated with water and wind erosion is likely to be greater
than that for tillage erosion. Furthermore, the loss of carbon, nitrogen and phosphorus from mobilization sites might set in motion
a degenerative feedback, whereby associated declines in plant productivity further increase erosion vulnerability and hence nutrient
loss. Indeed it is well established that the resistance of soils to erosion is closely linked to the stabilizing influence of organic matter 29
and vegetation cover 30. Conversely at deposition sites, nutrient and
carbon contents may rise, leading to greater primary productivity
and a positive feedback on soil fertility, plant growth and resistance
to erosion.

Undisturbed
grassland
<2
Pasture
Cropland
under
conservation

250
14

Hl
36

125

0.1

Cropland

Ll
31.6

0.5

Hl
58

10

Ll
61

100

Erosion rate (Mg ha1 yr1)

Figure 2 | interplay between soil erosion, land use/soil management

and carbon cycling at sites of erosion. The blue shaded area reflects
possible combinations of carbon residence time (1/decomposition
rate) and erosion rates as a function of land use/management. The
numbers (g C m2 yr1) and size of the circles represents the maximum
size of the carbon sink (positive, green) or source (negative, red) (see
Supplementary Information S2). For croplands, the data represent highinput systems (HI, low sensitivity of yield decline to erosion, 4% per
0.1 m erosion) and low-input systems (LI, high sensitivity of yield decline
to erosion, 15% per 0.1 m erosion).

likely to be of greatest significance in nutrient poor environments,

such as the nutrient poor soils of Africa and Australia. In these
regions, soil erosion associated with reduced vegetation cover and
the loss of soil carbon can trigger catastrophic shifts to a severely
degraded state15. The acceleration of erosion by these mechanisms
may precipitate land-use change37, which itself changes the rate of
biogeochemical cycling, thereby influencing atmospheric composition and climate change38, and further disrupting carbon, nitrogen and phosphorus cycling.
Our analysis shows that agricultural landscapes are far from
static: the accelerated rates of erosion experienced at present
are causing major modifications to the terrestrial carbon, nitrogen and phosphorus cycles. We need to consider soils as mobile
systems to make accurate predictions about the consequences of
global change for terrestrial biogeochemical cycles and climate
feedbacks. This is of primary importance, given that the imbalance between carbon and nutrient fluxes owing to erosion threatens the sustainability of food production and human welfare in
many parts of the world.

references

1. Davidson, E. A. & Janssens, I. A. Temperature sensitivity of soil carbon

decomposition and feedbacks to climate change. Nature
440, 165173 (2006).
2. Heimann, M. & Reichstein, M. Terrestrial ecosystem carbon dynamics and
climate feedbacks. Nature 451, 289292 (2008).
3. Stallard, R. F. Terrestrial sedimentation and the carbon cycling: coupling
weathering and erosion to carbon burial. Glob. Biogeochem. Cycles
12, 231257 (1998).
4. Batjes, N. H. ISRIC-WISE Global Data Set of Derived Soil Properties on a
0.5 by 0.5 Degree Grid (Version 3.0): Report 2005/08 (ISRICWorld Soil
Information, 2005).
5. http://faostat.fao.org/
6. Smil, V. Agricultures largest harvest. BioScience 49, 299308 (1999).
7. Seitzinger, S. P., Harrison, J. A., Dumont, E., Beusen, A. H. W. &
Bouwman, A. F. Sources and delivery of carbon, nitrogen, and phosphorus
to the coastal zone: An overview of global nutrient export from watersheds
(NEWS) models and their application. Glob. Biogeochem. Cycles
19, GB4S01(2005).

nature geoscience | VOL 3 | MAY 2010 | www.nature.com/naturegeoscience

2010 Macmillan Publishers Limited. All rights reserved

313

progress article
article | FOCUS
progress
8. Beusen, A. H. W., Dekkers, A. L. M., Bouwman, A. F., Ludwig, W. &
Harrison, J. Estimation of global river transport of sediments and associated
particulate C, N, and P. Glob. Biogeochem. Cycles 19, GB4S05(2005).
9. Smil, V. in Encyclopedia of Global Environmental Change: Volume 3, Causes
and Consequences of Global Environmental Change (ed. Douglas, I.)
536542 (Wiley, 2002).
10. Lal, R. Soil erosion and the global carbon budget. Environ. Int.
29, 437450 (2003).
11. Van Oost, K. et al. The impact of agricultural soil erosion on the global carbon
cycle. Science 318, 626629 (2007).
12. Van Hemelryck, H., Fiener, P., Van Oost, K. & Govers, G. The effect of soil
redistribution on soil organic carbon: an experimental study. Biogeosciences
Discuss. 6, 50315071 (2009).
13. Cole, J. et al. Plumbing the global carbon cycle: integrating inland
waters into the terrestrial carbon budget. Ecosystems 10, 172185 (2007).
14. van de Koppel, J., Rietkerk, M. & Weissing, F. J. Catastrophic vegetation shifts
and soil degradation in terrestrial grazing systems. Trends Ecol. Evol.
12, 352356 (1997).
15. Berhe, A., Harte, J., Harden, J. & Torn, M. The significance of the
erosion-induced terrestrial carbon sink. BioScience 57, 337346 (2007).
16. Harden, J. W. et al. Dynamic replacement and loss of soil carbon on eroding
cropland. Glob. Biogeochem. Cycles 13, 885901 (1999).
17. Rosenbloom, N. A., Harden, J. W., Neff, J. C. & Schimel, D. S.
Geomorphic control of landscape carbon accumulation. J. Geophys. Res.
111, G01004 (2006).
18. Fontaine, S. et al. Stability of organic carbon in deep soil layers controlled by
fresh carbon supply. Nature 450, 277280 (2007).
19. Gabet, E. J. & Mudd, S. M. A theoretical model coupling chemical weathering
rates with denudation rates. Geology 37, 151154 (2009).
20. Riebe, C., Kirchner, J., Granger, D. & Finkel, R. Strong tectonic and weak
climatic control of long-term chemical weathering rates. Geology
29, 511514 (2001).
21. Jacinthe, P. A., Lal, R. & Kimble, J. M. Carbon dioxide evolution in runoff from
simulated rainfall on long-term no-till and plowed soils in southwestern Ohio.
Soil Till. Res. 66, 2333 (2002).
22. Inoue, Y., Baasansuren, J. Watanabe, M., Kamei, H. & Lowe, D. J. Interpretation
of pre-ad 472 Roman soils from physicochemical and mineralogical properties
of buried tephric paleosols at Somma Vesuviana ruin, southwest Italy.
Geoderma 152, 243251 (2009).
23. van Groenigen, K.-J. et al. Element interactions limit soil carbon storage.
Proc. Natl Acad. Sci. USA 103, 65716574 (2006).
24. Rao, W., Chen, J., Luo, T. & Liu, L. Phosphorus geochemistry in the Luochuan
loess section, North China and its paleoclimatic implications. Quat. Int.
144, 7283 (2006).
25. Filippelli, G. M. The global phosphorus cycle: Past, present, and future
Elements 4, 8995 (2008).

314

NaTure geOScieNce doi: 10.1038/ngeo838

26. Chadwick, O., Derry, L., Vitousek, P., Huebert, B. & Hedin, L. Changing
sources of nutrients during four million years of ecosystem development.
Nature 397, 491497 (1999).
27. Stocking, M. A. Tropical soils and food security: The next 50 years. Science
302, 13561359 (2003).
28. Sharpley, A. N. The selective erosion of plant nutrients in runoff.
Soil Sci. Soc. Am. J. 49, 15271534 (1985).
29. Tisdall, J. & Oades, J. Organic matter and water-stable aggregates in soils.
J. Soil Sci. 33, 141163 (1982).
30. Elwell, H. & Stocking, M. Vegetal cover to estimate soil erosion hazard in Rhodesia.
Geoderma 15, 6170 (1976).
31. Woodward, F. I., Bardgett, R. D., Raven, J. A. & Hetherington, A. M. Biological
approaches to global environment change mitigation and remediation.
Curr. Biol. 19, 615623 (2009).
32. Reay, D., Dentener, F., Smith, P., Grace, J. & Feely, R. Global nitrogen deposition
and carbon sinks. Nature Geosci. 1, 430437 (2008).
33. Kaye, J. P. & Hart, S. C. Competition for nitrogen between plants and soil
microorganisms. Trends Ecol. Evol. 12, 139143 (1997).
34. Bardgett, R. D. The Biology of Soil: A Community and Ecosytems Approach 1st
edn (Oxford Univ. Press, 2005).
35. Wardle, D. A., Walker, L. R. & Bardgett, R. D. Ecosystem properties
and forest decline in contrasting long-term chronosequences. Science
305, 509513 (2004).
36. Wardle, D. A., Bardgett, R. D., Walker, L. R. & Bonner, K. Among- and
within-species variation in plant litter decomposition in contrasting longterm chronosequences. Funct. Ecol. 23, 442453 (2009).
37. Bakker, M. M. et al. Soil erosion as a driver of land-use change. Agr. Ecosyst.
Environ. 105, 467481 (2005).
38. Feddema, J. J. et al. The importance of land-cover change in simulating future
climates. Science 310, 16741678 (2005).

acknowledgements

K.V.O is a research associate of the Fonds de la Recherche Scientifique (FNRS),

Belgium, and is supported by the Communaut Franaise de Belgique (convention number 09/14-022).

author contributions

J.Q. led the writing of the paper. K.V.O. conducted the model simulations and contributed to the writing, together with G.G. and R.B.

additional information

The authors declare no competing financial interests. Supplementary information

accompanies the paper on www.nature.com/naturegeoscience.

nature geoscience | VOL 3 | MAY 2010 | www.nature.com/naturegeoscience

2010 Macmillan Publishers Limited. All rights reserved