Opinion

Resources for Crop Production: Accessing

the Unavailable
John J. Ewel,1,* Laura A. Schreeg,2 and Thomas R. Sinclair3

An acute imbalance between human population and food production is pro- Highlights
jected, partially due to increasing resource scarcity; dietary shifts and the Natural ecosystems and undomesti-
cated plants have solved many
current course of technology alone will not soon solve the problem. Natural
resource-acquisition problems – pro-
ecosystems, typically characterized by high species richness and perennial blems challenging agriculture with
growth habit, have solved many of the resource–acquisition problems faced by economic and societal constraints on
fertilizer, water and fossil energy.
crops, making nature a likely source of insights for potential application in
commercial agriculture. Further research on undomesticated plants and natural Symbiotic N2 fixation input can best be
ecosystems, and the adaptations that enable them to meet their needs for N, P, enhanced by focus on host plant.

and water, could change the face of commercial food production, including on Biologically mediated extraction of
marginal lands. phosphate from soils is widespread –
geographically and phylogenetically –
in nature.
Need for Fresh Perspectives on Food Production
Human malnutrition and starvation are major challenges for humanity. Further exacerbating the Efficacy of water use can be increased
problem, the current annual rate of human population growth, 1.10% globally [1], although by temporal regulation of transpiration,
declining, continues to outpace developing country projected increases in crop yields – 0.83% by deeper-rooted crops, and by
hydraulic redistribution via roots from
for maize, 0.86% for wheat, and 0.63% for rice [2]. There is clear evidence that yield increases in wetter to drier soil where it can
these crop species are stagnating in some of the more productive regions of the world [3]. The become available to companion
situation is exacerbated by the impact of agriculture on the resources required for crop species.
production, including the three on which we focus here – N, P, and water. Furthermore, there
is real concern about the species and genetic diversity of modern cropping. The experience of
the attack by Southern Corn Leaf Blight in 1970–71 [4] in the US looms as an example of the
dangers of genetic uniformity within a crop species. Included in the concern of about lack of
diversity is the heavy reliance on annual plants without the benefit of a substantial presence of
perennial species.

Here, we offer research suggestions to invigorate agriculture and crop productivity by looking to
plant adaptations and interactions among plants in nature for potential solutions in cropping
systems (Figure 1, Key Figure). Natural ecosystems have withstood the test of evolutionary time
and resource deprivation, resulting in plants and plant communities that are resilient when
faced with low amounts of available nutrients and water. The probability is high that the world’s
flora of some 390 000 species includes many with functional and structural attributes that have 1
Department of Biology, University of
not yet been deployed in crop production. In particular, we give attention here to the issues of Florida, Gainesville FL 32611, USA
diversity, N use, P acquisition, water use, and perennial growth habit. 2
Bureau for Food Security, U.S.
Agency for International Development,
Washington DC, WA 20004, USA
While we recognize that undernourishment is a problem with many facets – political, economic, 3
Crop and Soil Sciences Department,
geographic, cultural, and agronomic [5,6] – we focus here on biological research directions that North Carolina State University,
are underexploited and have the potential to increase crop yields of commercial agriculture (i.e., Raleigh, NC 27695, USA

farming enterprises of any size that sell products off the farm) in the near future. Some of our
suggested approaches have the potential to increase yields on existing commercial crop lands, *Correspondence:
but most are directed at making commercial production feasible on environmentally and [email protected] (J.J. Ewel).

Trends in Plant Science, February 2019, Vol. 24, No. 2 https://doi.org/10.1016/j.tplants.2018.10.008 121
© 2018 Elsevier Ltd. All rights reserved.
Key Figure
Resources for plant growth can be present yet unavailable

High internal N

N2

NH4

s
ate
l
Carboxy

PO4

Metal
oxide

s
PO 4
Enzyme

Soil organic
maƩer

Figure 1. Natural systems include many processes that facilitate plant acquisition of those resources – processes
underutilized in commercial agriculture. Examples are depicted here for water (blue), redistributed via roots as conduits
from deep to shallow soil; nitrogen (orange) transformed from N2 to NH3 by microbial symbionts, and retained for recycling
in large-stature plants; and phosphorus (red), released from plant-unavailable forms by plant exudates.

economically marginal lands, including those currently relegated to subsistence agriculture with
the potential to move to the commercial agriculture essential for future development. Several of
the avenues we highlight have been well known to agronomists for decades but need renewed
research attention to stimulate commercial production.

122 Trends in Plant Science, February 2019, Vol. 24, No. 2

Diversity: Solution, or False Hope?
Hundreds of studies in natural and investigator-designed ecosystems have shown that func-
tional diversity can result in enhanced use of light, water, and nutrients, and diminished pest
problems [7]. However, farmers are in the business of capturing resources and managing pests
as means to an end – crop production – and that is where diversity is not necessarily the best
option. Although mixtures of functional groups often yield more than the average of their
component species in monoculture, they rarely yield more than an equal area of the most
productive species in the mixture. Not surprisingly, commercial farmers usually opt for the most
productive species, and they typically grow it in monoculture. Nevertheless, in commercial
agriculture, diversity of plant traits can help suppress pests, increase effectiveness of water use,
enhance pollination, diminish damage due to wind, ensure full use of growing-season sunlight
and precipitation, and restore soil organic matter, just as it does in natural systems.

Subsistence farmers improve resource use, augment dietary variety, and reduce risk through use
of species combinations (e.g., maize–bean–squash in Mesoamerica). Furthermore, some of the
benefits of diversity are routinely achieved in commercial agriculture by temporal diversity (e.g.,
crop rotations and relay cropping) or by spatial diversity (e.g., trap crops, companion crops, and
strip crops). Diversity is also deployed commercially where species identity and relative abundance
in the harvested product are not crucial – forage crops and plant mass for fuel, for example.

Mixing species in the same place at the same time, however, raises the challenges in
commercial cropping to another level. One biological constraint is the design of food-producing
plant assemblages of strong ecological combining ability [8]: mixtures of species having
complementary stature, phenology, and resource acquisition. Some farmers grow mixtures
of species (e.g., canola + pea + lentil) on a commercial scale and report ecological and
economic benefits, but replicated trials with monocultures matched to site conditions are
scarce. This is a research need and opportunity for teams of agronomists and ecologists.

Growing species mixtures for human food may be even more strongly constrained by a
technological issue: complexity management – sowing, weed control, inputs tailored to needs
of species, and harvesting. Consequently, an important research priority is machinery capable
of dealing with such complexity, likely using emerging sensory technology [9,10]. The speed,
precision, control over timing, economies of scale, relief from drudgery, and labor savings that
accompany mechanization are key to increasing food production.

N, the Key Ingredient

Some of the best opportunities for knowledge transfer from natural systems to agriculture may
lie in plant adaptations for mediating biogeochemical cycles and improving plant nutrition. An
essential component of amino acids and nucleic acids, N is of key importance because of its
critical role in plant growth and crop yield (Box 1). Natural grasslands, woodlands, and forests
export relatively small amounts of N through leaching, denitrification, and volatilization. They
retain and recycle most N, and losses are typically replenished by fixation of atmospheric N2 by
microbes, dry and wet atmospheric fallout, and soil reserves [11].

Commercial crops, in contrast, result in the removal of large amounts of N from fields as a
critical component of harvested plant product. High crop yields are only possible if soil stores of
N are replenished, typically by manufactured N fertilizer, applied globally at >100 Tg/year [12].
Furthermore, fertilizer N inputs and outputs tend to be geographically imbalanced. Where N is
scarce or expensive (e.g., much of Africa), soil reserves of N diminish; consequently, crop yields
are poor and sustained commercial production is difficult. In contrast, where N fertilizer

Trends in Plant Science, February 2019, Vol. 24, No. 2 123

 Box 1. Nitrogen
Crop yields closely track N inputs [40], but a broad range of problems face N application, ranging from scarcity to
harmful excess. The current research portfolio on N fertilizer placement, timing, and uptake needs to be increased, and
guidance may be obtained from natural ecosystems. Two properties of natural systems – retention/reuse and N2 fixation
– offer underexploited models potentially useful to agriculture.

Could Bigger Be Better?

The retain-and-recycle-N attribute of natural plant communities might be obtained in crop production through the
propensity of some species to develop large plant stature and to accumulate N in high concentration (see Figure 1 in
main text). High-N, large-stature plants might be crop species themselves or cover crops consisting of a single species
or a diverse mixture. The tissues of these species could be managed for N conservation and reuse based on the more
common use of shoots as green manures or on belowground organs (e.g., the unharvested residual from yam bean,
Pachyrhiza spp., [41]). In rice, work is underway to increase culm size and strength to obtain yield increases [42]; large
culms open the possibility of storing and recycling non-structural N (as well as phosphorus and potassium).

Tweak the Host Plants, not the Microbes

Nearly all N2 fixation research has focused on the bacterial partner in the symbiosis. This seemed a logical option, and
the bacteria lent themselves to ready selection for superior strains. Nevertheless, little has come of this research
because the N2 fixation rate by nodules containing minimally competent bacteria is largely regulated by the host plant
[43]. N2 fixation activity is closely integrated into N and water cycling within the host plant, and it is the flow of N and water
under the regulation of the host plant that determines nodule N2 fixation activity.

applications exceed plant uptake (e.g., China), off-site environmental quality and human health
suffer. It is not surprising then, that N has received more attention in both agronomic and
environmental research than any other mineral nutrient [11,13–15].

Biological fixation (Figure 1) would help solve many crop production challenges associated with
N, yet few N2-fixing species are part of the commercial agriculture portfolio. Only the composite
category of pulses (11 genera of grain legumes grown for human consumption) makes it onto
the list of major food crops [16], and except for soybean (Glycine max L. Merr.), grown for its
seed protein and oil, and clovers (Trifolium spp.) and alfalfa (Medicago sativa L.), both grown for
forage, the land area devoted to legumes is minor although growing in some regions [17].

P – Breaking it Loose from the Soil

P is the critical component of plant molecules involved in energy flow. Consequently, limited
plant uptake of phosphate (the plant-available form of P) severely reduces crop yields. Ironically,
many soils where yields are constrained by P do contain substantial P, but it is unavailable to
most crops because it is complexed with Al, Fe, or Ca, or it is locked into organic matter.
Unavailability can be corrected locally by application of manures, biosolids and P-rich com-
posts [18], but supply and distribution limit the scope of their use. At larger landscape scales,
unavailability is typically corrected with P fertilizer (coupled with lime applications on acid soils),
which will become increasingly difficult as global supplies of rock phosphate tighten and costs
increase [19].

Scarcity of readily available soil P is most acute and extensive in highly weathered (or old) soils of
the tropics and subtropics [20], but P can also become limiting to crop yield in younger soils
after long-term, repetitive harvests. Nevertheless, many plant species in natural ecosystems
have evolved to live, and even thrive, on soils with low available P [21–23]. Studies from
Australia, in particular, provide insights on plant success in mining P from soils having virtually
no readily available P [22,24]. The roots of these plants release particular carboxylates (in many
cases via specialized cluster roots) capable of replacing P that is tightly bound in soil mineral
complexes, making the P available to plants (Box 2).

124 Trends in Plant Science, February 2019, Vol. 24, No. 2

 Box 2. Resolving the Paradox of Soil Phosphorus
The paradox of P scarcity for plant growth in many soils is its presence in the soil in unavailable forms. Plants have
resolved this dilemma with a number of mechanisms, including translocation and reuse, high P-use efficiency (i.e.,
carbon gain per unit P), and, two that we highlight here – making metal-bound P available for uptake, and releasing
phosphate from organic soil P.

A number of plants release carboxylate (of certain small organic molecules) into the rhizosphere to facilitate release of P
from hydrous oxides of iron and aluminum (see Figure 1 in main text). Furthermore, some species meet their nutrient
needs by carrying nutrient acquisition one step further by combining P release and uptake with fixation of atmospheric
N2. This dual-resource-acquisition phenomenon is known for several domesticated legume genera, for example,
Cajanus (pigeon pea), Kennedia, Lupinus (lupines), and Cicer (chickpea) [44–46], and it is likely that among the
thousands of unstudied N2-fixing species there are many more that are effective at P acquisition. A two-pronged
research agenda is called for: screen undomesticated plant species for crop candidates that mine P and fix N2, and, in
crop breeding programs, screen genotypes for host plants with these dual traits.

Many plants depend heavily on P recycled from decomposing plant material and soil organic matter [47,48] (see Figure 1
in main text). While some phosphate is leached from senescent leaves to potentially become a no-cost P source for
microbes and plant roots [49], most P from soil organic matter is obtained by plants using enzymes to release the P
[50,51]. Organic soil P exists in several forms: some forms are more abundant than others [52], and some are more
readily accessible than others [53]. Pairing the particular chemical characteristics of soil organic P with plant-rhizosphere
mechanisms that mineralize that form of P would go a long way toward enhancing P uptake and recycling in many
agricultural systems.

In addition to cluster roots, natural systems have solved the dilemma of scant available P in
many other ways (Figure 1). For example, a recent study of Panamanian forests found that two-
thirds of 541 tree species were associated with low-P (<2 ppm dissolved phosphate) soils [23].
Nevertheless, few of those solutions that evolved in trees have worked their way into com-
mercial crop plants. One exception is the commercial application of plant-growth promoting
rhizobacteria (PGPRs), which demonstrably enhance plant growth, although scientists do not
yet fully understand the processes involved [25,26]. Eventual understanding of the mechanisms
that enable a rich diversity of species to perform well in low-P environments, and the ability to
fine-tune PGPRs for particular species and soils, could lead to an array of underexplored
options for food production.

The phosphate released from metal oxides or organic matter, as described in Box 2, may be
only briefly available for plant uptake, as it can quickly revert to low-availability chemical forms or
be immobilized by soil microbes. Thus, in addition to better information on P-release mecha-
nisms, research on immediate capture of released phosphate is needed. Abundance and
distribution of fine roots, and effective mycorrhizae for crop plants, are two areas that merit
further investigation.

Water, the Geographic Limiter

Globally, rain-fed agriculture is constrained by climate. Food production is constrained in
altitude and latitude by temperature, and, in regions where temperature permits agriculture,
production is centered on climates in which annual rainfall does not greatly deviate from
potential evapotranspiration. Wetter climates present a host of biological constraints (herbi-
vores, diseases, and weeds) and soil impoverishment due to high rates of nutrient leaching,
while drier areas mean abiotic constraints such as water scarcity and salinity [27].

In many regions of dry climate, water is available at depth in the soil but beyond the reach of the
roots of annual crops, at least during part of the growing season. There are two ways to access
that water: deep-penetrating roots by the crop itself, or potentially a companion species that

Trends in Plant Science, February 2019, Vol. 24, No. 2 125

 Box 3. Bioirrigation
Water moves in plant roots from high hydrostatic pressure to low hydrostatic pressure. One consequence of this
physical process is the possibility of water transport through roots from regions deep in the soil where water is plentiful
up to dry soil surface layers. Subsequent release of the water into the upper soil layers makes water available for uptake
for both the uplifting plant and its neighbors (see Figure 1 in main text). Hydraulic redistribution has been documented for
decades in tree-dominated natural ecosystems [54,55], but putting it into practice for crop production has proven
difficult.

Across 26 empirical studies, hydraulic redistribution averaged 0.3 mm/day (range 0.04–1.3) and accounted for an
average of about 15% of annual transpiration demand [56]. These rates are small, but sufficient to facilitate plant survival
during drought; a feature that may be particularly relevant in perennial production systems [57]. Interspecific transfer of
modest amounts of water via hydraulic redistribution has been documented between deep- and shallow-rooted
perennial forage legumes [58] and between annual crops – deep-rooted pearl millet (Pennisetum glaucum) and drought-
susceptible rice (cultivar NERICA-4) [59].

In addition to the small amount of water made available, interspecific competition is a barrier to use of hydraulic
redistribution in cropping systems: the lifting species is almost invariably larger than the potential beneficiary. One
innovative attempt to mitigate this problem involved detopping a deep-rooted forage species grown adjacent to oilseed
rape (Brassica napus); some additional water became available to the rape and enhanced its survival [60].

There is reasonable expectation that hydraulic redistribution could prove useful in perennial-crop agriculture and
agroforestry, but extending its application into annual cropping schemes is uncertain. Nevertheless, if successful, it
might reduce the interannual variance in crop yields in water-marginal environments where a small amount of additional
water at a crucial stage in the plant life cycle could mean the difference between no yield and some yield.

can tap into deep water and transport it for release into soil layers accessible by the crop plants
(Box 3 and Figure 1).

Breeding plants that use water at a reduced rate early in the growing season are another way to
address water limitations. Unless the water is subsequently lost to evaporation or weeds, early-
season conservation can make more water available to sustain physiological activity, particu-
larly seed growth, during late-season drought [28]. Early-season water conservation can be
achieved through partial stomatal closure either under high atmospheric vapor pressure deficit
or early in the soil drying cycle. While such stomatal response is expressed in ancestral lines of
crop species, selection pressure for high yields – commonly under well-watered conditions –
seems to have de-emphasized them. Recent research has now identified a few genotypes in
nearly all major crops with these traits, and commercial varieties are being developed specifi-
cally for their expression [29].

Perennial Hope
Perennial plants dominate almost all natural terrestrial ecosystems, and the environmental
advantages of long-lived plants – soil protection, buildup of soil organic matter, within-plant
nutrient storage, and more complete exploitation of soil water and nutrients – are well
documented [30]. Thus, it seems incongruous that annual plants dominate food production.
How did that come to be? The leading hypothesis is that the preadaptation of annual plants to
the disturbed sites deployed by humans for agriculture led to an easy path for domestication of
annuals [31]. The potential advantages of perennials are counterbalanced by several potential,
but not inevitable, disadvantages, including comparatively low yields, risk of pest buildups, and
long lead time prior to harvest.

Although efforts have been underway for decades to perennialize major grain crops, [32],
efforts to date constitute a small fraction of what has been dedicated to the development of
annual crops. Therefore, it is perhaps not surprising that annual grain lines developed to take full

126 Trends in Plant Science, February 2019, Vol. 24, No. 2

advantage of a favorable growing season remain more productive. Nevertheless, in today’s Outstanding Questions
world of changing interannual environments, and a near-future world in which fertilizers and What are the levels of benefit as
water are less readily available, it is appropriate to devote more attention to exploring and observed in natural ecosystems that
can be expected from temporal parti-
increasing the productivity of perennial species that are outside of current agricultural practices. tioning in resource use among species
An encouraging recent report [33] describes a perennial rice hybrid that produced yields in such practices as relay cropping and
comparable to those of locally used varieties through four cropping cycles; farmers preferred intensive crop rotations? Which bene-
fits are unobtainable except through
the perennial because of its lower labor requirements.
increasing the number of species?

One advantage of perennials is their resilience in response to varying environments. Another Are some benefits of functional diver-
advantage is that they could offer improved acquisition of available N, P, and water through the sity obtainable from attributes of single
annual oscillations of growth conditions. For example: species? Examples might include
 Gene sequencing evidence identifies certain phylogenetic lineages in which establishment of larger-stature plants that retain greater
nutrient mass; litter quality that fosters
plant–microbe symbioses that fix N2 are likely to be far more feasible than others [34]. Most of heterotrophic N fixation; and rapid
the precursor lineages are woody legumes [35]. growth and extensive resource acqui-
 Ectomycorrhizae are known for effective P acquisition on the poorest soils; all ectomycor- sition by perennial roots on herba-
ceous species.
rhizal plants are perennials [36,37].
 Water and nutrient acquisition from great soil depth is more readily accomplished by the
Nutrient retention in plant tissues
extensive, permanent root systems of perennials than by annuals. reduces the likelihood of loss through
 Perennial plants have a far greater array of chemical (both quantitative and qualitative), leaching, and recycling of previously
physical, and biological defenses (including symbiotic arthropods) against herbivores than do accessed nutrients reduces the need
for fertilizer. Are there crop species, or
annual plants [38]. potential crop species, that accrue
As soil carbon storage, erosion reduction, and fertilizer costs become increasingly important large amounts of essential mineral
criteria in crop choice, the potential value of perennials is likely to become more widely nutrients and retain a substantial por-
appreciated. tion in postharvest debris?

Which host-plant traits lend them-

Concluding Remarks selves most readily to modification
Increasing food production is challenge enough, without the added impediment of doing so in likely to lead to enhanced N2 fixation?
the face of potentially reduced supplies of fuels, fertilizers, and water. Clearly, we need
additional efforts on all fronts. Are there opportunities to domesticate
perennial plants that have high yield
potential? Annual grains have high
We have identified several plant adaptations and mechanisms found in natural ecosystems food yields, but perennials excel at
related to acquisition of N, P, and water that could lend themselves to cropping systems, and resource acquisition and nutrient
surely there are many more. Some processes in nature are reasonably well understood, while retention. So far, attempts to make
perennials of annuals have faced sig-
others require more research before they are likely to contribute to food production (see
nificant challenges. Is it time to take the
Outstanding Questions). opposite approach, and emphasize
the domestication of perennials?
Innovations with the potential to increase food production confront many barriers in addition to
limited biological research; a point we illustrate with N2-fixing crops. Given that symbiotic N2-
fixing plants typically have high protein content, and the agronomic benefits of legumes have
been touted by scientists for more than a century [39], why has the diversity of commercially
grown legumes not grown larger? Relatively small trading of legume products in the market
place, lack of farmer awareness of agronomic benefits of including legumes in crop rotations,
and limited investment in research including the lack of financial incentives for seed companies
to invest in self-pollinated species (such as most legumes) are among the reasons. Parallel
barriers would likely be encountered by any proposed change, whether it involved resource
acquisition or crop identity.

A common thread among the resources we highlight is that they are often present yet
unavailable: N2 in the atmosphere calls for biological fixation; P chemically bound in soil calls
for biologically mediated release; and water at depth calls for vertical transport. Today, much

Trends in Plant Science, February 2019, Vol. 24, No. 2 127

crop production operates on the basis of adding resources, rather than accessing potential
supplies that are biologically difficult to access. Resource scarcity is a matter of both amount
and accessibility. Cropping systems in many parts of the world would benefit from incorpo-
ration of traits and processes that enable them to access resources that are present but just
beyond reach.

Acknowledgments
Hannah L. Chelgren prepared the figure. We thank two anonymous reviewers for comments. The views expressed in this
publication do not necessarily reflect the views of the US Agency for International Development or the United States
Government.

References
1. United Nations (2017) World Population Prospects: The 2017 19. Cordell, D. and White, S. (2014) Life’s bottleneck: sustaining the
Revision, Key Findings and Advance Tables. Department of Eco- world’s phosphorus for a food secure future. Annu. Rev. Environ.
nomic and Social Affairs, Population Division. Working Paper No. Resour. 39, 161–188
ESA/P/WP/248 20. Walker, T.W. and Syers, J.K. (1976) Fate of phosphorus during
2. Alexandratos, N. and Bruinsma, J. (2012) World Agriculture pedogenesis. Geoderma 15, 1–19
Towards 2020/2050: the 2012 Revision. ESA Working paper 21. Haridasan, M. (2008) Nutritional adaptations of native plants of
No. 12-03, Food and Agriculture Organization of the United the cerrado biome in acid soils. Braz. J. Plant Physiol. 20,
Nations 183–195
3. Grassini, P. et al. (2013) Distinguishing between yield advances 22. Lambers, H. et al. (2013) Phosphorus nutrition of phosphorus-
and yield plateaus in historical crop production trends. Nat. sensitive Australian native plants: threats to plant communities in
Commun. 4, 2918 a global biodiversity hotspot. Conserv. Physiol. 1, 1–20
4. Ullstrup, A.J. (1972) The impacts of the southern corn leaf blight 23. Turner, B.L. et al. (2018) Pervasive phosphorus limitation of tree
epidemics of 1970–1971. Annu. Rev. Phytopathol. 10, 37–50 species but not communities in tropical forests. Nature 555,
5. Muller, A. et al. (2017) Strategies for feeding the world more 367–370
sustainably with organic agriculture. Nat. Commun. 8, 1290 24. Lambers, H. et al. (2011) Phosphorus nutrition of Proteaceae in
6. Poore, J. and Nemecek, T. (2018) Reducing food’s environmental severely phosphorus-impoverished soils: Are there lessons to be
impacts through producers and consumers. Science 360, learned for future crops? Plant Physiol. 156, 1058–1066
987–992 25. Pieterse, C.M. et al. (2016) The soil-borne supremacy. Trends
7. Hooper, D.U. (2005) Effects of biodiversity on ecosystem func- Plant Sci. 21, 171–173
tioning: a consensus of current knowledge. Ecol. Monogr. 75, 26. Verbon, E.H. and Liberman, L.M. (2018) Beneficial microbes
3–35 affect endogenous mechanisms controlling root development.
8. Harper, J.L. (1977) Population Biology of Plants, Academic Press Trends Plant Sci. 21, 218–229
9. Shi, Y. et al. (2016) Unmanned aerial vehicles for high-throughput 27. Ewel, J.J. (1999) Natural systems as models for the design of
phenotyping and agronomic research. PLoS One 11, e0159781 sustainable systems of land use. Agrofor. Syst. 45, 1–21
10. National Academies of Sciences, Engineering, and Medicine 28. Sinclair, T.R. et al. (2017) Limited-transpiration response to high
(2018) Science Breakthroughs to Advance Food and Agricultural vapor pressure deficit in crop species. Plant Sci. 260, 109–118
Research by 2030, National Academies Press 29. Sinclair, T.R. (ed.) (2017) Water-Conservation Traits to Increase
11. Robertson, G.P. and Vitousek, P.M. (2009) Nitrogen in agricul- Crop Yields in Water-Deficit Environments: Case Studies,
ture: balancing the cost of an essential resource. Annu. Rev. Springer
Environ. Resour. 34, 97–125 30. Crews, T.E. and DeHaan, L.R. (2015) The strong perennial vision:
12. FAO (2017) World Fertilizer Trends and Outlook to 2020. a response. Agroecol. Sustain. Food 39, 500–515
Summary Report, Food and Agriculture Organization of the 31. Van Tassel, D.L. et al. (2010) Missing domesticated plant forms:
United Nations can artificial selection fill the gap? Evol. Appl. 3, 434–452
13. Mosier, A.R. et al., eds (2004) Agriculture and the Nitrogen Cycle: 32. Crews, T.E. and Cattani, D.J. (2018) Strategies, advances, and
Assessing the Impacts of Fertilizer Use on Food Production and challenges in breeding perennial grain crops. Sustainability 10,
the Environment, Island Press 2192
14. Booth, M.S. et al. (2005) Controls on nitrogen cycling in terrestrial 33. Huang, G. et al. (2018) Performance, economics and potential
ecosystems: a synthetic analysis of literature data. Ecol. Monogr. impact of perennial rice PR 23 relative to annual rice cultivars at
75, 139–157 multiple locations of Yunnan Province of China. Sustainability
15. Galloway, J.N. et al. (2008) Transformation of the nitrogen cycle: http://dx.doi.org/10.3390/su10041086
recent trends, questions, and potential solutions. Science 320, 34. Werner, G.D.A. et al. (2014) A single evolutionary innovation
889–892 drives the deep evolution of symbiotic N2-fixation in angiosperms.
16. FAO (2017) The Future of Food and Agriculture: Trends and Nat. Commun. 5, 4087
Challenges, Food and Agriculture Organization of the United 35. Soltis, D.E. et al. (1995) Chloroplast gene sequence data suggest
Nations a single origin of the predisposition for symbiotic nitrogen fixation
17. Foyer, C.H. et al. (2016) Neglecting legumes has compromised in angiosperms. Proc. Natl. Acad. Sci. 92, 2647–2651
human health and sustainable food production. Nat. Plants 2, 36. Brundrett, M.C. (2002) Coevolution of roots and mycorrhizas of
16112 land plants. New Phytol. 154, 275–304
18. Castán, E. et al. (2016) Managing the value of composts as 37. Corrales, A. et al. (2018) Ectomycorrhizal associations in the
organic amendments and fertilizers in sandy soils. Agric. Ecosyst. tropics – biogeography, diversity patterns and ecosystem roles.
Environ. 224, 29–38 New Phytol. 220, 1076–1091

128 Trends in Plant Science, February 2019, Vol. 24, No. 2

38. Coley, P.D. and Barone, J.A. (1996) Herbivory and plant defenses 50. McGill, W.B. and Cole, C.V. (1981) Comparative aspects of
in tropical forests. Annu. Rev. Ecol. Syst. 27, 305–335 cycling of organic C, N, S and P through soil organic matter.
39. FAO (2017) Would You Judge a Book by Its Cover? Book Cover Geoderma 26, 267–286
Designs on Pulses and Legumes from the FAO Library Collec- 51. Tarafdar, J.C. and Claasen, N. (1988) Organic phosphorus com-
tions. Food and Agriculture Organization of the United Nations pounds as a phosphorus source for higher plants through the
40. Sinclair, T.R. and Weiss, A. (2010) Principles of Ecology in Plant activity of phosphatases produced by plant roots and micro-
Production, CAB International organisms. Biol. Fertil. Soils 5, 308–312

41. Castellanos, J.Z. et al. (1997) Symbiotic nitrogen fixation and yield 52. Condron, L.M. et al. (2005) The chemistry and dynamics of soil
of Pachyrhizus erosus (L) Urban cultivars and Pachyrhizus ahipa organic phosphorus. In Phosphorus: Agriculture and the Envi-
(Wedd) Parodi landraces as affected by flower pruning. Soil Biol. ronment (Sims, J.T. and Sharpley, A.N., eds), pp. 87–121, ASA-
Biochem. 29, 973–981 CSSA-SSSA

42. Hirano, K. et al. (2017) Engineering the lodging resistance mech- 53. Turner, B.L. (2008) Resource partitioning for soil phosphorus: a
anism of post-Green Revolution rice to meet future demands. hypothesis. J. Ecol. 96, 698–702
Proc. Jpn. Acad. B 93, 220–233 54. Richards, J.H. and Caldwell, M.M. (1987) Hydraulic lift: substan-
43. Sinclair, T.R. and Nogueira, M.A. (2018) Selection of host-plant tial nocturnal water transport between soil layers by Artemesia
genotype: the next step to increase legume N2 fixation activity. J. tridentata roots. Oecologia 73, 486–489
Exp. Bot. 69, 3523–3530 55. Dawson, T.E. (1993) Hydraulic lift and water use by plants:
44. Ae, N. et al. (1990) Phosphorus uptake by pigeon pea and its role implications for water balance, performance and plant-plant inter-
in cropping systems of the Indian Subcontinent. Science 248, actions. Oecologia 95, 565–574
477–480 56. Neumann, R.B. and Cardon, Z.G. (2012) The magnitude of
45. Veneklaas, E.J. et al. (2003) Chickpea and white lupin rhizosphere hydraulic redistribution by plant roots: a review and synthesis
carboxylates vary with soil properties and enhance phosphorus of empirical and modeling studies. New Phytol. 194, 337–352
uptake. Plant Soil 248, 187–197 57. Sinclair, T.R. (2000) Model analysis of plant traits leading to
46. Ryan, M.H. et al. (2012) Carbon trading for phosphorus gain: the prolonged crop survival during severe drought. Field Crops
balance between rhizosphere carboxylates and arbuscular Res. 68, 211–217
mycorrhizal symbiosis in plant phosphorus acquisition. Plant Cell 58. Pang, J. et al. (2013) Commensalism in an agroecosystem:
Environ. 35, 2170–2180 hydraulic redistribution by deep-rooted legumes improves sur-
47. Chapuis-Lardy, L. et al. (2001) Assessing organic phosphorus vival of a droughted shallow-rooted legume companion. Physiol.
status of Cerrado oxisols (Brazil) using 31P-NMR spectroscopy Plant. 149, 79–90
and phosphomonoesterase activity measurement. Can. J. Soil 59. Izumi, Y. et al. (2018) Water supply from pearl millet by hydraulic
Sci. 81, 591–601 lift can mitigate drought stress and improve productivity of rice by
48. Achat, D.L. et al. (2013) Contributions of microbial and physical- the close mixed planting. Plant Prod. Sci. 21, 8–15
chemical processes to phosphorus availability in Podzols and 60. Sekiya, N. et al. (2011) Applying hydraulic lift in an agroecosys-
Arenosols under a temperate forest. Geoderma 211–212, 18–27 tem: forage plants with shoots removed supply water to neigh-
49. Schreeg, L.A. et al. (2013) Nutrient-specific solubility patterns of boring vegetable crops. Plant Soil 341, 39–50
leaf litter across 41 lowland tropical woody species. Ecology 94,
94–105

Trends in Plant Science, February 2019, Vol. 24, No. 2 129