Page 1
Rapid #: -27733283
CROSS REF ID:
36821152710001865
LENDER:
YGM (SUNY Geneseo) :: Milne Library
BORROWER:
ORE (Oregon State University) :: Main Library
TYPE:
Article CC:CCG
JOURNAL TITLE:
Nature food
USER JOURNAL TITLE:
Nature Food
ARTICLE TITLE:
Soil health contributes to variations in crop production and nitrogen use efficiency
ARTICLE AUTHOR:
Xu, Jianming
VOLUME:
6
ISSUE:
6
MONTH:
YEAR:
2025-06-01
PAGES:
597 - 609
ISSN:
2662-1355
OCLC #:
Processed by RapidX:
10/19/2026 1:02:16 PM
This material may be protected by copyright law (Title 17 U.S. Code)
Page 2
User Name: =
Date and Time: = 2026-10-19
Job Number: = 299267460
Documents (1)
Client/Matter:
-None-
Search Terms:
Soil health contributes to variations in crop production and nitrogen use efficiency
Search Type:
NaturalAnd
Content Type
Narrowed by
news
-None-
1. Soil health contributes to variations in crop production and nitrogen use efficiency
|
About LexisNexis
|
Privacy Policy
|
Terms & Conditions
|
Copyright © 2020 LexisNexis
Page 3
Page 1 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
Soil health contributes to variations in crop production and nitrogen use
efficiency
Nature Food
April 2025
Copyright 2025 The Author(s), under exclusive licence to Springer Nature Limited All Rights Reserved
Section: Pg. 597-609; Vol. 6; No. 6; ISSN: 2662-1355
Length: 8735 words
Byline: Jianming Xu;
1221
Chenchen Ren;
3
Xiuming Zhang;
4
Chen Wang;
2
Sitong Wang;
2
Bin Ma;
2
Yan He;
2
Lingfei Hu;
2
Xingmei Liu;
2
Fangzhou Zhang;
2
Luotian Lu;
2
Shuyao Li;
2
Jiabao Zhang;
5
Yong-Guan Zhu;
6
Peter
Vitousek;
7
Baojing Gu;
1222
1
State Key Laboratory of Soil Pollution Control and Safety, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
jmxu@zju.edu.cn
3
Biosphere Sciences and Engineering, Carnegie Institution for Science, Stanford, CA, USA, 0000 0001 2323 7340,
grid.418276.e,
https://ror.org/04jr01610
4
,
International
Institute
for
Applied
Systems
Analysis,
Laxenburg,
,
Austria,
0000
0001
1955
9478,
grid.75276.31,
https://ror.org/02wfhk785
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
Page 4
Page 2 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
Body
Abstract
Soil health affects both food production and environmental quality. However, quantifying its impact poses a
substantial global challenge due to the scarcity of comprehensive soil health data and the complexity of
disentangling its effects from other variables. Here we integrate high-resolution global data on soil, climate and farm
management practices to assess the contribution of soil health to agricultural productivity. We show that soil
health is responsible for approximately 12% and 22% of global variations in crop production and nitrogen use
efficiency, respectively. While the influence of climate on crop yields is comparable to that of soil health, it is
substantially overshadowed by the role of agricultural management, which accounts for roughly 70% of the global
yield variation. In regions such as China, India and the central United States, the influence of soil health on crop
yields and nitrogen use efficiency is less pronounced due to the dominant effects of farming practices, including
the intensive use of fertilizers. Enhancing global soil health could increase crop yields by 7.8 Mt while reducing
nitrogen surplus by 8.1 Mt worldwide by 2050. It is crucial to achieve global sustainable development through
managing soil health beyond traditional agricultural practices and climate adaptation.
The role of soil health in crop productivity and environmental sustainability is often overlooked. Using a 0.5° × 0.5°
global dataset, this study shows that soil health accounts for 12% of yield variation and 22% of nitrogen use
efficiency variation. Enhancing global soil health could increase yields by 7.8 Mt and cut nitrogen surplus by 8.1
Mt by 2050.
Main
Nourishing a global population of 8 billion is a grand challenge. Annually, over 120 Mt of nitrogen fertilizers are
applied to boost global food production; yet, a substantial portion of this input is not absorbed by crops but lost to
the environment1, leading to substantial environmental consequences, including the degradation of air and water
quality, ecosystem health, and climate stability2. Increasing nitrogen use efficiency (NUE) to promote crop yields
while curbing environmental nitrogen surplus stands is critical for global sustainability3. Predominant research has
traditionally focused on the effects of agricultural management—such as fertilization strategies—and climate
impacts4,5, with soil health, particularly the role of soil microorganisms, being largely overlooked in discussions
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
5
State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of Sciences, Nanjing, ,
China, 0000000119573309, grid.9227.e,
https://ror.org/034t30j35
6
State Key Laboratory of Urban and Regional Ecology, Research Centre for Eco-Environmental Sciences, Chinese Academy of
Sciences, Beijing, , China, 0000000119573309, grid.9227.e,
https://ror.org/034t30j35
7
Department of Biology, Stanford University, Stanford, CA, USA, 0000 0004 1936 8956, grid.168010.e,
https://ror.org/00f54p054
1
State Key Laboratory of Soil Pollution Control and Safety, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
2
College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, , China, 0000 0004 1759 700X,
grid.13402.34,
https://ror.org/00a2xv884
bjgu@zju.edu.cn
Page 5
Page 3 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
around crop productivity and its environmental sustainability. Soil health here refers to a soil condition that can
benefit both yield and environmental quality during crop production (see Methods and Supplementary Information
for more details of soil health definition). Nevertheless, growing evidence highlights the critical role of soil health in
agriculture, particularly within the holistic ‘One Health’ framework, which emphasizes soil stewardship as a
fundamental pillar in the continuum of food production6.
Soil hosts a diverse microbial community7, maintaining optimal conditions for food production and ecosystem
services by regulating physical properties, chemical balance and biological activity8,9. Soil physics facilitates plant
growth by ensuring favourable conditions for water retention and root support10,11, also influencing greenhouse
gas emissions across various soil textures12. Concurrently, soil chemistry and biology enrich the soil with essential
nutrients and offer biological advantages that bolster crop growth13,14. Despite recognition of the importance of
soil health in agriculture, concrete, quantifiable indicators of soil health remain underdeveloped, and their
application in evaluating contributions to crop productivity is infrequent. With the adoption of improved farming
practices, the influence of soil health on achieving higher crop yields and diminishing environmental impacts on
global farmlands has become ever more pivotal. This is crucial not only for safeguarding food security but also for
lessening the burden of agricultural pollution, which is responsible for a major portion of environmental
contaminants across many regions worldwide.
Here we establish a comprehensive global soil health dataset at a 0.5° × 0.5° resolution for the world’s croplands.
This dataset encompasses not only conventional factors such as soil pH, nutrient content and bulk density, but also
accounts for soil bacteria, viruses and fauna (Supplementary Table 1). To quantify the contribution of soil health
to agricultural productivity, our dataset also integrates aspects of agricultural management and climatic conditions—
factors of great interest in prior research. Beyond assessing crop yields (indicated by crop harvested nitrogen per
area), we employ NUE (harvested crop nitrogen divided by total nitrogen input) as a metric to signify the
environmental impact of agricultural practices: the higher the NUE, the lower the detrimental environmental
implications. We integrated data from all crops to incorporate the impact of crop rotation interactions on yield and
NUE across cropland systems. Finally, we analyse the stewardship of soil health and other interrelated
components in the context of global crop production.
Results and discussion
Soil health, climate and management
To equitably assess the contributions of various factors affecting crop yield and NUE, we normalized all potential
influencing variables within each 0.5° × 0.5° grid cell based on their relationships with yield and NUE
(Supplementary Table 1). Subsequently, we grouped these factors into three categories: soil, climate and
management. ‘Soil’ encompasses the conditions and health parameters that underpin crop growth, including
factors such as soil pH, bulk density, soil organic carbon and soil bacteria (Methods). Globally, soil appears to
exhibit a consistent pattern for both crop yield and NUE (Fig. 1a,b), with higher scores occurring in higher latitude
regions such as northern Europe, Russia, New Zealand and the southern portions of Latin America (Fig. 1a).
Conversely, lower soil health scores are typically found in the arid zones of lower latitudes, including the southern
United States, northern Africa and India (Fig. 1b). This similar pattern indicates that enhanced soil health could
simultaneously improve crop yields and NUE.Fig. 1
Global pattern of soil, climate and management drivers for crop nitrogen yield and NUE.
a,b, Soil drivers for yield (a) and NUE (b). c,d, Climate drivers for yield (c) and NUE (d). e,f, Management drivers
for yield (e) and NUE (f). To determine each driver’s value per pixel, the average of each normalized indicator is
employed. The normalization direction of indicators is based on its relationship with yield and NUE, as shown in
Supplementary Table 1. The base map was applied without endorsement using data from the Database of Global
Administrative Areas (GADM) (
https://gadm.org/
).
Source data
Page 6
Page 4 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
Regarding climate (temperature and precipitation), we observed divergent patterns regarding crop yield (Fig. 1c)
and NUE (Fig. 1d). Regions with higher climate scores for crop yields tend to be in lower latitudes, including the
Amazon basin, central Africa and Southeast Asia. In contrast, the NUE is more favourable in higher latitude areas
like northern Europe and Canada, as well as in parts of Latin America4. Higher temperatures and greater
precipitation may bolster crop yields but also exacerbate nitrogen leakage to the environment15, thus diminishing
NUE. Meanwhile, these climate impacts on crop yield and NUE closely interact with management practices.
Agricultural management practices, including nitrogen inputs from chemical fertilizers and manure, pesticides, and
farm size, act as a broad indicator of farming practice changes. These changes encompass increased irrigation, the
adoption of machinery and advancements in technology associated with farm size expansion16,17 (further details
are provided in the ‘Management category’ section of the Supplementary Information). These management factors
exhibit distinct patterns in their influence on crop yield and NUE (Fig. 1e,f). Areas with high management scores,
which correlate with increased crop yields, are identifiable in China, India, South America, North America and
western Europe. These regions are characterized by intensive fertilizer and pesticide application and generally
possess more advanced agricultural technologies compared with other global areas1. On the other end of the
spectrum, regions such as Africa and central Europe receive lower scores due to insufficient fertilizer use, which is
a key limiting factor for yield3. Conversely, higher fertilizer inputs often result in a lower NUE due to the diminishing
yield returns with increased fertilizer usage.
Contributions from different factors
To elucidate the contributions and their functional pathways of soil, climate and management factors on crop yield
and NUE, we employed a structural equation model (SEM; Fig. 2). Among all assessed direct and indirect
influencers, nitrogen input emerges as the predominant factor, with a standardized deviation change of 0.56 in
yield for each standard deviation change in nitrogen input. This finding aligns with the notion that the advent of the
Haber–Bosch process for nitrogen fixation accounts for nearly half of the global crop production18. Soil chemical
content and additive pesticide use are positively corelated with crop yields. In contrast, several elements are
negatively associated with NUE, including nitrogen input, additive pesticide use, air temperature and soil chemical
components. Conversely, larger farm sizes and healthier soil physical and biotic conditions are positively correlated
with NUE, as these factors improve nutrient utilization by crops13,19. Although climatic factors contribute modestly
and directly to crop yield and NUE with standard coefficients of 0.07 and
−0.2,
respectively, they exert substantial
indirect effects via soil health. Air temperature negatively affects soil health, except for biotic features (Fig. 2),
while precipitation positively influences soil physical properties but negatively impacts soil chemical and biotic
aspects. Climate conditions shape soil health by altering moisture levels and texture, key factors influencing soil
viral diversity20, then affecting cropland production and NUE.Fig. 2
SEM of climate, management and soil impacts on agriculture yield and NUE.
**P < 0.05, ***P <0.01. TAP, total annual precipitation; MAT, mean air temperature. The P value is 0.000 except for
biotic
→
ln NUE, where it is 0.764. The numbers adjacent to the lines are the standardized path coefficients,
indicating the degree of standard deviation changes in dependent variables if each of the independent variables
changes by one standard deviation. The blue and red lines represent positive and negative effects, respectively.
The arrow width is proportional to the strength of the relationship. Direct effects are shown by arrows connecting
variables directly, while indirect effects occur when one variable influences another through intermediate variables.
The overall R2 values for the model are 0.78. The GFI of the SEM model can be found in the Supplementary
Information.
Based on the SEM results, we quantified the contributions of soil, climate and management factors to crop yield or
NUE fluctuations by multiplying the absolute values of the standardized coefficients by the standardized
independent variables (Methods), providing the relative contribution of each factor at the grid level (Fig. 3). Our
quantitative analysis reveals that agricultural management is the principal driver of crop yield variability, accounting
for 70% of the observed changes. This is because we considered not only nitrogen and pesticide inputs but also
farm size, which serves as an indicator of broader shifts in inputs16,17. In contrast, soil health and climate factors
are associated with 12% and 18% of crop yield changes, respectively (Table 1). This result closely aligns with
Page 7
Page 5 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
findings on the impact of soil health on crop productivity from large-scale field experiments across Europe21,
demonstrating strong agreement between our big data analysis and the experimental outcomes. In the case of
NUE, climate and management account for 34% and 44% of the contribution, respectively, with soil health
correlated with the remaining 22%.Fig. 3
Relative contribution of drivers on yield and NUE changes.
a,b, Global pattern of crop nitrogen yield (a) and NUE (b). c,d, Contribution of soil drivers to yield (c) and NUE
(d). e,f, Contribution of climate drivers to yield (e) and NUE (f). g,h, Contribution of management drivers to yield
(g) and NUE (h). The sum of contributions of drivers to yield or NUE is 100%. The base map was applied without
endorsement using data from the GADM (
https://gadm.org/
).
Source data
Table 1
Summary of effects from the SEM
Direct effect
Indirect effect
Total
Yield
NUE
Yield
NUE
Yield
NUE
Effect
Ratio
Effect
Ratio
Effect
Effect
Ratio
Ratio (%)
(%)
(%)
(%)
Climate
MAT
7
?0.2
16
0.070
0.158
18
34
TAP
0.07
0.045
0.059
Manage
Farm
?0.02
70
0.2
44
70
44
ment
size
N input
0.56
?0.24
Additive
0.14
?0.09
Soil
Chemic
0.04
12
?0.17
22
12
22
al
Physical
?0.07
0.09
Biotic
?0.02
0.002
Direct
effectIndirect
effectTotalYieldNUEYieldNUEYieldNUEEffectRatio
(%)EffectRatio
(%)EffectEffectRatio
(%)Ratio
(%)ClimateMAT7−0.2160.0700.1581834TAP0.070.0450.059ManagementFarm
size−0.02700.2447044N
input0.56−0.24Additive0.14−0.09SoilChemical0.0412−0.17221222Physical−0.070.09Biotic−0.020.002
Direct effects are derived from the standardized path coefficient from climate, management and soil to explain
variables based on Fig. 2. Indirect effects are calculated as the product of all effects in a single path. Ratio is the
absolute value of direct or indirect effects divided by the sum of the absolute values of all effects.
Spatially, crop yields and NUE exhibit variation across global regions (Fig. 3a,b). South America, the United
States, western Europe and China show higher crop yields exceeding 1,000 kg N
ha−1
yr−1.
In contrast, Africa,
central Europe and Australia have yields below 200 kg N
ha−1
yr−1.
NUE displays a different pattern, with high
efficiencies (>60%) in South America, the central United States, Europe and Africa, while Asia and Australia
generally fall below 50%. These disparities are attributed to the interplay of soil health, climate and agricultural
management across spatial scales (Fig. 3c–h). In regions with intensive agricultural management, such as Asia,
western Europe and parts of South America, many hotspots exhibit soil health contributions to crop yield
fluctuations below 10%, averaging at 6% (Fig. 3c). However, in North America, Africa and central to northern
Europe, where over-fertilization is not prevalent, soil health gains importance, associated with 30% or more of the
contribution. Remarkably, the contribution of soil health to NUE fluctuations is more substantial than to crop
yield, exceeding 50% in regions such as northern Europe, the southern United States, the Middle East, parts of
Page 8
Page 6 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
Latin America, South Africa and southeastern Australia (Fig. 3d). Yet, in areas with excessive fertilizer use such as
China and India, soil health contributes only 7% to NUE.
The role of climate in crop yield mirrors that of soil health, except in Southeast Asia, with less than 10%
contribution observed in China, western Europe, the central United States and parts of Latin America (Fig. 3e,f).
However, the influence of climate on NUE surpasses its impact on yield, with over 50% contribution noted in the
northern United States and Canada, northern Europe, most of Latin America and Africa, as well as South and
Southeast Asia. Conversely, the Mediterranean coast, the southern United States, parts of Latin America and
southern Australia exhibit about 7% climate contribution to NUE. Agricultural management consistently exerts a
dominant influence on global crop yield, with a prime example being China, where it accounts for 79% (Fig. 3g,h). It
also dictates NUE variations in China, northern India, western Europe and parts of South America, averaging at
60%, while other regions generally see contributions from management practices ranging from 10% to 40%, with
an average of 25%. It indicates that there are interactions among soil health, climate and management at different
spatial scales, emphasizing that localized soil health management is essential for both food production and
environmental sustainability.
Managing soil health
In the face of global change, existing strategies have been employed to manage agricultural practices and adapt to
climate variations to sustain and potentially enhance crop yield and NUE. Therefore, here we address soil health
management as a pivotal component for a sustainable agricultural future. To include soil health into the analysis of
future agriculture, we considered three Shared Socioeconomic Pathway (SSP)–Representative Concentration
Pathway (RCP) scenarios: SSP1–RCP2.6, SSP2–RCP4.5 and SSP5–RCP8.5, hereafter referred to as SSP126,
SSP245 and SSP585, with low, intermediate and high emissions and radiative forcing, respectively. For future
projections, we primarily used the most favourable scenario, SSP126, as our baseline to assess the global impact
of soil health management on nitrogen utilization and surplus. SSP126 envisions the least climate pressure and
the most favourable socioeconomic conditions.
Under the SSP126 scenario, a global yield reduction of approximately 9% by 2050 is anticipated when compared
with 2020 levels, with substantial geographical variability (Figs. 4 and 5). Notable yield decreases are projected
across eastern Europe, Mexico, southern Latin America and India, ranging from
−42%
to
−1%
(Fig. 4a), while
scattered increases may occur in Southeast Asia and Africa. This projected yield reduction is also expected to
associate with a global NUE decrease from roughly 0.50 to 0.47 (Fig. 5b), particularly pronounced in Mexico and
southern Latin America, where some areas see reductions exceeding 20% compared with their 2020 levels (Fig.
4c). However, the regional variation in nitrogen surplus diverges from that of crop yield and NUE (Fig. 4e). An
average increase of 80% in nitrogen surplus is projected for eastern Europe, parts of Africa, the northern United
States and southeastern Latin America. Conversely, regions like northern China, India, western Europe and Mexico
are expected to experience an average 19% decline in nitrogen surplus. Under the SSP245 and SSP585
scenarios, global patterns are similar, but they exhibit more substantial divergences (Supplementary Figs. 1 and
2).Fig. 4
Relative changes of yield, NUE and nitrogen surplus in 2050 under the SSP126 scenario.
a–f, Changes in yield (a,b), NUE (c,d) and nitrogen surplus (e,f). Panels a, c and e are relative changes compared
with those in 2020 under the SSP126 scenario. Panels b, d and f further considered the improvements in soil
management. The base map was applied without endorsement using data from the GADM (
https://gadm.org/
).
Source data
Fig. 5
Trends of crop nitrogen yield, NUE, nitrogen surplus and total crop production under SSP–RCP scenarios
towards 2050 and the cost–benefit of improvement of soil health.
Page 9
Page 7 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
a–d, Crop nitrogen yield (a), NUE (b), nitrogen surplus (c) and total crop production trends (d). Pink lines show
trends under different SSP–RCP scenarios. Green lines further considered the role of soil management. e, Costs
and benefits of improvement of soil health by 2050.
Source data
Integrating soil health management could substantially alter nitrogen usage and surplus on global croplands (Figs.
4b,d,f and 5 and Supplementary Figs. 1 and 2). Practices such as stover management, conservation tillage, soil
amendments and biological disease control, among others (Supplementary Fig. 3 and Supplementary Table 2), are
considered. Under the SSP126 scenario, the implementation of these measures is expected to substantially
mitigate crop yield reductions from a 5% decrease to a 6% increase globally (Fig. 5a), with reversed yield increases
in East and Southeast Asia, North America and South Africa (Fig. 4b). These increases would bring 7.8 Mt more
crop globally, while cutting nitrogen surplus by 8.1 Mt (Fig. 5). Accompanying the enhanced crop yields and less
nitrogen surplus, a surge in NUE is projected for most global regions, ranging from 0% to 20%, except for Middle
America, Argentina, Paraguay, Venezuela and eastern Europe (Fig. 4d). The dual increase in crop yield and NUE
is poised to result in a substantial decrease in nitrogen surplus, averaging 22%, particularly in western Europe,
Asia, North America, southern Africa and Brazil (Fig. 4f).
Cost and benefits
We conducted a cost–benefit analysis of the proposed soil health management practices under different scenarios
to assess their feasibility and inform policy decisions. The initial expenditure required to adopt sustainable
agricultural practices is substantial across all scenarios, standing at a uniform US$155 billion in 2020. This
uniformity symbolizes the considerable capital input needed to instigate a range of techniques such as cover
cropping, no-till farming, crop rotation and the application of organic amendments (Supplementary Table 11).
By 2050, the cost diverges substantially depending on the scenario (Fig. 5e). Under the SSP126 scenario, which
emphasizes sustainability, the implementation costs are the lowest at US$210 billion, aligning with the scenario’s
focus on long-term soil health investments that promote sustainable practices. In contrast, the SSP245 scenario,
which requires more intensive interventions to maintain productivity, sees costs rise to US$318 billion, reflecting the
need for more aggressive measures. The highest costs are observed in the SSP585 scenario, at US$333 billion,
highlighting a lack of emphasis on soil health and a preference for short-term gains over sustainable investments.
Despite these costs, the benefits of these investments are substantial and multifaceted. They encompass increased
grain production, health improvements, enhanced ecosystem services and substantial contributions to climate
change mitigation through carbon sequestration. Notably, the benefits grow markedly as we approach 2050,
particularly in terms of grain output and climate-related gains (Fig. 5). Quantifying these benefits reveals that the
total accrued advantages soar highest in the SSP126 scenario, reaching an impressive US$894 billion. This figure
not only eclipses the US$834 billion of SSP245 but also the US$790 billion associated with SSP585 (Fig. 5). These
figures bring to the forefront the profound economic and environmental dividends that can be reaped from a
commitment to sustainable soil management practices over the long haul. Such practices are not merely cost-
effective; they represent an investment in the very foundation of our food systems and the health of our planet.
Policy implications
The Anthropocene is characterized by human influence on the Earth’s geology and ecosystems, which includes the
profound impacts on soil health. As the bedrock of global food production, the health of our soils is a barometer
for the sustainability of human civilization itself. With the mounting evidence suggesting that protecting soil health
is not only beneficial but essential, there is an urgent need to incorporate comprehensive soil health measures into
global agricultural policies9. Given the favourable cost–benefit analysis, such measures are within reach and carry
the potential for widespread benefits. Here we expand on the proposed policy recommendations.
Enhanced investment in soil health
Page 10
Page 8 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
Governments and international bodies should intensify their commitments to soil health by substantially increasing
funding for research and development. This could include setting up dedicated soil health centres of excellence,
expanding extension services to educate farmers on soil health best practices, and supporting citizen science
initiatives that engage local communities in soil monitoring and conservation efforts. Public–private partnerships
could also play a role in fostering innovation in soil health technologies. To cultivate resilience in food systems,
soil health programmes must be context-specific, accounting for the vast diversity of global ecosystems. Research
should be geared towards developing a deep understanding of regional soil types and their responses to different
management practices. Such knowledge will enable the crafting of bespoke soil health improvement plans that
could range from biochar application in depleted soils to precision agriculture techniques that optimize inputs and
reduce waste22.
Policy integration and climate-smart soil strategies
Soil health should be intricately woven into the fabric of agricultural, environmental and climate policies23. This
means designing policies that not only incentivize sustainable soil management but also ensure that these practices
are resilient to climate variability. Climate-smart soil strategies should be developed with an emphasis on adaptation
and mitigation, incorporating measures such as improved irrigation for water conservation, agroforestry for
enhanced biodiversity and integrated pest management to reduce chemical use24. To ensure the long-term
success of soil health policies, a robust framework for monitoring and reporting changes in soil quality across
different landscapes is required23. This framework should facilitate the collection of standardized data, enabling the
tracking of progress and the effectiveness of soil health interventions. Reporting mechanisms could help in sharing
best practices globally and fostering international cooperation on soil conservation.
Soil health as part of global sustainability goals
Education initiatives aimed at raising awareness about the importance of soil health should be scaled up. School
curricula at all levels should include modules on sustainable agriculture and soil conservation. Public campaigns
could spotlight the role of healthy soils in food security and climate change, potentially shifting consumer behaviour
towards more sustainable food choices and supporting the market for sustainably produced goods. Soil health
should be explicitly included in the SDGs and other international targets. By doing so, it would elevate the profile of
soil health and encourage the integration of soil conservation into broader development and sustainability
objectives.
In essence, these expanded policy recommendations advocate for a holistic, forward-thinking approach to soil
health. They encourage the incorporation of soil stewardship into the larger environmental and agricultural
discourse, ensuring that soil health is not seen in isolation but as part of a broader ecosystem that supports all life
on Earth. With commitment and collaboration across sectors and regions, these policies can contribute
substantially to a more sustainable and secure global food system.
Limitations
Our study, while providing valuable insights, has some limitations that could impact the robustness of our findings.
The evaluation of the relative contributions of agricultural yield and NUE was based on 19 indicators. However,
the limited scope of these indicators may potentially influence the precision of quantifying their relative
contributions. While our analysis primarily focused on soil health and climate, it is important to recognize that
crop varieties and farming practices also play pivotal roles in agricultural productivity. Different crop varieties exhibit
varied responses to soil conditions, nutrient availability and climate, which can alter yield outcomes and NUE.
Future research should aim to refine the soil health dataset and incorporate more detailed data on crop varieties
and management practices to provide a more comprehensive and accurate assessment of the factors driving crop
yields and NUE globally.
Additionally, omitting the square terms of temperature and precipitation in the SEM may introduce some bias into
our quantification. Furthermore, the data for soil health indicators were sourced from various channels with
different resolutions, introducing uncertainties and potential estimation errors in our analysis. Our reliance on cross-
Page 11
Page 9 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
sectional data, rather than longitudinal panel data, also presents challenges in effectively capturing temporal
changes. Utilizing cross-sectional data for future projections introduces an inherent level of uncertainty.
Our study relies on global models and datasets25, 26–27, which may not fully capture localized soil health
interventions. Despite our efforts to incorporate finer-scale data, the 0.5° × 0.5° resolution may overlook some local
variations and interpretations. While our approach provides a broad perspective on the global benefits and costs of
improving soil health, it does not account for the complex realities faced by farmers with varying levels of access
to knowledge, tools and market infrastructure, particularly for smallholder farmers with limited resources. This
generalization could lead to discrepancies in the estimated costs and benefits across different farming contexts. To
address this gap, future research should incorporate ground-truthing through survey-based or field-based
evaluations involving human subjects. This approach would provide more granular data, capturing the unique
challenges and outcomes experienced by diverse agricultural communities at the local scale. Such detailed,
localized data would not only validate and refine the model projections but also enhance their applicability and
relevance across a wider range of farming systems globally, ensuring that recommendations are better tailored to
the specific needs of smallholder farmers and other vulnerable groups.
Despite these limitations, our study represents a systematic analysis of indicators influencing agriculture from the
perspectives of soil, climate and management. We underscore the crucial role of soil health, quantifying its
contribution to yield and NUE, and emphasize the importance of managing soil health for attaining both food
security and environmental sustainability, introducing a novel perspective on the promotion of sustainable
agricultural development.
Methods
Data sources
We categorized our indicator data into three main groups: soil, climate and management. Supplementary Table 3
details the specific indicators and their respective data sources within these categories. The data year specifications
are as follows: nitrogen budget, cropland distribution, population distribution and climate data all correspond to the
year 2020. However, the physical, chemical and microbial property data of the soil do not have a specific year, as
the samples come from a wide range of sources. We utilized nitrogen budget data at a spatial resolution of 0.5° ×
0.5°, including fertilizer input, biological nitrogen fixation, deposition and manure, sourced from the Integrated
Model to Assess the Global Environment–Global Nutrient Model (IMAGE–GNM) database28. Cropland data,
measured in square kilometres per grid at the same spatial resolution, were obtained from the History Database of
the Global Environment (HYDE v. 3.2)29. These datasets represent distinct scales: soil health data, primarily
measured belowground; management data, reflecting aboveground practices; and climate data, representing
surface-level conditions. Each scale influences specific stages of crop growth through unique mechanisms, as
reflected in crop yield and NUE. Soil conditions predominantly affect root development, providing crucial support
for crop growth. Aboveground management practices interact with soil properties, shaping nutrient uptake and crop
productivity, which subsequently influence crop yield and NUE. Climate indicators define the environmental
conditions that further regulate crop growth and yield. Despite operating at different scales, these factors collectively
drive the key processes that determine crop performance and productivity.
Farm size was determined by dividing cropland area by population (cropland population density) with normalization
performed in the negative direction30. We consider farm size a key management factor, as changes in farm size
indicate the shifts in practices, such as increased use of machinery, irrigation techniques, technology and
knowledge31,32. In contrast, small farms, particularly those under two hectares, tend to overapply chemical
fertilizers and pesticides, as observed in China32,33. Given that many management factors are correlated, leading
to multicollinearity, farm size serves as a broad indicator representing a wide range of farming practice changes.
Soil physical and chemical properties, such as soil bulk density, soil pH and soil organic carbon density, were
extracted from the surface layer (0–30 cm) at a spatial resolution of 1 km × 1 km. The mean values of these
properties were calculated. Temperature and precipitation data for the year 2020, at a spatial resolution of 0.5° ×
0.5°, were sourced from version 4 of the Climatic Research Unit Time-Series monthly high-resolution gridded
multivariate climate dataset (CRU TS v. 4.06)34. Yearly averages for temperature and total precipitation were
Page 12
Page 10 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
calculated based on monthly data. All specific metrics were then trimmed to match the extent of the cropland raster
and ensure spatial coherence.
For our scenario analysis, we gathered cropland area, population, temperature and precipitation data under three
SSP–RCP scenarios: SSP1–RCP2.6, SSP2–RCP4.5 and SSP5–RCP8.5. Cropland area data until 2050 for these
scenarios were obtained from the Land-Use Harmonization (LUH2) ‘extension’ datasets35. Population data until
2050 were derived from gridded datasets for population and economy under SSPs36. Monthly temperature and
precipitation data covering the period 2015–2050, at a spatial resolution of 0.9° × 1.25°, were acquired from the
World Climate Research Programme Coupled Model Intercomparison Project Phase 6 (CMIP6)37. These data
were then transformed into year-average temperature and year-total precipitation.
Crop yield and NUE
To quantify the contribution of soil health to global crop yield, it is essential to identify a unified indicator that
reflects plant growth conditions across diverse global regions. Given the variation in harvest amounts among
different crops, including both grain and straw, we use the nitrogen content in the harvest to represent crop yield
across cropland system.12
Here, subscript i denotes grid. N yield represents the total nitrogen harvested per harvest area from croplands,
encompassing both grain (N grain) and straw (N straw) across over 170 different crops globally. Unharvested straw
is considered part of the soil in the cropland system. NUE measures the capacity of croplands to utilize nitrogen. It
is defined as the ratio of nitrogen harvested from croplands to the total nitrogen input (N total) for crop
production, which includes synthetic fertilizers (N fertilizer), manure (N manure, from both livestock and human
sources), biological nitrogen fixation (N BNF), irrigation (N irrigation) and nitrogen deposition (N deposition).345N
surplusi refers to the nitrogen that is not harvested, the majority of which is lost to the environment, while some
may accumulate in the soil and potentially be utilized in the future. Despite the uncertainty surrounding its future
use, N surplus can still serve as an indicator of the environmental impact associated with nitrogen surplus. From a
long-term nitrogen balance perspective, current-year yields also absorb residual nitrogen from previous years,
making nitrogen surplus a reasonable proxy for environmental impacts. Although some lost nitrogen may be
converted into environmentally benign dinitrogen gas (N2), this process still represents environmental impact due to
the energy consumption and emissions involved in nitrogen fertilizer production. All of these nitrogen-related
indicators are estimated using the Coupled Human and Natural Systems (CHANS) model and IMAGE–GNM38,39.
We corrected the raster data from the IMAGE–GNM based on the nitrogen budget of the CHANS model. Some
raster values exceeded reasonable limits, and we set them to the upper limit of the reasonable range. The upper
limit was determined based on Supplementary Table 16.
Framework for quantifying soil health contributions
To integrate data from diverse sources, along with crop yield and NUE, we first resampled all datasets to a uniform
spatial resolution of 0.5° × 0.5° in ArcGIS using a fishnet-based statistical method. The nitrogen budget data, which
are central to this study’s focus on crop nitrogen yield and NUE, were used as the resolution benchmark for
harmonization. For instance, although bulk density data are available at a finer resolution of 1 km × 1 km, aligning
all datasets to 0.5° × 0.5° was necessary to ensure compatibility with phosphorus content, historical temperature
and precipitation datasets (‘Data sources’). As all datasets were provided at the gridded level, this resampling
approach ensures consistency and spatial alignment across indicators (Supplementary Fig. 34).
To ensure consistency across soil factors for large-scale analysis of soil health impacts on crop yield and NUE, we
normalized all data at the raster scale. Indicators were standardized on a scale of 0–1 based on their correlation
direction with crop yield (Supplementary Information). This standardization enabled us to compile various soil
factors in a consistent manner, facilitating large-scale analysis of soil health impacts on crop yield and NUE based
on field experiments. The indicators were then grouped into three main categories: soil health, climate and
management (Supplementary Information). Soil health in this study refers to soil conditions that support both crop
growth and environmental sustainability, characterized by higher yields and lower environmental impacts. It
includes three key sub-categories—physical, chemical and biotic properties—each substantially influencing crop
Page 13
Page 11 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
production. We then included average annual temperature and total annual precipitation as climate control
variables. Management were grouped into three sub-categories: farm size, nitrogen input and additive inputs
(Supplementary Information).
Next, we developed models linking crop yield and NUE (dependent variables) to soil, climate and management
variables (independent variables; see ‘Statistical analysis’). We integrated data from all crops to assess NUE
holistically, as the NUE of specific crop varieties is unlikely to change notably. However, crop mixes and rotation
interactions influence NUE in croplands1. By combining all crops, we aimed to examine how soil health affects
nitrogen yield and NUE across cropland systems. Standardized coefficients from the statistical analysis were used
to quantify each variable’s contribution to crop yield and NUE (‘Contribution quantification’). To ensure the
reliability and robustness of our quantifications, we employed an alternative method to validate our findings and
assessed the uncertainties (Supplementary Information). Finally, we conducted scenario analyses to explore
potential pathways for enhancing crop yield and NUE. We further evaluated the feasibility of these pathways using
cost–benefit analysis.
Statistical analysis
The SEM employed in this study offers a comprehensive approach to simultaneously consider and manage multiple
independent and dependent variables. It provides insights into the strength of relationships between various factors
and the overall model fit. It is a robust method for combining with theoretical frameworks, research design and
control of confounding variables. Stata 16.0 software was utilized for SEM analysis to delve into the pathways and
magnitudes of climate conditions, soil conditions and agricultural management factors’ direct and indirect effects on
agricultural yield and NUE. The direct effect refers to the effect of an independent variable on a dependent variable.
In the SEM path diagram, the coefficient along the direct path between two variables is the direct influence. Indirect
effect is the influence transmitted through mediating variables, where one variable affects the target variable via
others. For example, in the model, the effect of mean air temperature (MAT) on yield through physical is MAT
→
physical
→
yield, and the path coefficient MAT
→
physical:
−0.64,
physical
→
yield:
−0.07.
Therefore, the indirect
effects of MAT on yield are
−0.64
×
−0.07
= 0.045. This is calculated by multiplying all path coefficients along the
mediating variables. The total effect is the sum of both direct and indirect effects. The contribution of each
category is the absolute value of the direct or indirect effects divided by the sum of the absolute values of all effects.
The final SEM results provided insights into direct and indirect effects of soil, climate and farming management, with
standardized path coefficients presented in Fig. 2, Table 1 and Supplementary Table 14.
Here the analytical process of SEM involved determining potential interaction relationships between factors
impacting cropland yield and NUE, incorporating theoretical and empirical knowledge. It was constructed based on
well-established agricultural production theories and empirical evidence (Supplementary Table 1), organizing the
factors that influence crop yield and NUE into logical pathways. Consequently, our model encompasses multi-level
relationships. Beginning with a comprehensive model incorporating all original- variables, we iteratively refined the
model by eliminating variables that did not contribute substantial information, guided by variable significance and
model testing indices (for example, total annual precipitation (TAP)
→
yield versus TAP
→
NUE, whether to include
the variable virus_richness and so on). This iterative process allowed us to identify the optimal model and select
classification variables. The final models, chosen from several alternatives, were assessed based on goodness of fit
statistics and the comparative fit index. The commonly used indices—root mean square error of approximation
(RMSEA, where RMSEA < 0.1 indicates a good model fit) and goodness of fit index (GFI, where GFI > 0.9 indicates
a good model fit)—were employed to evaluate model fit. To meet these criteria and avoid multicollinearity issues,
we incorporated temperature and precipitation but excluded their square terms in the SEM. The calculation of GFI is
obtained by comparing the empty model with the original model. The chi-squared metric does not have a fixed
range, and its size depends on the complexity of the model and sample size. The chi-squared test is meaningless in
this study because of the enormous sample size (n = 24,959). The code availability for the SEM building process is
presented in the Supplementary Information. Explanation of GFIs are in Supplementary Table 4. The overall R2
value for the SEM is 0.78.
Contribution quantification
Page 14
Page 12 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
Based on the SEM results, we obtained the absolute values of standardized coefficients for yield and NUE
associated with each variable (Supplementary Table 5). To evaluate the contribution of independent variables to
yield or NUE fluctuations, we calculated a score by multiplying the absolute value of the standardized coefficients
(
β
) obtained from the SEM by the absolute value of the standardized independent variables (Xij). This score
allowed us to determine the relative contribution of each driving variable to yield and NUE at the grid level using
the following equations:67where subscript i, j and k denote grid, independent variable and category, respectively.
Scoreij represents the contributing score of driving variable j in grid i.
βj
is the standardized coefficient of variable j
to yield or NUE. Xij corresponds to variable j in grid i, which have been normalized by standard deviation.
Σ
kScoreij
is the sum of the contribution scores of driving variables in category k, including soil health, climate and
management.
ΣScore
i,total is the total contributing score of 19 driving variables. The relative contribution of
driving category k in grid i on yield or NUE on grid level is Contributionik (%).
Scenario analysis
To predict the agriculture affected by soil health, we also used a multiple regression model to conduct the cross-
sectional analysis, while controlling for confounding factors. We first estimated the following equation to explore
how climate and nitrogen use affect soil variables:8where subscript i and j denote grid and independent variable,
respectively. Soili denotes the variables of soil health, including physical, chemical and biotic sub-category
variables. Xij includes climate variables (temperature, precipitation and their quadratic terms), log-transformed
population density and its quadratic term. We use temperature and precipitation as standard meteorological
indicators to represent for climate change, consistent with previous research31, 32–33. Moderate temperatures
promote crop growth, while extreme temperatures can cause heat stress and yield losses32,40. Precipitation is
essential as well, particularly in rainfed croplands, as it is a primary water source for crop growth41,42.
βj
is the
coefficient that needs to be estimated.
α
is a constant, and
μ
is an error item. Population density is chosen over
normalized farm size in the future prediction equation to avoid estimation errors caused by the variability of
normalized values, which are influenced by the data distribution.
We then estimated how soil properties affect yield and NUE change using the following equation:9where ln
Agriculturei refers to log-transformed crop nitrogen yield or NUE; Xij includes climate variables (temperature,
precipitation and their quadratic terms), log-transformed population density and its quadratic term, nitrogen use
variables (log-transformed nitrogen fertilizer input and fertilizer–manure ratio) and additive variables. Soili denotes
the variables of soil health, including physical, chemical and biotic sub-category variables.
β
and
γ
are the
coefficients that need to be estimated. Detailed results are in Supplementary Table 9. The summary statistics of
variables are listed in Supplementary Table 10.
To incorporate future projections, we obtained grid-level temperature and precipitation data (2020–2050, every
decade) from the SSP–RCP database to calculate their changes compared with 2020. Additionally, we used data
on grid-level rural populations and cropland areas to derive population density data and their changes compared
with 2020. The changes in soil variables under SSP–RCP scenarios towards 2050 were calculated as
follows:10where f8 is the function related to equation (8) and t is the year.
Finally, the agriculture variables including crop nitrogen yield and NUE under SSP–RCP scenarios towards 2050
were calculated based on equations (11) and (12):1112where f9 is the function related to equation (9),
Δ
denotes
the difference in values between year t and 2020, and
γ
is the coefficient estimated from equation (9). is the
simulated yield or NUE under SSP–RCP scenarios.
To enhance agricultural performance, we assume improvements in soil properties, including physical, chemical and
biotic aspects. We expect a 5% improvement in 2030, 10% in 2040 and 15% in 2050. Based on the coefficients
estimated from a multiple regression model in Supplementary Table 9, positive improvements are anticipated in the
physical score, while negative improvements are expected in the chemical and biotic scores. For grid i, the
predicted value of agricultural variables under soil properties management is calculated as follows:1314where
ρtk
represents the proportion of improvement in soil properties and is the simulated yield or NUE with soil management
under SSP–RCP scenarios. Global mean changes are shown in Fig. 5, while the global patterns of yield and NUE
Page 15
Page 13 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
changes under different scenarios (SSP126, SSP245 and SSP585) are depicted in Fig. 4, and Supplementary Figs.
1 and 2, respectively.
Cost–benefit analysis
Implementation cost
Costs of soil health improvements refer to the near-term expenses and potential income surpluses associated with
adopting regenerative soil practices, including investment in new equipment, seeds, amendments and labour; short-
term reductions in yields; surplus of land in production; and the transition costs as practices are put in place. Here
we identified nine key options (o) for soil health improvement based on the literature review (Supplementary Table
11). We break down the implementation costs (ICo,t) of these mitigation options into three distinct components: (1)
the investment cost (INo,t) over the entire lifecycle of the management options, which may include initial
investment, establishment, training and any necessary replacements or renovations; (2) fixed operating costs
(FOo,t) of repairs, maintenance and administration of the equipment or facilities used; and (3) variable operating
costs (VOo,t) of the immediate expenses or saving on the changing labour, energy, fertilizer, irrigation and any
other inputs. Here we mainly refer to the methodology used in the GAINS model43 and collected related
parameters and indicators of soil health management options from previous literature26,44, and supplemented
with expert judgements to calculate global implementation costs:15
All cost figures from model calculations are measured in constant 2020 US$ in this study. Detailed cost-related
parameters used for specific management options can be found in Supplementary Table 12. The implementation
cost does not account for the marginal cost increase of individual abatement strategies due to technological
advancement.
Benefit assessment
Benefits of soil health improvements accrue over the long term and can be financial, environmental and social,
specifically: (1) financial benefits mainly refer to higher crop yields (Grainbenefit,t); (2) improved human health
(HHbenefit,t) through better water and air quality; (3) environmental benefits (EHbenefit,t) include improved water
and nutrient retention, increased biodiversity and ecosystem services from oxygen (O2) release; and (4) the benefit
of climate change mitigation (CCbenefit,t) through carbon sequestration (fixing carbon dioxide in the air) by crop
photosynthesis and soil organic matter accumulation. Here the total benefits arising from improved soil health
(SHbenefit,t) are defined as the sum of the above-mentioned four types of benefit, as shown in equations (16)–
(20):1617181920where
ΔProduction
i,o,t is the estimated crop production change (kg N
yr−1),
which is derived
from the yield change multiplied by harvest area, due to the implementation of soil health management options o;
Grainprice,t represents the regional weighted grain price (U$ per kg N);
Δ
Ei,o,t is the estimated soil pollutant
reduction in the form of Nr (kg N
yr−1);
and HCostN,t is the weighted nitrogen-related human health damage cost
derived from previous peer-reviewed studies25,45,46. Here the ecosystem benefits are the sum of avoided
nitrogen-related damage cost and the increased oxygen release for the ecosystem. ECostN,t is the regional
weighted damage cost of nitrogen waste to the ecosystem in terms of water and air pollution (2020, US$ per kg N;
Supplementary Table 12).
ΔNPP
i,t refers to the changed net primary productivity (NPP) due to the implementation
of healthy soil options. NPP is frequently used in large-scale studies to estimate the carbon sequestration and
oxygen release of various land types. The values 1.2 and 1.63 are the fixed parameters for the estimation of O2
release and C sequestration for every 1. 00 kg of organic matter produced by vegetation based on the
photosynthesis equation47. is the price of the industrial oxygen production48 and is used here to monetize the
ecosystem value of increased oxygen release; PC is the carbon price49 to monetize the climate benefit.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Acknowledgements
Page 16
Page 14 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
This study was supported by the National Natural Science Foundation of China (42325707 and 42261144001) and
National Key Research and Development Project of China (2022YFE0138200).
Notes
Supplementary informationThe online version contains supplementary material available at
https://doi.org/10.1038/s43016-025-01155-6
.Publisher’s note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Bibliography
[1].Zhang, X. Managing nitrogen for sustainable developmentNature (2015). Vol. 528. Pg 51-59.
[2].Steffen, W. Planetary boundaries: guiding human development on a changing planetScience (2015). Vol. 347.
[3].Stevens, CJ. Nitrogen in the environment: excess nitrogen causes problems in developed nations, but
nitrogen-poor soils threaten food security elsewhereScience (2019). Vol. 363.
[4].Ren, C. Climate change unequally affects nitrogen use and losses in global croplandsNat. Food (2023). Vol. 4.
Pg 294-304.
[5].Ju, XT. Reducing environmental risk by improving N management in intensive Chinese agricultural systemsProc.
Natl Acad. Sci. USA (2009). Vol. 106. Pg 3041-3046.
[6].Larkin, RP. Soil health paradigms and implications for disease managementAnnu. Rev. Phytopathol. (2015).
Vol. 53. Pg 199-221.
[7].Ma, B. A genomic catalogue of soil microbiomes boosts mining of biodiversity and genetic resourcesNat.
Commun. (2023). Vol. 14.
[8].Bünemann, EK. Soil quality – a critical reviewSoil Boil. Biochem. (2018). Vol. 120. Pg 105-125.
[9].McBratney, A. Field, DJ. Koch, A. The dimensions of soil securityGeoderma (2014). Vol. 213. Pg 203-213.
[10].Dunbabin, VM. Modelling root–soil interactions using three-dimensional models of root growth, architecture and
functionPlant Soil (2013). Vol. 372. Pg 93-124.
[11].Moraes, MTD. Soil compaction impacts soybean root growth in an Oxisol from subtropical BrazilSoil Tillage
Res. (2020). Vol. 200.
[12].Shakoor, A. Effect of animal manure, crop type, climate zone, and soil attributes on greenhouse gas emissions
from agricultural soils—a global meta-analysisJ. Clean. Prod. (2021). Vol. 278.
[13].Salim, N. Raza, A. Nutrient use efficiency (NUE) for sustainable wheat production: a reviewJ. Plant Nutr.
(2020). Vol. 43. Pg 297-315.
[14].Bertrand, M. Earthworm services for cropping systems. A reviewAgron. Sustain. Dev. (2015). Vol. 35. Pg 553-
567.
[15].Sinha, E. Michalak, AM. Balaji, V. Eutrophication will increase during the 21st century as a result of
precipitation changesScience (2017). Vol. 357. Pg 405-408.
Page 17
Page 15 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
[16].Ren, C. Fertilizer overuse in Chinese smallholders due to lack of fixed inputsJ. Environ. Manag. (2021). Vol.
293.
[17].Ren, C. Ageing threatens sustainability of smallholder farming in ChinaNature (2023). Vol. 616. Pg 96-103.
[18].Erisman, JW. Sutton, MA. Galloway, J. Klimont, Z. Winiwarter, W. How a century of ammonia synthesis
changed the worldNat. Geosci. (2008). Vol. 1. Pg 636-639.
[19].van Groenigen, JW. Earthworms increase plant production: a meta-analysisSci. Rep. (2014). Vol. 4.
[20].Ma, B. Biogeographic patterns and drivers of soil viromesNat. Ecol. Evol. (2024). Vol. 8. Pg 717-728.
[21].Romero, F. et al. Soil health is associated with higher primary productivity across Europe. Nat. Ecol. Evol. 8,
1847–1855 (2024).
[22].Guo, X. Liu, H. Zhang, J. The role of biochar in organic waste composting and soil improvement: a
reviewWaste Manag. (2020). Vol. 102. Pg 884-899.
[23].Amelung, W. Towards a global-scale soil climate mitigation strategyNat. Commun. (2020). Vol. 11.
[24].Campanhola, C. & Pandey, S. (eds) Sustainable Food and Agriculture (Academic Press, 2019).
[25].Gu, B. Abating ammonia is more cost-effective than nitrogen oxides for mitigating PM2.5 air pollutionScience
(2021). Vol. 374. Pg 758-762.
[26].Gu, B. Cost-effective mitigation of nitrogen pollution from global croplandsNature (2023). Vol. 613. Pg 77-84.
[27].Brink, C. et al. in The European Nitrogen Assessment: Sources, Effects and Policy Perspectives (eds Sutton,
M. A. et al.) 513–540 (Cambridge Univ. Press, 2011).
[28].Beusen, AHW. Van Beek, LPH. Bouwman, AF. Mogollón, JM. Middelburg, JJ. Coupling global models for
hydrology and nutrient loading to simulate nitrogen and phosphorus retention in surface water – description of
IMAGE–GNM and analysis of performanceGeosci. Model Dev. (2015). Vol. 8. Pg 4045-4067.
[29].Klein Goldewijk, K. Beusen, A. Doelman, J. Stehfest, E. Anthropogenic land use estimates for the Holocene –
HYDE 3.2Earth Syst. Sci. Data (2017). Vol. 9. Pg 927-953.
[30].WorldPop Dataset (WorldPop, 2020);
https://www.worldpop.org
/
[31].Zhu, P. Warming reduces global agricultural production by decreasing cropping frequency and yieldsNat.
Clim. Change (2022). Vol. 12. Pg 1016-1023.
[32].Agnolucci, P. Impacts of rising temperatures and farm management practices on global yields of 18 cropsNat.
Food (2020). Vol. 1. Pg 562-571.
[33].Ortiz-Bobea, A. Ault, TR. Carrillo, CM. Chambers, RG. Lobell, DB. Anthropogenic climate change has slowed
global agricultural productivity growthNat. Clim. Change (2021). Vol. 11. Pg 306-312.
[34].Harris, I. Osborn, TJ. Jones, P. Lister, D. Version 4 of the CRU TS monthly high-resolution gridded multivariate
climate datasetSci. Data (2020). Vol. 7.
[35].Ma, L. Global rules for translating land-use change (LUH2) to land-cover change for CMIP6 using
GLM2Geosci. Model Dev. (2020). Vol. 13. Pg 3203-3220.
[36].Jiang, T., Su, B. & Wang, Y. National and provincial population and economy projection databases under
Shared Socioeconomic Pathways (SSP1–5)_v2. Science Data Bank
https://doi.org/10.57760/sciencedb.01683
(2024).
Page 18
Page 16 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
[37].Swart, N. C. et al. Amon_CanESM5 _r1i1p1f1: CMIP6 Monthly Data for Surface Air Temperature and
Precipitation (Canadian Centre for Climate Modelling and Analysis, accessed 15 October 2023);
https://aims2.llnl.gov/search
[38].Gu, B. Ju, X. Chang, J. Ge, Y. Vitousek, PM. Integrated reactive nitrogen budgets and future trends in
ChinaProc. Natl Acad. Sci. USA (2015). Vol. 112. Pg 8792-8797.
[39].Zalasiewicz, J. Williams, M. Steffen, W. Crutzen, P. The new world of the AnthropoceneEnviron. Sci. Technol.
(2010). Vol. 44. Pg 2228-2231.
[40].Lesk, C. Stronger temperature–moisture couplings exacerbate the impact of climate warming on global crop
yieldsNat. Food (2021). Vol. 2. Pg 683-691.
[41].He, L. Rosa, L. Solutions to agricultural green water scarcity under climate changePNAS Nexus (2023). Vol. 2.
[42].Wang, X. Global irrigation contribution to wheat and maize yieldNat. Commun. (2021). Vol. 12.
[43].Klimont, Z. & Winiwarter, W. Integrated Ammonia Abatement—Modelling of Emission Control Potentials and
Costs in GAINS. IIASA Interim Report IR-11-027 (IIASA, 2011).
[44].Beach, RH. Global mitigation potential and costs of reducing agricultural non-CO2 greenhouse gas emissions
through 2030J. Integr. Environ. Sci. (2015). Vol. 12. Pg 87-105.
[45].Sobota, DJ. Cost of reactive nitrogen release from human activities to the environment in the United
StatesEnviron. Res. Lett. (2015). Vol. 10.
[46].Giannadaki, D. Giannakis, E. Pozzer, A. Lelieveld, J. Estimating health and economic benefits of reductions in
air pollution from agricultureSci. Total Environ. (2018). Vol. 622. Pg 1304-1316.
[47].Chen, Q. Research on the evaluation method of the ecosystem services of Earth’s critical zoneJ. Nanjing Univ.
Nat. Sci. (2022). Vol. 58. Pg 1070-1086.
[48].Yue, D. Wang, L. Geng, R. Wang, Q. Dai, Y. Initial assessment of seaweed farming ecological value in coastal
waters of ChinaJ. Agric. Sci. Technol. (2014). Vol. 16. Pg 126-133.
[49].Strefler, J. Alternative carbon price trajectories can avoid excessive carbon removalNat. Commun. (2021). Vol.
12.
[50].He, X. et al. A global dataset of soil total phosphorus concentration in (semi-)natural terrestrial ecosystems.
figshare
https://doi.org/10.6084/m9.figshare.14583375.v9
(2021).
[51].Tedersoo, L. et al. Global diversity and geography of soil fungi. Science346, 1256688 (2014).
[52].Delgado-Baquerizo, M. et al. A global atlas of the dominant bacteria found in soil. Science359, 320–325 (2018).
[53].van den Hoogen, J. et al. Soil nematode abundance and functional group composition at a global scale.
Nature572, 194–198 (2019).
[54].Phillips, H. R. P. et al. Global distribution of earthworm diversity. Science366, 480–485 (2019).
[55].Tang, F. H. M. et al. Risk of pesticide pollution at the global scale. Nat. Geosci.14, 206–210 (2021).
Classification
Language: ENGLISH
Page 19
Page 17 of 17
Soil health contributes to variations in crop production and nitrogen use efficiency
Publication-Type: Magazine
Journal Code: 43016
Subject: SOIL HEALTH (94%); ENVIRONMENT & NATURAL RESOURCES (90%); HEALTH CARE
INFORMATION (90%); SOIL CONSERVATION (90%); SUSTAINABLE DEVELOPMENT (90%); CORPORATE
ENVIRONMENTAL RESPONSIBILITY (89%); ENVIRONMENTALISM (89%); PRODUCTIVITY (89%);
SUSTAINABILITY (89%); SUSTAINABLE AGRICULTURE (89%); LIFE FORMS (79%); CLIMATE CHANGE (78%);
NATURAL SCIENCES (78%); POLLUTION & ENVIRONMENTAL IMPACTS (78%); AIR QUALITY (73%);
RESEARCH REPORTS (72%)
Organization: STANFORD UNIVERSITY (59%); INTERNATIONAL INSTITUTE FOR APPLIED SYSTEMS
ANALYSIS (59%); State Key Laboratory of Soil Pollution Control and Safety; Zhejiang University; College of
Environmental and Resource Sciences; Zhejiang University; Biosphere Sciences and Engineering; Carnegie
Institution for Science; International Institute for Applied Systems Analysis; State Key Laboratory of Soil and
Sustainable Agriculture, Institute of Soil Science; Chinese Academy of Sciences; State Key Laboratory of Urban
and Regional Ecology, Research Centre for Eco-Environmental Sciences; Chinese Academy of Sciences;
Department of Biology; Stanford University
Industry: SOIL HEALTH (94%); CROP PRODUCTION (92%); AGRICULTURE (90%); HEALTH CARE
INFORMATION (90%); SUSTAINABLE DEVELOPMENT (90%); FERTILIZERS (89%); SUSTAINABLE
AGRICULTURE (89%); AGRICULTURAL CHEMICALS (77%)
Geographic: EAST CHINA (92%); ZHEJIANG, CHINA (92%); CHINA (79%); INDIA (79%)
Load-Date: June 23, 2025
End of Document
Document Outline