Pastizales FAO 2023

ISSN 0254-6019
Global assessment of
soil carbon in grasslands
From current stock estimates to
sequestration potential
FAO ANIMAL PRODUCTION AND HEALTH / PAPER 187

FAO ANIMAL PRODUCTION AND HEALTH / PAPER 187
Global assessment of
soil carbon in grasslands
From current stock estimates to
sequestration potential
by
Marta Dondini
Food and Agriculture Organization of the United Nations
Manuel Martin
National Research Institute for Agriculture, Food and the Environment
Camillo De Camillis and Aimable Uwizeye
Jean-Francois Soussana
National Research Institute for Agriculture, Food and the Environment
Timothy Robinson and Henning Steinfeld

Rome, 2023
Required citation:
Dondini, M., Martin, M., De Camillis, C., Uwizeye, A., Soussana, J.-F., Robinson, T. & Steinfeld, H. 2023.
Global assessment of soil carbon in grasslands – From current stock estimates to sequestration potential.
FAO Animal Production and Health Paper No. 187. Rome, FAO. https://doi.org/10.4060/cc3981en
The designations employed and the presentation of material in this information product do not imply the
expression of any opinion whatsoever on the part of the Food and Agriculture Organization of the United
Nations (FAO) concerning the legal or development status of any country, territory, city or area or of its
authorities, or concerning the delimitation of its frontiers or boundaries. Dashed lines on maps represent
approximate border lines for which there may not yet be full agreement. The mention of specific companies
or products of manufacturers, whether or not these have been patented, does not imply that these have been
endorsed or recommended by FAO in preference to others of a similar nature that are not mentioned.
The views expressed in this information product are those of the author(s) and do not necessarily reflect the
views or policies of FAO.
ISSN 0254-6019 [Print]

ISSN 2664-5165 [Online]
ISBN 978-92-5-137550-1
© FAO, 2023, last update 23/02/2023
Some rights reserved. This work is made available under the Creative Commons Attribution-NonCommercial-
ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo/legalcode).
Under the terms of this licence, this work may be copied, redistributed and adapted for non-commercial purposes,
provided that the work is appropriately cited. In any use of this work, there should be no suggestion that FAO
endorses any specific organization, products or services. The use of the FAO logo is not permitted. If the work is
adapted, then it must be licensed under the same or equivalent Creative Commons licence. If a translation of this
work is created, it must include the following disclaimer along with the required citation: “This translation was
not created by the Food and Agriculture Organization of the United Nations (FAO). FAO is not responsible for the
content or accuracy of this translation. The original English edition shall be the authoritative edition.”
Disputes arising under the licence that cannot be settled amicably will be resolved by mediation and arbitration
as described in Article 8 of the licence except as otherwise provided herein. The applicable mediation rules will
be the mediation rules of the World Intellectual Property Organization http://www.wipo.int/amc/en/mediation/
rules and any arbitration will be conducted in accordance with the Arbitration Rules of the United Nations
Commission on International Trade Law (UNCITRAL).
Third-party materials. Users wishing to reuse material from this work that is attributed to a third party, such
as tables, figures or images, are responsible for determining whether permission is needed for that reuse and for
obtaining permission from the copyright holder. The risk of claims resulting from infringement of any third-party-
owned component in the work rests solely with the user.
Sales, rights and licensing. FAO information products are available on the FAO website (www.fao.org/publications)
and can be purchased through publications-sales@fao.org. Requests for commercial use should be submitted via:
www.fao.org/contact-us/licence-request. Queries regarding rights and licensing should be submitted to:
copyright@fao.org.
Cover photo: © Tibor Végh via Unsplash

iii
Contents
Acknowledgements v
Abbreviations and acronyms vi
Context viii
Findings ix
1. Introduction 1
2. Methods 5
2.1 FAO LEAP guidelines 5
2.2 Framework and methodology development 5
2.3 Assessing the baseline soil carbon stocks of grassland systems 7
2.4 Assessing carbon input levels needed to maintain current soil
organic carbon stocks 11
2.5 Assessing soil organic carbon sequestration potential in grassland
systems 11
3. Results 15
3.1 Global baseline soil organic carbon stocks 15
3.2 Assessment of current carbon stock levels 20
3.3 Soil organic carbon sequestration potential 26
4. Discussion 29
4.1 Baseline soil organic carbon 29
4.2 Soil organic carbon balance 36
4.3 Grassland potential to sequester soil carbon 37
4.4 Sources of uncertainties regarding the baseline soil organic
carbon stocks 38
5. Conclusions and way forward 43
References 47
iv
FIGURES
1. Percent increase of soil organic carbon in response to improved
management 13
2. Regional total (cumulative) soil organic carbon (SOC) estimated for
the year 2010 by the RothC model for improved and unimproved
grassland worldwide 15
3. Regional average of total organic (plant and excreta) carbon input
to the soil in unimproved and improved grasslands 19
4. Regional averages of carbon inputs needed to maintain current
levels of carbon in the soil in unimproved and improved grasslands 20
5. Regional carbon balance (tonnes C/ha/year) for unimproved
and improved grassland systems 22
6. Soil organic carbon (SOC) sequestration potential after 20 years
of application of best management practices for all available grassland
soils (i.e. those not excluded from the analysis as high SOC or sandy soils) 26
7. Correlation matrix of main variables used to drive the RothC model 39
MAPS
1. Spatial distribution of unimproved and improved grassaland systems 9
2. Baseline soil organic carbon (SOC) stocks (tonnes C/ha) in the top 30 cm
in improved grasslands 17
3. Baseline soil organic carbon (SOC) stocks (tonnes C/ha) in the top 30 cm
in unimproved grasslands 18
4. Global carbon input levels (tonnes C/ha/year) needed to maintain
current soil organic carbon stocks under improved and unimproved
grasslands 21
5. Carbon balance (tonnes C/ha/year) for improved grassland systems 24
6. Carbon balance (tonnes C/ha/year) for unimproved grassland systems 25
7. Annual increase in soil organic carbon (SOC) in the top 30 cm
on all available grassland soils globally (i.e. those not excluded from
the analysis as high SOC or sandy soils) 27
TABLES
1. CCI-LC classes reclassified into improved and unimproved grasslands 8
2. Sensitivity analysis of model results (SOC stocks) to changes on main
variables used to drive the RothC model 39
BOXES
1. Assessing the effect of changing management practices on SOC
case study Eastern Africa 31
2. Assessing the effect of pasture intensification on SOC
case study Paraguay 34
v
Acknowledgements
This document was prepared by the Food and Agriculture Organization of the United
Nations (FAO) Livestock Environmental Assessment and Performance (LEAP) Partnership
and the National Research Institute for Agriculture, Food and the Environment (INRAE),
under the direction of Henning Steinfeld (FAO) and technical guidance from Jean-Francois
Soussana (INRAE). The document received inputs from several colleagues on data compi-
lation and analysis including Giuseppina Cinardi (FAO), Alessandra Falcucci (FAO), Monica
Rulli (FAO), Giuseppe Tempio (FAO) and Dominik Wisser (FAO) with specific inputs from
Martial Bernoux (FAO) and Skalsky Rastislav (International Institute for Applied Systems
Analysis, IIASA).
The main authors of this document were: Marta Dondini, Soil Organic Carbon Model-
ling Specialist (FAO), Manuel Martin, Researcher (INRAE), Camillo De Camillis, LEAP Man-
ager (FAO), Aimable Uwizeye, Livestock Policy Officer (FAO), Jean-Francois Soussana, Senior
Scientist (INRAE), Timothy Robinson, Senior Livestock Policy Officer (FAO), and Henning
Steinfeld, Chief NSAL (FAO).
The document was peer-reviewed by Pete Smith (Institute of Biological and Environ-
mental Sciences, University of Aberdeen, United Kingdom of Great Britain and Northern
Ireland), Guillermo Peralta (FAO), Carolina Olivera Sanchez (FAO), Anne Mottet (FAO), Félix
Teillard (FAO) and by the Global Soil Partnership Secretariat led by Ronald Vargas (FAO).
Moreover, the document was reviewed by the Editorial Board of the FAO Animal Production
and Health Division.
Support for editing came from Delanie Kellon. Sara Giuliani provided the communica-
tion and publication support. Administrative support was done by Eva Pardo Navarro and
Maria Pilar Schneider Cruces. Enrico Masci developed the design and layout and Claudia
Ciarlantini provided the production coordination.
vi
Abbreviations and acronyms
4p1000 4 per 1000 initiative

AGDM above-ground dry matter
BAU baseline scenario
BIO microbial biomass
C0 estimated carbon input
C3 plants in which the carbon dioxide is initially fixed by the Calvin cycle in
photosynthesis
C4 plants where the initial carbon fixation occurs in the outer mesophyll
cells, and the Calvin cycle occurs in the inner bundle sheath cells
CAGR carbon in the above-ground residues
Cbal carbon balance
CBGR carbon in the below-ground residues
CExc carbon input from animal excreta
CRes carbon input from plant residues
DM dry matter
DPM decomposable plant material
FAO Food and Agriculture Organization of the United Nations
GAEZ Global Agro-Ecological Zones
GAUL Global Administrative Unit Layers
GHG greenhouse gas
GIS geographic Information System
GLC_SHARE Global Land Cover SHARE
GLEAM Global Livestock Environmental Assessment Model
GPP gross primary production
GSOCseq global soil organic carbon sequestration
GSP Global Soil Partnership
HUM humidified organic matter
HWSD Harmonized World Soil Database
IIASA International Institute for Applied Systems Analysis
INRAE National Research Institute for Agriculture, Food and the Environment
INT intervention scenario
IOM inert organic matter
IPCC Intergovernmental Panel on Climate Change
KJWA Koronivia Joint Work on Agriculture
LC land cover
LCA life-cycle assessment
LEAP Livestock Environmental Assessment and Performance Partnership
NDC nationally determined contribution
vii
NPP net primary production

RothC Rothamsted carbon model
RPM resistant plant material
SDGs sustainable development goals
SOC soil organic carbon
SOC0 initial soil organic carbon stock
SOCBAU soil oganic carbon under the baseline scenario
SOCINT soil organic carbon under the intervention scenario
TAG technical advisory group
UN United Nations
CHEMICAL ELEMENTS AND FORMULAE

C carbon
CO2 carbon dioxide
N nitrogen
UNITS
°C degree Celsius
Gt gigatonne, metric unit equivalent to 1 billion (109) tonnes
Pg petagram, mass unit equivalent to 1015 grams.
t time expressed in years
Mt megatonne, metric unit equivalent to 1 million (106) tonnes
viii
Context
The adoption of the Paris Agreement in 2015 paved the way for countries to commit to the
international response to climate change through the transition to a low-emission economy
and the development of a climate-resilient future. In livestock systems, and indeed in the
whole agricultural sector, there is a need to balance the benefits of animal-source foods and
livestock keeping for nutrition, health and well-being, with the urgent need to reduce green-
house gas (GHG) emissions to tackle the climate crisis, which also threatens food security.
Grasslands contain approximately 20 percent of the world’s soil organic carbon (SOC)
stocks, which implies that they play a significant role in the global carbon and water cycles
(Puche et al., 2019). Soils can act as both sources and sinks of carbon and many grasslands
have suffered losses of SOC because of anthropogenic activities such as intensive livestock
grazing, agricultural uses and other land-use activities. This trend, however, could be reversed
by stimulating plant growth, capturing carbon in the soil, and protecting carbon in highly
organic soils.
Given the important economic, nutritional and environmental roles that grassland sys-
tems play globally, the Livestock Environmental Assessment and Performance Partnership
(FAO LEAP Partnership) funded this study to illustrate the state of soil carbon stocks in
grassland systems and their potential to sequester carbon in the soil.
The aim of this report is to estimate the baseline SOC stocks of grasslands in the year
2010, assess the carbon input levels needed to maintain current SOC stocks, and determine
if such carbon input is available under current conditions. For these purposes, we defined
improved grasslands as managed systems, and unimproved grasslands as systems close to
semi-natural environments. Furthermore, this report aims to estimate the SOC sequestra-
tion potential of grasslands if management practices known to improve SOC sequestration
are implemented worldwide.
©Jonathan Petersson on Pexels
ix
Findings
SOIL CARBON STOCKS IN GRASSLANDS

The present study provides a spatially explicit report on the state of grassland soils and can
be used as a baseline for future work to explore the impacts of livestock management on
soil carbon at country and farm levels. Globally, there was an estimated annual uptake of
63.5 megatonne (Mt) of carbon (C) in the year 2010 in grassland soils to a depth of 30 cm,
with unimproved systems storing slightly higher amounts than improved systems (33.8 vs
29.8 Mt C). On average, in the year 2010 the SOC stock under unimproved grasslands was
53 tonnes C/ha and 50 tonnes C/ha in improved grasslands. The greatest SOC stocks were
found in temperate regions characterized by low decomposition rates and high grassland
productivity. In comparison, the lowest SOC stocks were observed in arid to semiarid grass-
land soils characterized by low biomass production and organic matter decomposition, thus
reducing carbon inputs into the soil. Climatic conditions explained much of the variability of
SOC stock in grassland soils to 30 cm depth, followed by the carbon input to the soil from
plant and animal sources, and clay content.
All together these results highlight the importance of the interaction between climate and
grassland management, with the latter playing a crucial role in the quality and quantity of
organic material entering the soil. Indeed, stabilization of SOC also depends on several soil
properties such as soil pH, which contributes to regulating soil nutrient availability and soil par-
ticles, which protect soil organic matter by stabilizing them against microbial mineralization.
The lack of incentives for farmers to improve management practices, and the difficulty
of accurately monitoring SOC stocks and changes are the main reasons for not including
SOC in several countries’ nationally determined contributions (NDCs) and national com-
munication reports. The results of this report could support the inclusion of SOC targets
in NDCs, which will improve NDCs’ comprehensiveness and transparency for tracking and
comparing policy progress across NDCs.
The uncertainty regarding the input variables, and their distribution and allocation to
different land uses, together with intrinsic model uncertainties, should be carefully taken
into consideration when using the results arising from this work on the current state of the
carbon in the soil, and its potential to be sequestered in grassland soil. The estimation of
the global soil carbon stock is still quite uncertain, and improved geostatistical methods are
urgently needed to reduce the propagation of such uncertainties in soil models. To improve
the accuracy of input data, such as soil, animal and vegetation properties, and C exchange
information, it is crucial to generate local datasets, especially from underrepresented
regions (e.g. Africa), and explore differences among existing datasets.
ASSESSMENT OF CURRENT CARBON STOCK LEVELS

The majority of grassland soils seem to receive enough organic material to maintain cur-
rent carbon stock levels. Improved grasslands needed, on average, higher carbon inputs
x
than unimproved systems to sustain current SOC stocks (2.1 vs 1.3 tonnes C/ha/year).
Moreover, the positive soil carbon balance found in both improved and unimproved sys-
tems globally indicates a potential increase in SOC stocks. Despite such a positive trend
globally, the large spatial variability of these estimates highlights that the soil state at coun-
try level could differ greatly from the global estimates. The majority of grasslands have a
positive carbon balance meaning that the land is stable or even under improved biophysical
conditions. However, negative carbon balance was found in East Asia, Central and South
America, and Africa south of the Equator, meaning that current SOC stocks are likely to be
decreasing due to anthropogenic stresses combined with climatic conditions. No specific
global measurements are currently available, and it is worth mentioning that the diversity of
situations – in terms of climate, soils and management practices – might have been crucial
for soil carbon dynamics in these areas, as represented by the variability of the carbon input
values in grassland systems.
The findings of this analysis show that there is room for additional carbon storage in
some grassland soils. The main recommendations for grassland systems are to prioritize car-
bon returns in deteriorated soils that have a negative carbon balance, and to protect SOC
in areas – particularly under unimproved grasslands – with high carbon stocks. Grasslands
could contribute to the recarbonization of degraded land and the results of the present
study can highlight hotspots where interventions on grasslands are needed to preserve or
increase SOC in the long term.
SOIL CARBON SEQUESTRATION POTENTIAL

This study found that if the SOC content in the 0–30 cm depth layer of available grass-
lands increased by 0.3 percent after 20 years of the application of management practices
that enhance SOC sequestration, 0.3 tonnes C/ha/year could be sequestered. Sub-Saharan
Africa and South Asia show the highest potential for carbon storage on a per hectare basis
(0.41 and 0.33 tonnes C/ha/year, respectively), followed by Oceania, North America and
East Asia. Low levels of SOC on grasslands, much of it with serious degradation issues,
provide the opportunity to enhance SOC sequestration.
The 4p1000 initiative has identified an aspirational sequestration target of 3.5 Pg
C/year to provide substantive global mitigation. Our estimates suggest that 17 percent
of this target could be reached in the top 30 cm of grasslands and continue over at least
20 years after adoption of SOC enhancing management, such as the incorporation of
animal manures, agroforestry and rotational grazing. This requires that grasslands increase
SOC storage between 0.18–0.41 tonnes C/ha every year. Our estimates do not account for
differences in climate and important soil process issues, notably nutrient and water limita-
tions, biomass production and turnover rates. However, sequestering carbon via increases
in the soil component on grasslands is an achievable and potentially effective route to
quickly increasing carbon sequestration in the near term. Emphasis on future work should
be placed on spatially explicit studies to explore the impacts of livestock management
practices at country level and to monitor management-induced carbon sequestration in
livestock-based ecosystems at farm level.
Despite the large technical potential to sequester carbon in soils, there are often sig-
nificant limitations to achieving that potential in any particular place and within specific
xi
farming systems. In addition, there may be trade-offs with productivity, food security or
hydrologic balances, as well as concerns regarding other GHGs, such as N2O. Therefore, for
a full system budget, it is imperative to include estimates of changes in methane emissions
in order to understand the environmental impacts of management practices on the full
grassland system. Future work should be focused on including soil carbon estimates in life
cycle analyses. The main challenges would be to develop a methodology to allocate SOC
stocks to different livestock units and to account for temporal and spatial dynamics of car-
bon in the soil. Nevertheless, this would enable accurate life cycle assessment of livestock
systems as well as the development of targeted livestock sector-driven national policies for
climate change mitigation and adaptation, and food security.
1
1. Introduction
The adoption of the Paris Agreement in 2015 paved the way for countries to commit to the
international response to climate change through the transition to a low-emission economy
and the development of a climate-resilient future. Since the global annual carbon dioxide
(CO2) emissions from fossil fuels and all other sources are ~10 Gt of carbon (Boden, Marland
and Andres, 2017), soil organic carbon (SOC) has been proposed as a plausible partial climate
mitigation strategy which can offset part of the greenhouse gas emissions (GHG) derived
from anthropogenic activity, with an estimated global sequestration potential of 30–60 Gt of
carbon (Lal, 2004; Sommer and Bossio, 2014), and might buy time while low-carbon tech-
nologies are being developed and adopted. In livestock systems, there is a need to balance
the benefits of animal-source foods and livestock keeping for nutrition, health and well-be-
ing, with the urgent need to reduce GHG emissions to tackle the climate crisis, which also
threatens food security.
In 2017, the 23rd Conference of Parties adopted the Koronivia Joint Work on Agri-
culture (KJWA) to discuss the role of agriculture in climate action while considering the
vulnerability of the sector to climate change and addressing food security. The KJWA plays
a crucial role in enabling the livestock sector to contribute to climate action by mobilizing
knowledge, technology, finance and capacity. It acknowledges the strategic importance
of livestock including key areas such as improved soil carbon sequestration in grazed
grasslands, improved nutrient use and manure management and improved livestock man-
agement systems (Uwizeye et al., 2021). It is therefore evident that assessing the current
state of grassland systems, and their potential to sequester carbon in the soil, is of key
importance for understanding the trade-offs between grassland services on food security,
biodiversity conservation and climate mitigation, and how current grassland management
could be improved to meet climate targets.
Grasslands are ecological communities dominated by grasses with little to no tree or shrub
cover. Some grasslands are natural, while other grasslands have been created from other
forms of vegetation, notably forest. Humans use grasslands for grazing, but not all grasslands
are grazed by domesticated animals. Some may be protected (i.e. grazing is prohibited) and
others are located in regions that simply cannot support them (Garnett et al., 2017).
Grasslands are among the largest ecosystems in the world, occupying 3.5 billion ha
(FAOSTAT, 2016), of which almost 2 billion ha are used for grazing livestock (FAOSTAT,
2016; Mottet et al., 2017).
Natural grasslands (often called rangelands) are dominated by perennial grasses whose
species composition has not been altered to improve livestock productivity.
Improved grasslands (often called pastures) are more intensively maintained, and highly
productive. These grasslands have been modified by sowing more nutrient-rich grasses or
legumes, and by using fertilizers, other amendments and sometimes irrigation to support
more intensive livestock grazing. Improved pastures are species poor. Sometimes the grass
2 Global assessment of soil carbon in grasslands – From current stock estimates to sequestration potential
©Ivan Bandura on Unsplash
is mowed to produce silage for winter feed. The animals themselves may receive feed
supplements, in which case the dung they deposit loads the soil with externally produced
nutrients (Garnett et al., 2017).
Semi-natural grasslands can be broadly defined as ‘habitats created by low-intensity,
traditional farming, or, in some cases, the natural vegetation on poor soils or in exposed
locations’ (Garnett et al., 2017). The semi-natural grassland is a very fluid habitat, which is
amenable for conversion to (and from) arable land and to improved grassland through cul-
tivation, re-sowing and fertilizer application (Garnett et al., 2017). While they tend to pro-
voke a great deal of definitional debate, ‘semi-natural’ grasslands have been defined here
to distinguish them from more intensively managed pastures and from ‘natural’ grasslands.
Soils store significant amounts of carbon as soil organic matter, globally about 2.3 times
more than the carbon in atmospheric CO2 and 3.5 times more than the carbon in all living
terrestrial plants (Yang et al., 2019). Global grasslands are important components of the
terrestrial carbon cycle, storing 119–121 Gt C (Erb et al., 2018) in vegetation biomass, about
343 Gt C in the top one meter of soil (Conant et al., 2017) and a potential soil sequestration
rate of 0.5 tonnes C/ha per year (Henderson et al., 2015; Conant et al., 2017).
1. Introduction 3
The SOC stock in grasslands is determined by the balance between carbon inputs and
outputs. Carbon inputs are derived naturally from the annual photosynthetic carbon uptake
of all leaves in a grassland in the form of root exudates and litter (i.e. gross primary production
(GPP)). The net primary production (NPP) of grasslands is the net carbon stored as new plant
material before harvest and other losses. Additional carbon inputs are derived from animal
manure. Carbon losses occur through natural processes of respiration, decomposition, ero-
sion, leaching, fire and removal of biomass by grazing animals, and by human interventions
through biomass harvesting. Management interventions, such as mowing versus pasture and
grazing intensity (i.e. the fraction of NPP consumed by grazing animals), may particularly
influence the SOC stocks. When grasslands are grazed, biomass ingested by animals con-
tains digestible and non-digestible organic compounds. The non-digestible carbon fraction
(25–40 percent) of the intake is returned to the soil through excreta (i.e. dung and urine). The
digestible part is respired as CO2 shortly after intake (Chang et al., 2015). Only a small fraction
serves to increase animal mass (e.g. muscles) or to form animal products (e.g. milk) which are
exported from the grassland ecosystem (Soussana, Tallec and Blanfort, 2010). Another small
part of the digested carbon is emitted in the form of methane by ruminant enteric fermen-
tation and manure management systems (Sejian et al., 2012).
The soils of managed grasslands contain ~20 percent of the world’s SOC stocks, which
implies that they play a significant role in the global carbon and water cycles (Puche et al.,
2019). Soils can act as both sources and sinks of carbon and many grasslands have suf-
fered losses of SOC because of anthropogenic activities such as intensive livestock grazing,
agricultural uses and other land-use activities. This trend, however, could be reversed by
practices aimed at stimulating root and plant growth (e.g. grazing and nutrient cycling) and
by helping carbon move from above ground to below ground, where it can be captured.
These practices can also stabilize productivity and generate significant social, economic and
environmental benefits.
Current literature suggests no clear relationships between grazing management and
carbon sequestration (Conant et al., 2017). However, positive carbon sequestration was
reported for light-to-moderate grazing intensities (Abdalla et al., 2018), while overgrazing
was found to have a negative effect on SOC stocks (Dlamini, Chivenge and Chaplot, 2016).
The interactions of carbon and nitrogen in soils are of great importance for regulating the
main ecological processes such as nutrient cycling and energy flow (Sardans, Rivas-Ubach
and Peñuelas, 2012). Sufficient nitrogen needs to be available for plants to grow, and there-
fore for soils to sequester carbon. This can be provided in the form of bacterial nitrogen
fixation or the application of mineral fertilizers or organic amendments containing nitrogen
(Liu et al., 2020). These nitrogen inputs to the soil can promote carbon sequestration but
would also cause methane and nitrous oxide to be emitted. Hence, the net GHG balance
will depend upon whether the sequestration gains outweigh these other emissions.
Generally, best management techniques yielding increases in SOC stocks rely on the
management of grazing intensities, as well as increasing forage production through
improved species. The sequestration of soil carbon arising from grassland management
could, therefore, be significant (Lorenz and Lal, 2018) and this, in turn, could have a pos-
itive effect on soil health and other ecosystem services. One critical co-benefit of building
carbon in soil is improved nutrient availability and cycling, which can improve soil fertility
while reducing the need for chemical fertilizers. Numerous soil functions and ecosystem
services depend on SOC and its dynamics. Improvements in soil health, along with an
increase in the availability of water and nutrients, increase soil’s resilience against extreme
climate events (e.g. drought, heat wave) and impart disease-suppressing attributes, which
in turn can also improve animal health. Enhancing and sustaining soil health is also perti-
nent to achieving the goals of the United Nations Decade on Ecosystem Restoration and
advancing the Sustainable Development Goals (SDGs) outlined in the United Nations 2030
Agenda for Sustainable Development (e.g. alleviating poverty, reducing hunger, improving
health, climate action, life on land, and promoting economic development).
This study is part of the deliverables of the Livestock Environmental Assessment and
Performance Partnership (FAO LEAP Partnership). The FAO LEAP Partnership is a mul-
ti-stakeholder initiative that is committed to improving the environmental performance of
livestock supply chains, whilst ensuring their economic and social viability. It is composed of
three stakeholder groups: governments, private sectors, and civil society and non-govern-
mental organizations (NGOs). FAO LEAP Partnership develops comprehensive guidance and
methodology for understanding the environmental performance of livestock supply chains,
and to shape evidence-based policy measures and business strategies. Technical advisory
groups (TAGs) – groups of experts from academia, private sectors and NGOs – are formed
to develop the guidance and methodology for measuring environmental performance.
The soil carbon TAG conducted the background research and developed the core techni-
cal content of the guidelines for measuring and modelling soil carbon stocks and stock
changes in livestock production systems (FAO, 2019). The aim of these guidelines is a
harmonized, international approach for estimating SOC stock and stock changes in live-
stock production systems. A set of methods and approaches is recommended for use by
individual farmers or land managers, those undertaking life cycle assessment of livestock
products, policy makers, and regulators at local, regional or national scales.
Given the important economic, nutritional and environmental roles that grassland
systems play globally, the FAO LEAP Partnership funded this study to illustrate the state of
soil carbon stocks in grassland systems and their potential to sequester carbon in the soil.
The specific objectives of this work were:
• To assess the baseline SOC stocks of grasslands in the year 2010;
• To assess the organic carbon input levels needed to maintain current SOC stocks, and
determine if such carbon input is available under current conditions; and
• To obtain a first estimate of SOC sequestration potential of grasslands, if manage-
ment practices known to improve SOC sequestration are implemented worldwide.
5
2. Methods
2.1 FAO LEAP GUIDELINES

LEAP guidelines for SOC assessment describe the approaches to model SOC stocks and
changes in livestock production systems. Three modelling approaches have been recom-
mended in the LEAP guidelines, namely: empirical models (Level 1), soil models (Level 2)
and ecosystem models (Level 3).
Empirical models estimate SOC stocks and changes using an empirical approach, which
usually represents the observed relationships between SOC stocks or SOC changes and
defined variables (environmental and/or management), such as soil texture, climate, land
use or management practices (Grigal and Berguson, 1998; Davidson and Janssens, 2006).
The LEAP guidelines for SOC assessment recommend using these models to provide a first
estimate of the expected SOC change direction or amplitude.
Soil models estimate SOC stocks and changes by simulating SOC dynamics through
time, considering the effects of climatic and soil factors together with land use and man-
agement variables. Models at this level are process-oriented; they are generally used to pre-
dict SOC dynamics based on different conceptual carbon pools or compartments that vary
in size via inputs, decomposition rates and stabilization mechanisms. Soil models focus on
the processes mediating the movement and transformations of soil carbon only. Each soil
organic matter pool within a model is characterized by its position in the model structure
and its decay rate. Decay rates are usually expressed by first-order rate kinetics (Paustian
et al., 1997) for the concentration of the pool over time. To include estimates of changes
in SOC stocks in life-cycle assessments (LCA), the LEAP guidelines for SOC assessment rec-
ommend using at least a Level 2 model to estimate SOC after a land management change.
Ecosystem models are process-oriented and consider the effects of climate, soil, land
use and management variables on SOC dynamics. However, these models simulate soil
processes other than carbon turnover that may have a direct or indirect impact on SOC
dynamics. Thus, Ecosystem models are built by different sub-models simulating above-
and below-ground plant biomass, soil water dynamics, nutrient dynamics, and their
interactions.
The purpose of the present study was to focus on soil carbon only, therefore Ecosystem
models were not included, while both Empirical and Soil models were integrated into the
methodology.
2.2 FRAMEWORK AND METHODOLOGY DEVELOPMENT

The LEAP recommendations set the benchmark for the development of a framework for
assessing SOC stocks and potential SOC sequestration in grassland systems at global level.
The Tier 2 approach (Level 2 – soil model) recommended in the LEAP guidelines for SOC
assessment was used to estimate the baseline SOC stocks in grassland systems, providing
global reference conditions of grassland soils for the year 2010.
©Gaudenis G. on Unsplash
Among the Soil models suggested in the LEAP guidelines, the Rothamsted Carbon model
(i.e. RothC; Coleman and Jenkinson, 1996) was selected, as it is one of the most commonly
used soil process-based models. RothC simulates the turnover of organic carbon in non-wa-
terlogged topsoil using a monthly time step to estimate total SOC. The model has been wide-
ly tested and used at the plot, field, regional and global scales, using data from long-term
field experiments from different locations (Diels et al., 2004; Pramod et al., 2021).
RothC uses a pool-type approach, describing SOC as pools of inert organic matter (IOM),
humus (HUM), microbial biomass (BIO), resistant plant material (RPM) and decomposable
plant material (DPM). During the decomposition process, material is exchanged between the
SOC pools according to first-order rate equations. These equations are characterized by a spe-
cific rate constant for each pool. These rates are adjusted according to rate modifiers which
are dependent on the temperature, moisture, and crop cover of the soil. The decomposition
process results in gaseous losses of carbon dioxide (CO2). The type of vegetation influences
the distribution of carbon inputs into the RPM and DPM pools, hence the DPM:RPM ratio
typically depends on the vegetation type. In RothC, four vegetation types are considered:
croplands, improved grasslands, unimproved grasslands, and forests with a DPM:RPM ratio
of 1.44, 1.44, 0.67 and 0.25 respectively. For a given total carbon input and mineralization
rate, land use with lower values of the DPM:RPM ratio will exhibit higher total SOC stocks.
2. Methods 7
The Tier 1 approach (Level 1 – empirical model) recommended in the LEAP guidelines
for SOC assessment was selected to explore the global potential of grassland systems to
sequester carbon. This analysis illustrates where and how much carbon might be sequestered
if – through improved practices and management – SOC on grasslands can be increased by
a generally accepted (as attainable) moderate amount, based on the medium sequestration
scenario of Sommer and Bossio (2014) and Zomer et al. (2017). This empirical approach
has been chosen over a process-based methodology to reduce uncertainties on manage-
ment data. As countries, and even farms, adopt different practices based on ecological and
socio-economic constraints, it can be difficult to determine a spatially explicit distribution of
management practices tailored to increase soil carbon. Instead, an empirical approach was
adopted to estimate the percent increase of SOC attainable under improved management,
such as the incorporation of animal manures, agroforestry, rotational grazing, or other prac-
tices that are known to increase soil carbon at the decadal scale. This approach provides a
general framework for countries with limitations on their ability to implement more complex,
intensive data requiring, process-oriented modelling approaches.
The two approaches presented here require specific input data and distinct modelling
assumptions. Full details on data requirements and model initialization of both approaches
are given in the following sections. Due to the nature of the methodology used, the esti-
mated baseline SOC stocks for the year 2010 and the soil carbon sequestration potential
of grassland systems will be analyzed and discussed independently.
2.3 ASSESSING THE BASELINE SOIL CARBON STOCKS OF GRASSLAND

SYSTEMS
The estimation of changes in SOC by the Tier 2 Soil model, due to either land-use or man-
agement changes, requires model initialization. Initialization refers to setting the initial SOC
condition (total SOC and SOC of the different pools) at the start of the period over which
stocks will be estimated, so that further simulated results are realistic estimates.
The Harmonized World Soil Database (HWSD) version 1.2 was used to provide initial
soil conditions in the model (FAO, IIASA, ISRIC, ISS-CAS & JRC, 2012). The HWSD is a
30 arc-second raster database with over 15 000 different soil mapping units that combines
existing regional and national updates of soil information worldwide with the information
contained within the 1:5 000 000 scale FAO-UNESCO Soil Map of the World. The HWSD
provides soil data to a depth of 1 meter at a resolution of 30 arc s (approximately 1 km),
for the dominant soil types in each grid cell. The soil properties used from this database
to drive the RothC model for the top 30 cm soil depth were: organic carbon content, bulk
density, coarse fragments and clay fraction. The RothC model is run for the dominant soil
type (percentage of grid cell area > 50 percent) in each grid cell at a soil depth of 30 cm.
RothC requires monthly precipitation and air temperature data which are used to
determine temperature-based rate modifiers for various soil processes. The annual monthly
statistics on averaged mean temperature and sum of precipitation were derived from the
AgMERRA climate dataset (i.e. 0.5 deg spatial resolution) (Ruane, Goldberg and Chryssan-
thacopoulos, 2015) for years from 1980 to 2010.
Land cover was estimated using the Climate Change Initiative (CCI) Land Cover (LC)
data v2.0.7 (ESA, 2017) for the year 2010. The CCI-LC predictions (300 meters spatial
TABLE 1
CCI-LC classes reclassified into improved and unimproved grasslands
CCI-LC class New class used in this study
Mosaic cropland (>50%) / natural vegetation (tree, shrub, herbaceous cover) (<50%) Improved
Mosaic natural vegetation (tree, shrub, herbaceous cover) (>50%) / cropland (<50%) Improved
Grassland Improved
Mosaic tree and shrub (>50%) / herbaceous cover (<50%) Unimproved
Mosaic herbaceous cover (>50%) / tree and shrub (<50%) Unimproved
Shrubland Unimproved
Sparse vegetation (tree, shrub, herbaceous cover) (<15%) Unimproved
resolution) were grouped into two main categories reflecting the grassland categories
given in the RothC model, namely improved and unimproved grasslands. We reclassi-
fied CCI-LC classes into improved and unimproved grasslands on the RothC assumption
that improved systems are managed systems, while unimproved grasslands are close to
semi-natural environments (Table 1). A full description of the CCI-LC classes and their
direct link with the Land Cover Classification System developed by the United Nations can
be found in the CCI-LC product user guide (ESA, 2017).
Simulations were only performed on these two categories and their distribution is shown
in Map 1. For improved grassland, we used a DPM:RPM ratio of 1.44 (i.e. 59 percent of the
plant material is DPM and 41 percent is RPM). For unimproved grassland, a ratio of 0.67 was
used.
The annual plant residue input was estimated following the Intergovernmental Panel on
Climate Change (IPCC) methodology (IPCC, 2006). The carbon input from plant residues
(CRes) was calculated using the sum of above-ground and below-ground residues, which
were then converted into their carbon content.
= + Eq.1
Where CAGR is the carbon in the above-ground residues and CBGR is the carbon in the
below-ground residues.
Following Chapter 11 of the IPCC methodology (IPCC, 2019), the above-ground res-
idues can be estimated as a function of the total yield and then converted into carbon
content as follows:
=( × 0.3) × 0.475 Eq. 2
Where AGDM is the above-ground dry matter.

MAP 1
Spatial distribution of unimproved and improved grassaland systems
2. Methods
Source: United Nations Geospatial. 2020. Map of the World. United Nations. Cited 22 August 2022. www.un.org/geospatial/file/2285/download?token=puayKYRA
modified with data from ESA, 2017.
9
The same approach can be used to estimate the carbon content from the below-ground
biomass residues:
=( × 0.9) Eq. 3
Where BGDM is the below-ground dry matter.
Data on above-ground dry matter per hectare for the year 2010 were estimated by
the GLEAM 2.0 model (FAO, 2020a). Data on fresh matter yields per hectare, and their
respective land area, were taken from a modified version of Global Agro-Ecological Zones
(GAEZ 3.0) (IIASA and FAO, 2012) and Haberl et al. (2007) to estimate the above-ground
net primary productivity for grassland. To this scope, the fresh matter was converted into
dry matter content, with dry matter being 90 percent of fresh (Opio et al., 2013).
Carbon input from animal excreta (CExc) was estimated from the nitrogen deposited
in 2010, which was calculated by the GLEAM 2.0 model. The function of the GLEAM
‘Manure’ module is to calculate the losses of nitrogen through manure management sys-
tems and the rate at which excreted nitrogen is applied and deposited (FAO, 2020a). The
deposited nitrogen is then converted into carbon by applying a mean C:N ratio of 17.5
(FAO, 2015a), and then used as input in the RothC model.
All above input variables were harmonized to a 1 km spatial resolution and the RothC
model was first initialized using a long ‘spin-up’ simulation (i.e. 10 000 years) (Coleman
and Jenkinson, 1996), to estimate the soil pool ratios by iteratively adjusting the carbon
inputs such that the SOC from the HWSD dataset was reproduced. Then a short ‘warm-up’
simulation of 30 years was used to estimate the soil conditions for the year 2010 (FAO,
2019; FAO, 2020a). The year 2010 was chosen for consistency with the data derived from
the GLEAM 2.0 model and to align the potential SOC estimated by the RothC model with
the GHG emissions estimated by the GLEAM 2.0 model.
The Global Administrative Unit Layers (GAUL) database (FAO, 2015b) was used to ana-
lyze results by region. This choice was made to align the results of this study with the LCA
analysis on livestock systems performed with the GLEAM model.
The Pearson correlation test was used to investigate the dependence between multiple
variables used to drive the RothC model and the simulated SOC (Smith and Smith, 2007).
In particular, correlation tests were performed among simulated SOC (baseline SOC), initial
SOC (HWSD soil carbon data), potential evapotranspiration, air temperature, precipitation,
and organic carbon inputs to the soil.
The sensitivity of RothC to different input parameters was investigated to quantify the effects
of such parameters on the simulated SOC (Smith and Smith, 2007). Only one parameter was
changed at a time, while the others were kept constant. Simulations were run to assess how SOC
was affected by changes in average temperature (increased/decreased by a range from -2 degree
Celsius (°C) to +2 °C with an increment of 1 °C), initial SOC content (decreased/increased by a
range from -50 percent to +50 percent with an increment of 20 percent or 30 percent), and total
carbon inputs (decreased/increased by a range from -50 percent to +50 percent with an incre-
ment of 20 percent or 30 percent). For each scenario, the relative change in SOC was calculated
as a percentage. All analysis was performed using R software version 4.0.3 (R Core Team, 2013).
2. Methods 11
2.4 ASSESSING CARBON INPUT LEVELS NEEDED TO MAINTAIN CURRENT

SOIL ORGANIC CARBON STOCKS
The RothC model is designed to run in two modes: “forward” in which known carbon
inputs are used to calculate the changes in soil organic matter, and “inverse” when carbon
inputs are calculated at an equilibrium state for 10 000 years from known changes in soil
organic matter. In this study, the model was run at equilibrium in inverse mode to predict
the plant input required to maintain current SOC levels. Prior to initializing the model in
inverse mode, the equation developed by Falloon et al. (1998) was used to estimate the size
of the IOM pool from the known SOC stock. The remaining SOC stock (i.e. the total stock
minus IOM) is then used as the input variable, and carbon inputs into the soil are iteratively
adjusted until this input value is reached.
1.139
= 0.049 × Eq. 4
When run in inverse mode, RothC needs only two input data related to management. The
first one is the number of months when soils are left bare. This input variable was set to zero
months for all grasslands, as grassland ecosystems are never left bare. The second input var-
iable is the proportion of carbon inputs to the soil that consists of organic amendments. The
organic carbon inputs to the soil are mainly the result of plant residues and additions of animal
manure and other organic products. In RothC, the fate of carbon provided by plant residues
and organic amendments is specific, reflecting their difference in terms of decomposability.
Therefore, to use RothC for estimating the amount of carbon input needed to maintain current
levels of SOC stocks, the proportion between CRes and CExc was estimated.
The estimated carbon input (C0) was then compared against total carbon input (CRes + CExc)
to assess if the carbon input needed to maintain current SOC levels is available under current
conditions (Martin et al., 2021). The carbon balance (Cbal) of a given soil is therefore defined
as the difference between available carbon inputs (CRes + CExc) and the carbon input (C0), as
estimated with the RothC model.
Cbal = (CRes + CExc) - C0 Eq. 5
If Cbal differs from zero, the steady-state hypothesis is currently not valid. Cbal < 0 indicates
that the current total carbon input is not sufficient to sustain existing SOC stocks, hence
resulting in a SOC declining trend. If Cbal > 0, SOC stocks might be on an increasing trend.
The Global Administrative Unit Layers (GAUL) database (FAO, 2015b) was used to ana-
lyse results by region.
2.5 ASSESSING SOIL ORGANIC CARBON SEQUESTRATION POTENTIAL IN

GRASSLAND SYSTEMS
The global potential of grassland systems to sequester carbon in the soil was estimated by using
an empirical approach based on a methodology developed by Zomer et al. (2017). This approach
estimates the percent increase of SOC attainable after 20 years of improved management prac-
tices. A geospatial analysis is then used to estimate the potential attainable SOC sequestration
in grassland systems and to identify opportunities for SOC sequestration worldwide.
The initial SOC (tonnes C/ha), bulk density (kg/m3) and sand content (weight percent) at
0 – 30 cm soil depth were extracted from the Soils Grid (i.e. 250 m spatial resolution - ISRIC
World Soil Information) (Hengl et al., 2014).
The Global Land Cover SHARE Beta-Release v1.0 (i.e. GLC_SHARE) (FAO, 2014) was used
to identify grassland extent and distribution. This geospatial database provides an estimate of
the percent of land-cover area within a 1 km grid cell. The GLC_SHARE dataset was resam-
pled to allow for the analysis and geoprocessing at the finer 250 m (0.002083333 degrees)
resolution of the soil data. The designation grassland is based upon the UN Land Cover Clas-
sification System, and includes any geographic area dominated by natural herbaceous plants
(grasslands, prairies, steppes and savannahs) with a cover of 10 percent or more, irrespective
of different human and/or animal activities, such as grazing, selective fire management, etc.
Woody plants (trees and/or shrubs) can be present assuming their cover is less than 10 percent.
The geospatial analysis used to estimate the potential attainable increase of SOC on grass-
land after twenty years is described in detail in Sommer and Bossio (2014) and Zomer et al.
(2017). The increase in percent-SOC in response to improved management was described in
Sommer and Bossio (2014) with a four-parameter sigmoid function of the form:
= 0 + − 0 Eq. 6
−
1+
where SOC0 is the initial SOC content (percent), a and b are empirical constants, t is the
time expressed in years, and t0 is the year where the slope of the curve is largest (i.e. the
highest annual sequestration rate). The parameters for the scenario based upon Sommer
and Bossio (2014) were:
= 0.697; = 11.5; 0 =4
The percent increase of SOC after 20 years was calculated from this curve (Figure 1) and
resulted in a value of 0.27.
Bulk density was used to first convert SOC (tonnes C/ha) (as presented in the Soils
Grid 250 m data) into SOC (percent). The estimated percentage increase of SOC (i.e. 0.27
percent increase) was then added to SOC (percent), and the result was converted back to
SOC (tonnes C/ha).
High SOC soils (i.e. soils with a weighted average bulk density (0–30 cm) equal to or
less than 1.0 kg/m3 and/or with more than 400 tonnes C/ha) were excluded from further
analysis. Sandy soils (i.e. sand content at 15 cm equal to or greater than 85 percent) were
also excluded from further analysis. These soils were excluded because their potential for
sequestering carbon would be negligible.
2. Methods 13
FIGURE 1
Percent increase of soil organic carbon (SOC) in response to improved management
0.45
0.4
0.35
0.3
0.25
%C
0.2
0.15
0.1
0.05
0
0 5 10 15 20 25 30 35 40 45 50 55
years
Source: Sommer, R. & Bossio, D. 2014. Dynamics and climate change mitigation potential of soil organic carbon sequestration.
Journal of Environmental Management, 144: 83–87. https://doi.org/10.1016/j.jenvman.2014.05.017
All result grids were converted into World Sinusoidal projection to allow for area calcu-
lations. The GLC_SHARE – Dominant (Class 3 = Grassland) dataset (FAO, 2014), in percent
area of a 1 km grid cell, was resampled to 250 m and multiplied times the various results
(tonnes C/ha) to calculate actual total tonnes of carbon in each grid cell (i.e. given the
actual area of grassland in that grid cell). The Global Administrative Unit Layers (GAUL)
database (FAO, 2015b) was used to analyse results by region.
15
3. Results
3.1 GLOBAL BASELINE SOIL ORGANIC CARBON STOCKS

Globally, grassland soils stored an estimated 63.5 Mt of carbon in the year 2010 at 30 cm
soil depth, with unimproved systems storing slightly higher amounts than improved systems
(33.8 vs 29.8 Mt C). In unimproved grassland systems, the Russian Federation and the Amer-
icas stored the highest amount of SOC among all regions, while South Asia and Eastern
Europe stored less than 1 Mt C each. In improved grasslands, Central and South America
was the region with the highest SOC socks (5.6 Mt C), followed by the Russian Federation
(5.1 Mt C) and East Asia (4.9 Mt C). Eastern Europe was the only region with SOC stocks
values below 1 Mt C, while the other regions ranged from 1.3 Mt C in South Asia to 3.5 Mt
C in North America (Figure 2).
On average, in the year 2010 the SOC under unimproved grasslands was 53 tonnes C/ha
and 50 tonnes C/ha in improved grasslands. Global distribution of SOC is strongly influenced
FIGURE 2
Regional total (cumulative) soil organic carbon (SOC) estimated for the year 2010
by the RothC model for improved and unimproved grassland worldwide
8
7 Unimproved Improved
6
Total C (Mt C)
5
4
3
2
1
0
Russian Federation
Central & South America
North America
Sub-Saharan Africa
East Asia
Oceania
Western Europe
West Asia & Northern Africa
South Asia
Eastern Europe
Source: UN. 2020. Map of the World, modified with data from Coleman, K. & Jenkinson, D.S. 1996. RothC-26.3 - A Model for the
turnover of carbon in soil. In: Powlson, D.S., Smith, P., Smith, J.U., eds. Evaluation of Soil Organic Matter Models. NATO ASI Series,
38: 237-246. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-61094-3_17
©Ralph (Ravi) Kayden on Unsplash
by temperature and precipitation. SOC is generally lower in the tropics where it is hotter
and/or drier, and higher in the cooler and wetter latitudes. The spatial distribution of SOC in
improved (Map 2) and unimproved grasslands (Map 3), and its contribution to total carbon
stock differs substantially from the northern to the southern hemispheres. Most of the world’s
SOC is stored at northern latitudes, particularly in the permafrost and moist boreal regions.
In contrast, large areas of grassland in East Asia, across sub-Saharan Africa and some areas
in North America are found on low carbon density soils.
The regions of the Russian Federation, Europe and North America store the greatest
amount of soil carbon on a per hectare basis in improved systems, with 76 tonnes C/ha,
61 tonnes C/ha and 60 tonnes C/ha, respectively (Map 2). In unimproved systems, these
same regions store higher amounts of carbon in the soil, with values ranging from
92 tonnes C/ha in the Russian Federation to 56 tonnes C/ha in North America (Map 3).
The Russian Federation region accounts for more than 50 percent of all SOC stocks
globally. Together with North America, these two regions appear to have not suffered
human-induced soil degradation.
MAP 2
Baseline soil organic carbon (SOC) stocks (tonnes C/ha) in the top 30 cm in improved grasslands
3. Results
modified with data from Coleman and Jenkinson, 1996.
17
18
MAP 3
Baseline soil organic carbon (SOC) stocks (tonnes C/ha) in the top 30 cm in unimproved grasslands
modified with data from Coleman.
Global assessment of soil carbon in grasslands – From current stock estimates to sequestration potential
3. Results 19
In improved systems, moderate amounts of SOC stocks were found in Central and South
America, East Asia, and West Asia and Northern Africa, ranging from 52 to 56 tonnes C/ha.
South Asia, Oceania, and sub-Saharan Africa regions have very low amounts of SOC, account-
ing for just 3.9 percent of the global total (Map 2).
In unimproved systems (Map 3) moderate amounts of SOC stocks were only found in
Central and South America (49 tonnes C/ha), while in all other regions SOC stock level is
lower than average global figures, ranging from 32 tonnes C/ha (sub-Saharan Africa) to
35 tonnes C/ha (South Asia).
Plant residues and animal manure also affect the SOC stocks. Globally, the average total
yearly carbon input to the soil, for the reference baseline year 2010, was estimated to be
3.23 tonnes C/ha/year in improved systems. In the regions of sub-Saharan Africa and Cen-
tral and South America, the total yearly carbon input to the soil was estimated to be higher
than average figures, with values under improved grasslands reaching 6.7 tonnes C/ha/
year and 5.8 tonnes C/ha/year, respectively (Figure 3). On the other hand, lower amounts
of total carbon inputs to the soil were found in the Russian Federation (2.0 tonnes C/ha/
year) and West Asia and Northern Africa regions (2.1 tonnes C/ha/year). The average total
yearly carbon input to the soil, for the reference baseline year 2010, was estimated to be
2.35 tonnes C/ha/year in unimproved systems, with estimates close to the regional average
of total organic carbon input of all world regions (Figure 3).
FIGURE 3
Regional average of total organic (plant and excreta) carbon input to the soil
in unimproved and improved grasslands
8
Carbon inputs (tonnes C/ha/year)
7
6
5
4
3
2
1
0
Sub-Saharan Africa
Western Europe
Oceania
East Asia
North America
South Asia
Eastern Europe
Russian Federation
South Asia
Eastern Europe
Sub-Saharan Africa
Western Europe
East Asia
Oceania
Russian Federation
North America
Improved Unimproved
3.2 ASSESSMENT OF CURRENT CARBON STOCK LEVELS

Improved grasslands needed, on average, higher carbon inputs than unimproved systems
to sustain current SOC stocks (2.1 vs 1.3 tonnes C/ha/year).
Some of the highest carbon input needs in improved systems were found in sub-Saharan
Africa (6 tonnes C/ha/year), and Central and South America (5 tonnes C/ha/year). On the con-
trary, the Russian Federation, and West Asia and Northern Africa require low carbon inputs to
the soil, with values below 1 tonne C/ha/year. The same regional distribution has been found
in unimproved systems, with the Russian Federation requiring only 0.3 tonnes C/ha/year to
maintain current levels of carbon in the soil, while parts of Central and South America need
more than double the average required amount of carbon inputs (Figure 4).
Some of the highest carbon input needs were found in areas where high SOC stocks
are associated with high mineralization coefficients related to the mild moist conditions,
or with high SOC stocks and sandy soils. At the opposite end of the spectrum, other areas
exhibited low carbon input requirements because of low SOC stocks and moderate to low
mineralization levels. Therefore, the resulting global distribution of carbon inputs to main-
tain current levels of SOC stocks (Map 4) is strongly affected by the interaction of climate
and current soil conditions.
FIGURE 4
Regional averages of carbon inputs needed to maintain current levels of carbon in the soil
in unimproved and improved grasslands
7
Carbon inputs (tonnes/ha/year)
0
Sub-Saharan Africa
East Asia
Western Europe
Oceania
South Asia
Eastern Europe
North America
Russian Federation
Sub-Saharan Africa
South Asia
East Asia
Western Europe
Oceania
Eastern Europe
North America
Russian Federation
Improved Unimproved
MAP 4
Global carbon input levels (tonnes C/ha/year) needed to maintain current SOC stocks under improved and unimproved grasslands
3. Results
21
In our framework, the estimated carbon inputs are those needed to maintain cur-
rent SOC stocks at steady state. We compared the total carbon input (CRes + CExc) to
the estimated carbon inputs and the difference was termed the carbon balance (Cbal).
The current carbon balance may be used to assess if current SOC stocks are increasing
or decreasing. The majority of grassland soils seem to receive enough organic material
to maintain current carbon stock levels. On average, Cbal in current grassland systems
is 1.1 tonnes C/ha/year, for both improved and unimproved systems, hence indicating
increasing SOC stocks.
At regional level, unimproved grasslands in Eastern Europe and the Russian Federation
show the highest positive balances with values reaching 1.7 and 1.5 tonnes C/ha/year,
respectively, while the lowest positive carbon balance (0.6 tonnes C/ha/year) was found in
East Asia (Figure 5). Improved systems in the Russian Federation seem to follow the same
pattern of the unimproved grasslands, with the highest Cbal of all regions (1.5 t C/ha/yr).
On the other hand, we found that improved systems in East Asia are close to equilibrium
conditions (0.1 tonnes C/ha/year), followed by sub-Saharan Africa (0.6 tonnes C/ha/year)
and Central and South America (0.8 tonnes C/ha/year) (Figure 5).
FIGURE 5
Regional carbon balance (tonnes C/ha/year) for unimproved
and improved grassland systems
1,8
Carbon balance (tonnes C/ha/year)
1,6
1,4
1,2
1,0
0,8
0,6
0,4
0,2
0,0
Russian Federation
North America
Eastern Europe
Western Europe
Oceania
South Asia
Sub-Saharan Africa
East Asia
Eastern Europe
Russian Federation
Western Europe
North America
South Asia
Oceania
Sub-Saharan Africa
East Asia
Improved Unimproved
3. Results 23
However, it is important to note that for both grassland systems, several countries have
a negative Cbal. Analysis indicated that available carbon inputs to the soil were lower than
estimated carbon inputs needed to preserve current SOC stocks, and consequently were
not sufficient to maintain stocks at steady state. In improved systems, the highest negative
Cbal was found in Indonesia (-6.7 tonnes C/ha/year), the Philippines (-5.1 tonnes C/ha/year),
Colombia (-4.5 tonnes C/ha/year), Malaysia (-3.9 tonnes C/ha/year) and Uruguay (-3.3 tonnes
C/ha/year), meaning that current SOC stocks are likely to be decreasing due to anthropogenic
stresses combined with climatic conditions (Map 5). Negative Cbal values were also found in
unimproved systems in Colombia (-6.2 tonnes C/ha/year), Indonesia (-5.3 tonnes C/ha/year)
and Mexico (-0.9 tonnes C/ha/year), among others (Map 6).
24
MAP 5
Carbon balance (tonnes C/ha/year) for improved grassland systems
Note: negative values indicate that current stocks cannot be maintained. Carbon balances were calculated under current climatic conditions.
Global assessment of soil carbon in grasslands – From current stock estimates to sequestration potential
MAP 6
Carbon balance (tonnes C/ha/year) for unimproved grassland systems
3. Results
Note: negative values indicate that current stocks cannot be maintained. Carbon balances were calculated under current climatic conditions.
25
3.3 SOIL ORGANIC CARBON SEQUESTRATION POTENTIAL

Estimates of global soil carbon stocks, trends and sequestration potential are now central
to important topics, such as food security and climate change. In this context, regenerative
forms of grazing can provide carbon removal from the atmosphere and storage in the soil.
Well-adapted grazing systems – with improved pasture and optimized grazing regimes –
have the potential to increase SOC on degraded grassland, or on grassland soils that have
not reached their full carbon sequestration potential.
The Russian Federation showed the highest potential for carbon storage with an average
SOC stock of 191 tonnes C/ha after 20 years of implementation of best management prac-
tices (Figure 6). However, soils of the Russian Federation are already fairly carbon-dense on
average (186 tonnes C/ha) and these soils have likely reached their full sequestration poten-
tial, so even the best management practices will not provide a further accumulation of SOC.
The annual increment (on a per hectare basis) ranged from 0.18 to 0.41 tonnes C/ha across
the various regions (Figure 6). Sub-Saharan Africa and South Asia show the highest potential
for carbon storage on a per hectare basis (0.41 and 0.33 tonnes C/ha/year, respectively), fol-
lowed by Oceania, North America, and East Asia. Western Europe and Eastern Europe have
the lowest annual increments (0.20 tonnes C/ha and 0.18 tonnes C/ha respectively), which
result in a negligible sequestration potential (Figure 6). In general, areas of the southern hemi-
sphere with low carbon stocks show a large potential for soil carbon storage (Map 7).
FIGURE 6
Soil organic carbon (SOC) sequestration potential after 20 years
of application of best management practices for all available grassland soils
(i.e. those not excluded from the analysis as high SOC or sandy soils)
250 0.45
0.4
tonnes C/ha/year
200 0.35
tonnes C/ha
0.3
150
0.25
0.2
100
0.15
50 0.1
0.05
0 0
North America
Russian Federation
Western Europe
Eastern Europe
East Asia
Oceania
South Asia
Sub-Saharan Africa
potential SOC sequestered baseline SOC stocks SOC sequestration rate

Note: Results are given by regional averages per hectare (bars) and their annual increment (dots)).
Source: UN. 2020. Map of the World, modified with data from Sommer, R. & Bossio, D. 2014. Dynamics and climate change
mitigation potential of soil organic carbon sequestration. Journal of Environmental Management, 144: 83–87.
https://doi.org/10.1016/j.jenvman.2014.05.017; Hengl, T., de Jesus, J.M., MacMillan, R.A., Batjes, N.H., Heuvelink, G.B.M., Ribeiro, E.,
Samuel-Rosa, A., Kempen, B., Leenaars, J.G.B., Walsh, M.G. & Ruiperez Gonzalez, M. 2014. SoilGrids1km — Global Soil Information
Based on Automated Mapping. PLOS ONE, 9(8): e105992. https://doi.org/10.1371/journal.pone.0105992
MAP 7
Annual increase in soil organic carbon (SOC) in the top 30 cm on all available grassland soils globally
3. Results
(i.e. those not excluded from the analysis as high SOC or sandy soils)
Note: maps were produced based upon a geospatial analysis of datasets from the SoilsGrids250 database.
modified with data from Sommer and Bossio, 2014 and Hengl et al., 2014.
27
29
4. Discussion
4.1 BASELINE SOIL ORGANIC CARBON

In the year 2010 there was an estimated global annual uptake of 63.5 Mt of carbon in
grassland soils to a depth of 30 cm, with unimproved systems storing slightly higher amounts
than improved systems (33.8 vs 29.8 Mt C). Such a difference, even if small, is an expected
result as unimproved systems are less managed compared to improved systems, with the
latter more likely to lose carbon under human-induced activities. On average, the present
study shows that in the year 2010 the SOC in global grasslands was about 51 tonnes C/ha
to a soil depth of 30 cm (i.e. 53 tonnes C/ha in unimproved and 50 tonnes C/ha in improved
systems). Grassland SOC stocks may be as low as 25 tonnes C/ha for herbaceous grasslands
in deserts, and as high as 160 tonnes C/ha for evergreen shrub grasslands in boreal regions
(Petri et al., 2010; Lorenz and Lal, 2018). These results are comparable with our study, where
we found SOC stocks of 25 tonnes C/ha, or lower, in regions with arid climate (Beck et al.,
2018) and higher SOC stocks, above 80 tonnes C/ha, in cold climates (Beck et al., 2018).
An early study from Sombroek and colleagues reported estimated SOC stocks of 124 tonnes
C/ha to a 1 m soil depth (Sombroek, Nachtergaele and Hebel, 1993) in grasslands, while
the Special Report on Land Use, Land-Use Change, and Forestry reported a soil carbon
stock in temperate grasslands of about 236 tonnes C/ha to a 1 m soil depth (IPCC, 2019).
A more recent meta-analysis by Dlamini, Chivenge and Chaplot (2016) estimated SOC
stocks to 30 cm depth between 1 and 400 tonnes C/ha, with an average of 50 tonnes C/ha.
Our study can be directly compared only with the latter study, which reported estimates to
a soil depth of 30 cm. Despite the large SOC stock range, the average SOC stock values are
comparable (i.e. 51 tonnes C/ha in our study vs 50 tonnes C/ha in Dlamini, Chivenge and
Chaplot [2016]). Indeed, SOC stocks vary greatly among climatic regions and soil types, as
reported in the few global studies focused on quantifying soil carbon in grasslands.
The results arising from the present study provide additional information on the SOC
spatial distribution for both grassland systems. The greatest SOC stocks were found in
temperate regions due to lower average temperatures compared to humid, subhumid,
tropical, and semiarid regions resulting in lower decomposition rates, hence in the accu-
mulation of SOC. Greater SOC stocks under wet climates could also be attributed to the
high productivity of grasslands in wet environments. In comparison, the lowest SOC stocks
were observed in arid to semiarid grassland soils due to the low rainfall amounts resulting
in low biomass production and organic matter decomposition, hence a reduction of carbon
inputs into the soil. Indeed, stabilization of SOC also depends on several soil properties
such as soil pH, which contributes to regulating soil nutrient bioavailability, organic matter
turnover and an array of soil processes (Kemmitt et al., 2006), and soil clay and silt particles,
which protect soil organic matter by stabilizing it against microbial mineralization (Six et
al., 2002). Climatic conditions explained much of the variability of SOC stock in grassland
©Liz99 on Unsplash
soils to 30 cm depth, followed by the carbon input to the soil. In grassland systems, carbon
inputs to the soil are associated with grass biomass production, grazing intensities and
ruminant stocking density. In our framework, this parameter was derived from the GLEAM
model, a tool that enabled comprehensive, disaggregated and consistent analysis of the
environmental performance of global livestock production systems. The GLEAM model uses
a herd model coupled with an IPCC (2006) Tier 2 approach to computing emissions, there-
by enabling key characteristics of the livestock populations (e.g. herd structures, animal
performance, rations and manure management) to be captured in the calculations. Further,
GLEAM adopts a life-cycle approach and calculates the emissions arising along the supply
chain from cradle to retail point. Finally, the reliance on geographical information systems
(GIS) provides spatially explicit analysis and flexibility in combining datasets and aggregating
results (MacLeod et al., 2018). By ‘soft-coupling’ (i.e. a link between two individual mod-
els where a result of one model is integrated as an input parameter to the other model)
GLEAM and RothC, it was possible to include detailed information about the contribution
of livestock systems (e.g. N deposited, animal intake and distribution) to SOC stocks and to
estimate the first spatially explicit baseline scenario for the year 2010. The results presented
4. Discussion 31
here are meant to provide an estimate of the SOC levels in grassland systems in 2010; such
estimates are a starting point for further analysis, and in particular for designing and test-
ing management practices that could be beneficial to mitigating climate change without
compromising food security.
It is well known that among land-based GHG removal technologies, soil carbon
sequestration practices play a role in delivering agroecosystem resilience, climate change
adaptability, food security and improving nutrition. However, the effect of such practices
should be analysed at local scale because local socio-economic constraints, legislation and
environmental factors should be considered when designing interventions intended to
mitigate climate change. Therefore, these results are intended to act as a baseline when
identifying locations where interventions should be a priority (due to high soil degrada-
tion) and provide a baseline when quantifying the effect of such practices on soil carbon
sequestration. Two case studies are presented in Box 1 and Box 2 to highlight the applica-
bility of this framework at local level: to quantify changes in SOC stocks 30 years after the
establishment of fodder gardens in East Africa (Box 1), and to assess the effect of pasture
intensification in Paraguay (Box 2).
BOX 1
Assessing the effect of changing management practices on SOC
case study Eastern Africa
The establishment of fodder gardens is a practice that results in high productivity

under repetitive cutting, palatable fodder, and high protein content (20−25 percent).
This practice provides firewood, is easy to establish, and also has the potential to
improve soil quality through N fixation by legumes.
One tree species used in agroforestry systems that had remarkable success in
conserving soil, nutrient cycling, and nutrient retention is calliandra (Calliandra
calothyrsus). Calliandra, indigenous to Central America, is a small tree that reaches
about 10 m in height, has a deep root system, and is an aggressive pioneer species,
often found in disturbed areas such as roadsides, riverbanks and shifting cultivation
plots (Palmer, Macqueen and Gutteridge, 1994). Calliandra grows naturally in moist,
tropical regions up to an altitude of 1 500 m (Paterson, 1994), with annual rainfall
between 700 and 3 000 mm, and annual temperatures ranging between 22 and 28 °C.
Because of the limited size of the farms, research focused on integrating the trees
into existing cropping systems rather than on planting them in monoculture fodder
banks. Farmers preferred planting trees in hedges around the farm compound, and
in hedges along contour bunds.
In African countries, more than 40 000 smallholders (Kenya and Uganda) have
established fodder gardens with calliandra as a practice to raise milk production,
improve cow health and shorten the calving interval.
(Cont.)
Fodder trees require little or no cash investment or land taken away from producing
food or other crops. The only inputs required are seeds and minimal amounts of labour.
Moreover, this practice could provide other services, such as the provision of natural
fencing, and erosion control (Kabirizi, Mpairwe and Mutetikka, 2004).
Despite the potential of this practice to provide several ecological and socio-
economic benefits, little is known about its potential to sequester carbon in the soil.
For this purpose, the RothC and GLEAM models were soft coupled to study the effect
on SOC of the establishment of fodder gardens in mixed systems. GLEAM (v2.0) was
run to estimate the N deposited under a business as usual (baseline) scenario, and a
second run was performed to estimate the N deposited after the establishment of
fodder gardens (intervention) in Eastern African countries (Ethiopia, Kenya, Uganda
and the United Republic of Tanzania). Following the literature, 1 kg DM of calliandra
was added to the diets of adult females producing milk in dairy cattle systems. The
N deposited estimates for both the baseline and intervention scenarios were used as
input to the RothC model to estimate the change in soil carbon 30 years after the
establishment of fodder gardens (i.e. the RothC model was run for 30 years under
baseline and intervention conditions). Under the two scenarios, all model inputs (i.e.
weather, soil pH, soil bulk density and soil texture) were kept constant, except for the
input from animal excreta. For both scenarios, N deposited was converted to carbon
by applying a C:N ratio of 17.5 (FAO, 2015a) and used as input to the model runs.
Changes in SOC after 30 years of the application of fodder gardens were
calculated as follows:
∆C = SOCINT - SOCBAU
Where ∆C is SOC change, SOCINT is the SOC under the intervention scenario and SOCBAU
is the SOC under the baseline scenario.
The estimated SOC stocks under BAU conditions in Eastern African grasslands range
from 5.3 to 93.3 tonnes C/ha, with mean values of 40.9 tonnes C/ha. The mean SOC
stock is in line with data presented by Tessema et al. (2020) in their review of SOC stocks
and changes in grasslands in Eastern African countries. This metanalysis of local direct
measurements of SOC reported a mean initial SOC stock of 43.8 tonnes C/ha, and a
sequestration potential under grassland, following different management interventions,
of 1.8 tonnes C/ha per year.
The RothC model estimated a potential increase in soil carbon – after the
establishment of fodder gardens in mixed systems – of 0.9 tonnes C/ha, which leads to
about 0.03 tonnes C/ha per year. This result is lower than the SOC potential reported
by Tessema et al. (2020). It is, however, important to highlight that the two studies
are not directly comparable due to the background information and assumptions
used to determine the SOC potential.
(Cont.)
4. Discussion 33
More specifically, in the publications by Tessema et al. (2020), eight different

management practices were examined: enclosure, improved management (using
rotations and adding different inputs such as manure, fertilizer, etc.), free grazing, light/
heavy grazing, fencing, restoration measures, and conversion from natural forest to
grazing. The intervention simulated by the RothC model is based on the establishment
of fodder gardens with calliandra as a practice to increase milk production. The main
changing factor applied to the modelling of soil carbon is the change in animal
excreta, derived from a partial change in animal feed, with an application rate of 50
percent. The difference in carbon input between baseline and intervention is quite
low (data not shown), however, even a small change in carbon input to the soil from
animal excreta has the potential to increase carbon sequestration over 30 years. Other
interventions and improved management practices, such as changing grazing intensity,
could provide an even greater increase in SOC. It should be noted that calliandra could
become invasive and cause ecological damage outside Central America; therefore,
more studies are needed in the area to quantify the long-term impacts of such practices
on soil health, as well as production and other ecosystem services.
Potential increase in soil carbon after the establishment of fodder gardens

in mixed systems in Eastern African countries
Sudan Yemen
Eritrea
Djibouti
Abyei Somalia
South Sudan
Ethiopia
Democratic
Republic of the Congo
Uganda Kenya
Rwanda
Burundi
United Republic of Tanzania
Zambia
Malawi
Mozambique
Malawi
Notes: Dashed lines on maps represent approximate border lines for which there may not yet be full agreement.
Final boundary between the Republic of Sudan and the Republic of South Sudan has not yet been determined.
Final status of the Abyei area is not yet determined.
Source: United Nations Geospatial. 2020. Map of the World. United Nations. Cited 22 August 2022. www.un.org/
geospatial/file/2285/download?token=puayKYRA modified with data from Coleman and Jenkinson, 1996.
BOX 2
Assessing the effect of pasture intensification on SOC
case study Paraguay
Over the last decades, despite heterogeneities at local scales, there has been overall a
global trend of grazing systems intensification in response to increasing demand for
livestock products and land competition. In Paraguay, increasing animal productivity
is on the country’s agenda, but environmental trade-offs should also be considered
to define tailored interventions which will not compromise soil health and other eco-
system services. Therefore, it is crucial to explore the effect of pasture intensification
on soil carbon dynamics. For this purpose, the RothC and GLEAM models were soft
coupled to study the effect of pasture intensification on beef systems. GLEAM (v2.0)
was run to estimate the N deposited under a business as usual (baseline) scenario; a
second run was performed to estimate the N deposited after the intensification of
grassland systems (intervention) in Paraguay. The following changes were made to
the GLEAM inputs compared to the baseline scenario: feed intake was increased by
6.5 percent, biomass was increased by 10 percent and 90 kg/ha per year of synthetic
N fertilizer was applied to the soil (N fertilizer was not applied in the baseline scenar-
io). GLEAM estimates of N deposited for both baseline (BAU) and intervention (INT)
scenarios were converted to carbon by applying a C:N ratio of 17.5 (FAO, 2015a) and
then used as input to the RothC model to estimate the change in soil carbon after 30
years of pasture intensification. To do so, the RothC model was run for 30 years for
both BAU and INT conditions. Between the two scenarios, all other model inputs (i.e.
weather, soil pH, soil bulk density and soil texture) were kept constant.
Changes in SOC after 30 years of pasture intensification were calculated as follows:
∆C = SOCINT – SOCBAU
Where ∆C is SOC change, SOCINT is the SOC under the intervention scenario and
SOCBAU is the SOC under the baseline scenario.
The RothC model estimated a mean potential SOC increase of 27 tonnes C/ha
after 30 years of pasture intensification in Paraguay, which corresponds to a yearly
accumulation of about 0.9 tonnes C/ha in the soil. However, changes in SOC differ
substantially across the country, with accumulation reaching 150 tonnes C/ha in the
northeast of the country. A substantial depletion in SOC is found in the north/north-
west, with a maximum loss of about 50 tonnes C/ha. Areas with positive change in SOC
experience an accumulation of carbon in the soil mainly because of the higher carbon
inputs to the soil from both plants and animals, compared to the baseline scenario.
In this case, synthetic N fertilizer does not inhibit SOC accumulation, a process that is
likely to occur in areas where animal density is low and not enough organic material is
returned to the soil from animal excreta.
(Cont.)
4. Discussion 35
Increasing SOC stocks under perennial grasses relies mainly on enhancing carbon
inputs from plant roots and residues. This can be achieved by managing plant
biomass removal from grazing or increasing forage production through improved
species, irrigation and fertilization, yielding increases in SOC stocks of as much as
10 percent (Conant et al., 2017). In our study, fertilization, yield and animal feeds
were increased compared to the baseline scenario. As a result, we estimated an
increase in SOC of 0.9 tonnes C/ha a year, which leads to an increase in SOC stocks of
about 7 percent. This increase is, however, only achievable in areas where initial soil
conditions, soil nitrogen and animal density are balanced, otherwise a depletion of
carbon will occur. Soil carbon is only one component of the carbon balance system.
Pasture intensification, achieved by increasing fertilization, yield and animal feeds,
has a large impact on methane emissions and other GHG fluxes. Therefore, for a full
system budget, it is imperative to include estimates of changes in methane emissions
following pasture intensification in order to understand the environmental impacts
of such an intervention on the full system.
Potential soil organic carbon (SOC) change (tonnes C/ha)

after 30 years of pasture intensification
SOC change (t C/ha)

Bolivia (Plurinational State of)
<7
7−0
0−20
20−35
35−60
>60
Paraguay
Brazil
Argentina
Source: United Nations Geospatial. 2020. Map of the World. United Nations. Cited 22 August 2022. www.un.org/
geospatial/file/2285/download?token=puayKYRA modified with data from Coleman and Jenkinson, 1996.
It should be noted that the framework presented here, as well as the SOC stock esti-
mates, follow the approach described in the LEAP guidelines for SOC assessment and the
GSOC-MRV Protocol, a FAO protocol for measurement, monitoring, reporting and verifi-
cation of SOC in agricultural landscapes (FAO, 2020b). However, several assumptions have
been used to generalize and simplify the methodology to be applied globally. Moreover,
datasets on soil, climate and carbon inputs (at both regional and global levels) carry levels
of uncertainty that should not be disregarded when applying this framework and the SOC
stock baseline estimates to analyze mitigation strategies. A detailed description of the
model uncertainties is given in section 4.5.
4.2 SOIL ORGANIC CARBON BALANCE

Different methods have been used to estimate carbon input to the soil for modelling
purposes. They range from inverse modelling and allocation functions to expert opinion.
The diversity of methods and contexts results in a great variety of estimates (Martin et al.,
2021). In our study we estimated carbon inputs in two ways: by IPCC Tier 2 methodology
using the GLEAM model to quantify the actual input to the soil; and by inverse modelling
to determine the carbon inputs needed to maintain current levels of SOC stocks. Overall,
estimated carbon inputs presented here were in line with other estimates found in the
literature using the RothC model, which ranged from 1.18 (Xu, Liu and Kiely, 2011) to
5.2 tonnes C/ha/year (Meersmans et al., 2013) for grasslands.
When estimated using an inverse modelling approach, the calculation of these carbon
input levels results from the interplay between observed SOC stocks, the SOC mineraliza-
tion rates and the quality of the incoming organic matter (i.e. plant residues and organic
amendments). Unimproved grasslands needed either less carbon input or fewer carbon
inputs than improved systems to maintain current SOC stocks. This might be explained by
the fact that unimproved grasslands are mainly located in high altitude regions with lower
soil organic matter mineralization rates due to lower temperatures as compared to other
warmer areas, and also by the quality of incoming plant material which has to be specified
in inverse modelling approaches. Martin et al. (2021) found the same result for French
grasslands using the same methodological approach used in this study.
The FAO Synthesis Report on the State of the World’s Land and Water Resources for
Food and Agriculture (FAO, 2021) reports that about 13 percent of grassland area is
degraded due to high anthropogenic pressure, which is driven by: agricultural expansion,
deforestation, fire extent and frequency, grazing density, population density, and ratio of
invasive/native species. Another 34 percent of the global grassland area has reduced bio-
physical status and it is defined as deteriorated. Soil organic carbon is a factor influencing
land deterioration, but other drivers, such as water erosion rate, wind erosion, water stress,
native species richness and above-ground biomass contribute to the overall biophysical
status of land. Interestingly 54 percent of the grassland area is reported to be in a stable
condition (FAO, 2021). Our global estimates on the state of carbon in grasslands show that
grasslands were, on average, close to equilibrium (1.09 tonnes C/ha/year for improved and
1.27 tonnes C/ha/year for unimproved grasslands). Positive Cbal values were found in areas
where the land is stable or even under improved biophysical conditions, such as grasslands
in the United States of America, Kazakhstan and Mongolia. These are areas under low
4. Discussion 37
human-induced pressure on land (FAO, 2021). The highest negative Cbal was found in East
Asia, Central and South America, and Africa south of the Equator, meaning that current
SOC stocks are likely to be decreasing due to anthropogenic stresses combined with climat-
ic conditions (Map 4). This is in accordance with the latest Synthesis Report on the State of
the World’s Land and Water Resources for Food and Agriculture (FAO, 2021), which identi-
fies these regions as severely affected by soil degradation (FAO, 2021). Most grasslands at
risk of human-induced land degradation are exposed to decreasing freshwater availability.
There are exceptions in Southern America and sub-Saharan Africa, where decreasing land
productivity and soil protection account for declining ecosystem services. In Asia, increasing
water stress contributes to grasslands at risk. In sub-Saharan Africa, grasslands are prone to
frequent and intense fire (FAO, 2021).
It is noteworthy that no specific global measurements are currently available, and that
the diversity of situations – in terms of climate, soils and management practices – might
have been crucial for soil carbon dynamics in these areas, as evidenced by the variability of
the carbon input values in grassland systems (Figure 3).
4.3 GRASSLAND POTENTIAL TO SEQUESTER SOIL CARBON

Increasing SOC in grassland areas globally is less likely than on croplands which are already
intensively managed (Smith et al., 2008). However, the adoption of improved management
practices on grasslands offers the opportunity to sequester significant amounts of carbon
in the near term, contributing to global mitigation efforts and restoring degraded lands.
The 4 per 1000 (4p1000) initiative has identified an aspirational sequestration target of
3.5 Gt C/year to provide substantive global mitigation. Our estimates suggest that about
17 percent of this target could be reached in the top 30 cm of grassland soils alone by
enhancing management through the incorporation of organic manures, some types of
agroforestry practices, or rotational grazing. However, it is important to note that we used
an empirical approach to determine the increase in percent-SOC in response to improved
management, but specific management practices have not been tested. We attempted to
do so for two case studies, by applying the RothC model in East Africa (Box 1), and Par-
aguay (Box 2), but more work is needed to understand the impact of individual practices
at global scale.
Our statistical estimate of soil carbon sequestration potential is one of the first attempts
at providing a detailed, spatially-disaggregated assessment at global level. Petri et al. (2010)
integrated demographic data with GIS to calculate potential per capita carbon sequestra-
tion and estimate the potential for land managers to engage in mitigation sequestration
schemes while using the land for their livelihoods. In their study, Petri et al. (2010) calculat-
ed the SOC potential after 20 years by applying sequestration factors for organic carbon as
a function of grassland typology, management status and climatic zones. Results are given
for climate and grassland types and the usage of different data sources makes it difficult to
compare against our estimates by continent. Globally, Petri et al. (2010) estimated a SOC
change of about 1.5 tonnes C/ha/year, which is somewhat higher than our conservative
estimates of 0.29 tonnes C/ha/year. This is mainly due to the high sequestration factors
used for estimating organic carbon sequestration after 20 years. In our study, we used a
global value of percent increase in SOC, which results in lower SOC potential compared to
the estimates reported by Petri et al. (2010). The spatial distribution of available grassland
soils is also different in the two studies. In calculating the SOC potential, highly organic
soils and sandy soils were excluded from the analysis, resulting in a reduced area stud-
ied and lower SOC potential compared to other published estimates (Petri et al., 2010).
The recently published Global SOC sequestration potential (GSOCseq) map (FAO, 2022)
reports a mean sequestration rate of 0.19 tonnes C/ha/year in grasslands under a sus-
tainable management scenario which implies a 20 percent increase in C inputs over a
20-year period. Our empirical approach is based on the assumption that the carbon
concentration in the soil would increase by 0.27 percent if management practices known
to improve SOC sequestration would be applied over a 20-year period worldwide.
Due to this preliminary assumption, our estimates are slightly higher than the GSOCseq
estimates for grasslands.
It should be noted that the estimated SOC sequestration potential, and its spatial distribu-
tion, is strictly dependent on the initial soil conditions (e.g. soil bulk density and texture), but
estimates do not account for differences in climate and important soil process issues, such as
carbon input and turnover rate. However, there is no published work specifically dedicated
to the prediction of soil carbon sequestration potential in grassland by process-based model-
ling. Indeed, the RothC model (Coleman and Jenkinson, 1996) has been previously used to
estimate the likely responses of soils to future climate and interactions with projected future
land-use changes (Gottschalk et al., 2012). A recent study conducted by Morais, Teixeira and
Domingos (2019) calculated global SOC dynamics for 80 specific land uses within broad land-
use classes (e.g. cropland, forest and grassland). Nevertheless, estimates of SOC sequestration
potential on grassland are still uncertain, and often included in vast assessments, hence dis-
entangling grassland impact on SOC sequestration appears challenging.
In this study, a simple statistical approach was applied to obtain a first estimate of
attainable SOC sequestration rates in grasslands, identifying regions with greater poten-
tial to sequester carbon after implementing management practices. It provides a general
framework for countries with limitations on implementing more complex, intensive data
requiring, process-oriented modelling approaches.
4.4 SOURCES OF UNCERTAINTIES REGARDING THE BASELINE SOIL

ORGANIC CARBON STOCKS
The results of the statistical analysis of all variables driving the RothC model show that the
modelled SOC stocks are correlated to climatic conditions (e.g. temperature and potential
evapotranspiration). This is an expected result as meteorological conditions directly affect
soil processes, such as mineralization rates, and indirectly affect the amount of organic
material (mainly plant residues) entering the soil. Moreover, modelled soil carbon is posi-
tively correlated to carbon inputs (Figure 7). This is also an expected result as inputs from
plant residues and manure are the main external sources of carbon entering the soil, which
will be stored in the soil depending on clay and climatic conditions.
In this study, analyses were performed to test the sensitivity effect of three main input
variables on the baseline SOC stocks. The input variables tested were initial SOC stocks, car-
bon inputs and air temperature (i.e. those with higher correlation with the modelled SOC
stocks). The relative change (percent) was calculated for four scenarios, as shown in Table 2.
4. Discussion 39
FIGURE 7
Correlation matrix of main variables used to drive the RothC model
SOC30
TEMP
SOC0
CLAY
PREC
Cin0
PET
1
PREC
0.75
TEMP
0.5
PET 0.25
Cin0 0
- 0.25
SOC30
- 0.5
CLAY
- 0.75
SOC0
-1
Note: (SOC0=initial SOC, SOC30=Soil Organic Carbon at the end of the 30 years model run (baseline SOC),
PET= potential evapotranspiration, TEMP=temperature, PREC=precipitation, Cin0= organic carbon input to the soil).
Source: UN. 2020. Map of the World, modified with data from Sommer, R. & Bossio, D. 2014. Dynamics and climate change mitigation
potential of soil organic carbon sequestration. Journal of Environmental Management, 144: 83–87. https://doi.org/10.1016/j.
jenvman.2014.05.017; Hengl, T., de Jesus, J.M., MacMillan, R.A., Batjes, N.H., Heuvelink, G.B.M., Ribeiro, E., Samuel-Rosa, A., Kempen,
B., Leenaars, J.G.B., Walsh, M.G. & Ruiperez Gonzalez, M. 2014. SoilGrids1km — Global Soil Information Based on Automated
Mapping. PLOS ONE, 9(8): e105992. https://doi.org/10.1371/journal.pone.0105992
TABLE 2
Sensitivity analysis of model results (SOC stocks) to changes on main variables used to drive the
RothC model
Variable Scenario Relative change
- 50% of initial carbon inputs -28.8%

- 20% of initial carbon inputs -11.5%
Carbon inputs
+ 20% of initial carbon inputs 14.4%
+ 50% of initial carbon inputs 28.8%
- 50% of initial SOC stocks -21.2%
- 20% of initial SOC stocks -8.5%
SOC stocks
+ 20% of initial SOC stocks 10.6%
+ 50% of initial SOC stocks 21.3%
Initial temperature +2 °C -18.1%
Initial temperature +1 °C -23.2%
Air temperature
Initial temperature -1 °C 34.9%
Initial temperature -2 °C 41.5%
The results showed that carbon input variability leads to ±30 percent change in the
baseline SOC stocks, while changes in initial SOC stocks had a relative change from the
baseline SOC stocks of about ±20 percent. Increasing air temperature will negatively impact
the baseline SOC stocks, with a relative change of up to -23.2 percent, while a decrease in
air temperature will have a relative change of up to 41.5 percent.
These results show that the major contributor to baseline SOC stock uncertainty is the
variability in carbon inputs. In this study, the plant residue inputs to the soil were estimated
from the dry matter yield estimates (Haberl et al., 2007), and animal excreta was derived
from the deposited nitrogen simulated by the GLEAM 2.0 model. Both datasets carry some
uncertainties; for instance, the above-ground dry matter is estimated from NPP, which in
turn can be estimated through various principles (e.g. light use efficiency, plant growth,
satellite information). To reduce uncertainties about soil carbon inputs, and therefore on
SOC estimates presented here and elsewhere, it is crucial to generate local datasets and
explore new and existing NPP datasets in order to improve the accuracy of plant residue
estimates. In addition to intrinsic methodological differences among different NPP prod-
ucts, differences in land use definition and distribution contribute to the large uncertainty
associated with carbon input estimates from plant residues.
Soil carbon inputs from animals have been estimated by applying a C:N ratio (from lit-
erature; see FAO, 2015a) to the N deposited. In this study, the C:N ratio was derived from
the literature as a global average value, and as such it could differ from regional figures.
However, it was observed that the sensitivity of the RothC model to animal excreta quality
is low (between 1.1 percent and 3 percent) (Jebari et al., 2020), so this variable does not
impact the SOC stocks modelled by RothC. The quantity of carbon entering the soil from
animal excreta impacts the SOC results. In this study, this variable was derived from GLEAM
estimates of deposited N, which follows a Tier 2 approach from IPCC (2006). In addition
to intrinsic methodological uncertainties linked to such an approach, which are extensive-
ly discussed in Opio et al. (2013), it is important to note that diversity in land maps and
definitions affects the quantification of N indicators (Kaltenegger et al., 2021). Moreover,
grassland definition and distribution affect all underlying input data and therefore exac-
erbate the uncertainty of the model results. In general, the ability to accurately estimate
carbon inputs to the soil is crucial in modelling soil processes and can be a major source
of uncertainty, as shown here and elsewhere (Hashimoto, Wattenbach and Smith, 2011;
Neumann et al., 2015; Martin et al., 2021).
Maps of soil properties are also known to carry a significant uncertainty due to the
limited freely available soil data needed to calibrate the statistical models used to derive
maps. In the statistical analysis of model sensitivity to input variables, the SOC stocks used
to initialize the model significantly affected modelling results. A comparison of three soil
datasets, namely SoilGrid (Hengl et al., 2014), HWSD (FAO, IIASA, ISRIC, ISS-CAS & JRC,
2012), which we used in this study, and the Northern Circumpolar Soil Carbon Database
(Hugelius et al., 2013) was conducted by Tifafi, Guenet and Hatte (2018) to quantify dif-
ferences in soil properties among datasets, and to evaluate them against soil data from
the United States of America, England, Wales and France. The results of this comparison
highlighted that global SOC stocks predicted by each product differ greatly, particularly for
boreal regions where differences can be related to large disparities in SOC concentration.
4. Discussion 41
Differences in other regions were mainly related to differences in soil bulk density estimates.
When comparing the three datasets versus ground truth data, a significant difference in
spatial patterns was found, with an underestimation in SOC stocks of more than 40 per-
cent compared to field data. The HWSD and SoilGrid maps were also compared globally
against the GSOCmap, the first global SOC map ever produced through a consultative and
participatory process involving member countries (FAO and ITPS, 2018). A larger agreement
between the GSOCmap and the HWSD than between the GSOCmap and the SoilGrids was
found. Positive and negative changes from the GSOCmap and the HWSD were irregularly
distributed, but the changes from the GSOCmap to the SoilGrids tended to be positive,
suggesting a major carbon pool predicted by the latter product (FAO and ITPS, 2018). The
estimation of the global soil carbon stock is still quite uncertain, and improved geostatistical
methods are urgently needed to reduce the propagation of such uncertainties on soil mod-
els. Moreover, soil and land-use distribution datasets are not linked and are often produced
on different timescales. This can lead to the allocation of initial SOC stocks to a land-use
that does not reflect the current condition.
The uncertainty regarding initial SOC stocks, and their distribution and allocation to
different land uses, together with model uncertainties, should be carefully taken into con-
sideration when using the results of this work on the current state of carbon in the soil and
its potential to be sequestered in grassland systems.
43
5. Conclusions and way forward
Soils contribute to the achievement of the UN SDGs through carbon sequestration.

By enhancing soil health and fertility, soils can play a crucial role in climate action (target
13.2), land degradation neutrality (target 15.3), and alleviating hunger (targets 2.1 and
2.4). Despite the undeniable technical potential to sequester carbon in soils, there are
often significant limitations to achieving that potential in any particular location and within
specific production systems.
The present study provides a spatially explicit report on the state of grassland soils, with
estimates of the SOC stocks for the year 2010. On average, in the year 2010, the SOC in
global grasslands was about 51 tonnes C/ha to a soil depth of 30 cm, with minor differ-
ences between improved and unimproved systems (53 vs 50 tonnes C/ha). The SOC stocks
presented in this report can be used as a baseline for future work to explore the impacts
of livestock management on soil carbon at country and farm levels. However, there is still a
strong necessity for additional data on current soil conditions, especially from underrepre-
sented regions. The approach used to develop the SOC baseline for grasslands follows the
methodology recommended by the Global Soil Partnership (GSP) for measurement, moni-
toring, reporting and verification of SOC in agricultural landscapes (FAO, 2020b). Examples
of practical application of our SOC stock estimates at national level, including specific
interventions, have been given in Box 1 and Box 2. A recent review of 184 countries’ initial
NDCs, found that only twenty-eight countries referred to SOC in their NDCs (Wiese et al.,
2021). Countries’ reasons for not including SOC in NDCs included the need to prioritize
goals of sustainable development and food security above climate mitigation, a lack of
incentives for farmers to improve management practices, and the difficulty of accurately
monitoring changes in SOC. The results of this report could therefore support the inclusion
of SOC targets in NDCs, which will improve NDCs’ comprehensiveness and transparency
for tracking and comparing policy progress across NDCs.
The majority of grassland soils seem to receive enough organic material to maintain
current carbon stock levels. However, improved grasslands needed, on average, higher
carbon inputs than unimproved systems to sustain current SOC stocks (2.1 vs 1.3 tonnes
C/ha/year). The positive soil carbon balance found in both improved and unimproved sys-
tems globally indicates a potential increase in SOC stocks. Despite such a positive trend
globally, the large spatial variability of these estimates highlights that the soil state at coun-
try level could differ greatly from the global estimates. Fifty-four percent of the grassland
area is reported to be in a stable condition (FAO, 2021); our global estimates support this
finding, showing positive carbon balance in areas where the land is stable or even under
improved biophysical conditions. On the other hand, negative carbon balance was found
in East Asia, Central and South America, and Africa south of the Equator, meaning that
current SOC stocks are likely to be decreasing due to anthropogenic stresses combined
with climatic conditions. The findings of this analysis show that there is room for additional
©Gabriel Jimenez on Unsplash
carbon storage in some grassland soils. The main recommendations for grassland systems
are to prioritize carbon returns in deteriorated soils that have a negative carbon balance,
and to protect SOC in areas – particularly under unimproved grasslands – with high carbon
stocks. Grasslands could contribute to the recarbonization of degraded land and the results
of the present study can highlight hotspots where interventions on grasslands are needed
to preserve or increase SOC in the long term.
The empirical approach used in this study made it possible to estimate the soil carbon
sequestration potential of available grasslands following the application of management
practices known to improve SOC sequestration or protection. Grasslands could sequester
0.3 tonnes C/ha/year in the 0−30 cm depth layer, which could be an important contribution
to global mitigation efforts. The adoption of improved management practices offers the
opportunity to sequester significant amounts of carbon in the near term, and potentially
to make an important contribution to global mitigation efforts. The 4p1000 Initiative has
identified an aspirational sequestration target of 3.5 Pg C/year to provide substantive global
mitigation. Our estimates suggest that 17 percent of this target could be reached in the top
30 cm of grasslands and continue over at least 20 years after adoption of SOC enhancing
5. Conclusions and way forward 45
management, such as the incorporation of animal manures, agroforestry and rotational

grazing. This requires that grasslands increase SOC storage between 0.18−0.41 tonnes C/ha
every year. Our estimates do not account for differences in climate and important soil
process issues, notably nutrient and water limitations, biomass production and turnover
rates. However, sequestering carbon via increases in the soil component on grasslands is
an achievable and potentially effective route to quickly increasing CO2 sequestration in the
near term. Despite the large technical potential to sequester carbon in soils, there are often
significant limitations to achieving that potential in any particular place and within specific
farming systems. In addition, there may be trade-offs with productivity, food security or
hydrologic balances, as well as concerns regarding other GHGs, such as N2O.
A full system analysis requires estimates of GHG emissions alongside SOC estimates, or
else misleading messages could arise from the investigation of independent components of
the system balance. The methodology presented here made it possible to assess SOC stocks
by soft-coupling the RothC soil carbon model with the GLEAM LCA model. Specifically,
the GLEAM model provided inputs which have been used to drive the soil model. Future
work should target including soil carbon estimates in full LCA studies. The main challenges
would be to develop a methodology to allocate SOC stocks to different livestock units and
to account for temporal dynamics of carbon in the soil. Nevertheless, the addition of a
soil carbon compartment in the GLEAM model will enable accurate life cycle assessment
in livestock systems as well as the design of targeted national policies for climate change
mitigation and food security through the livestock sector.
Potential users should take into consideration the uncertainties deriving from the under-
lying datasets and the limitations of the approaches used to produce the SOC estimates.
In this context, future technical work should focus on testing the impacts of different data
sources on the SOC estimates. In particular, newly published soil carbon maps, such as the
global SOC map (FAO and ITPS, 2018), and spatial data infrastructure such as the Global
Soil Information System (GLOSIS), could be used to refine global SOC stocks estimates.
GLEAM’s forthcoming update on livestock LCA analysis for the year 2015 will also be a
useful tool to acquire newer information on organic inputs to the soil, which could be used
as a driver for the soil model. In general, the use of alternative sources of all major inputs
used to estimate SOC stocks would be required to quantify model uncertainties and to
observe SOC stock changes over time.
The results arising from the present work provide a global overview of the state of soil
carbon in grasslands, and their potential to sequester carbon in the soil. The SOC stocks
and changes have been analysed to a soil depth of 30 cm. However, SOC can also be
stored at deeper soil layers and future work should aim to both refine current estimates as
well as develop new approaches to predict SOC in deep soil layers. Moreover, there is still
a significant need for additional data, especially from underrepresented regions (Merbold
et al., 2021), on current soil conditions, and on the effects of management practices on
SOC stocks and GHG emissions. Involving local experts and institutions is a fundamental
step for modelling improvements as well as knowledge exchange. In this context, the GSP
partnership has recently published the first ever country-driven global SOC sequestration
potential (GSOCseq) map (FAO, 2022). The GSOCseq map was developed based on the
submissions of national experts appointed by FAO Member Nations. A bottom-up approach
was used to establish a reliable, transparent and cost-effective mechanism to monitor,

report and verify changes in SOC stocks in agricultural areas. The methodology used is
based on the RothC model and it is constantly extended, improved and updated to better
characterize local SOC dynamics. The FAO LEAP and GSP partnerships are actively collab-
orating to improve the GSOCseq map and methodology in order to better characterize
soil carbon dynamics in grasslands by refining the definition of grassland (e.g. to include
improved and unimproved grasslands) and the sustainable soil management scenarios cur-
rently used in the GSOCseq map.
The work presented here represents a first step to quantify the state of grassland soils,
as well as to identify and prioritize areas with potential to increase SOC stocks through SOC
sequestration in order to enable accurate life cycle assessment of livestock systems as well
as the development of targeted livestock sector-driven national policies for climate change
mitigation and food security.
47
References
Abdalla, M., Hastings, A., Chadwick, D.R., Jones, D.L., Evans, C.D., Jones, M.B., Rees, R.M.
& Smith, P. 2018. Critical review of the impacts of grazing intensity on soil organic carbon
storage and other soil quality indicators in extensively managed grasslands. Agriculture,
Ecosystems & Environment, 253: 62–81. Cited 21 October 2021. https://doi.org/10.1016/j.
agee.2017.10.023
Aalde, H., Gonzalez, P., Gytarsky, M., Krug, T., Kurz W.A., Lasco, R.D., Martino, D.L.,
McConkey, B.G., Ogle, S., Paustian, K., Raison, J., Ravindranath, N.H., Schoene, D.,
Smith, P., Somogyi, Z., van Amstel, A., and Verchot, L. 2006. Generic Methodologies
Applicable to Multiple Land-Use Categories. Volume 4, Chapter 2. In: Eggleston H.S.,
Buendia L., Miwa K., Ngara T. and Tanabe K. 2006 IPCC Guidelines for National Greenhouse
Gas Inventories, Prepared by the National Greenhouse Gas Inventories Programme, (eds).
Published: IGES, Japan. Cited 21 July 2021. www.ipcc-nggip.iges.or.jp/public/2006gl/pdf/4_
Volume4/V4_02_Ch2_Generic.pdf
Beck, H., Zimmermann, N., McVicar, T., Vergopolan, N., Berg, A. & Wood E. 2018. Present
and future Köppen-Geiger climate classification maps at 1-km resolution. Scientific Data, 5:
180214. Cited 11 May 2022. https://doi.org/10.1038/sdata.2018.214
Boden, T.A., Marland, G. & Andres, R.J. 2017. Global, Regional, and National Fossil-Fuel CO2
Emissions. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S.
Department of Energy, Oak Ridge, Tenn., U.S.A. Cited 21 July 2021. https://doi.org/10.3334/
CDIAC/00001_V2017
Chang, J., Ciais, P., Viovy, N., Vuichard, N., Sultan, B. & Sousanna J. F. 2015. The greenhouse
gas balance of European grasslands. Global Change Biology, 21(10): 3748–3761. Cited 21
July 2021. https://doi.org/10.1111/gcb.12998
Coleman, K. & Jenkinson, D.S. 1996. RothC-26.3 - A Model for the turnover of carbon in
soil. In: Powlson, D.S., Smith, P., Smith, J.U., eds. Evaluation of Soil Organic Matter Models.
NATO ASI Series, 38: 237-246. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-
642-61094-3_17
Conant, R.T., Cerri, C.E.P., Osborne, B.B. & Paustian, K. 2017. Grassland management
impacts on soil carbon stocks: a new synthesis. Ecological Applications, 27(2): 662–668. Cited
21 July 2021. https://doi.org/10.1002/eap.1473
Davidson, E.A. & Janssens, I.A. 2006. Temperature sensitivity of soil carbon decomposition
and feedbacks to climate change. Nature, 440(7081): 165–173. Cited 21 July 2021.
https://doi.org/10.1038/nature04514
Diels, J., Vanlauwe, B., Van der Meersch, M.K., Sanginga, N. & Merckx, R. 2004. Long-
term soil organic carbon dynamics in a subhumid tropical climate: 13C data in mixed C3/
C4 cropping and modeling with ROTHC. Soil Biology and Biochemistry, 36(11): 1739–1750.
Cited 21 October 2021. https://doi.org/10.1016/j.soilbio.2004.04.031
Dlamini, P., Chivenge, P. & Chaplot, V. 2016. Overgrazing decreases soil organic carbon stocks
the most under dry climates and low soil pH: A meta-analysis shows. Agriculture, Ecosystem &
Environment, 221: 258–269. Cited 21 July 2021. https://doi.org/10.1016/j.agee.2016.01.026
Erb, K.H., Kastner, T., Plutzar, C., Bais, A.L.S., Carvalhais, N., Fetzel, T., Gingrich, S.,
Haberl, H., Lauk, C., Niedertscheider, M., Pongratz, J., Thurner M. & Luyssaert, S. 2018.
Unexpectedly large impact of forest management and grazing on global vegetation biomass.
Nature, 553: 73–76. Cited 21 July 2021. https://doi.org/10.1038/nature25138
ESA. 2017. Land Cover CCI Product User Guide Version 2. Tech. Rep. Cited 21 July, 2021
https://maps.elie.ucl.ac.be/CCI/viewer/download/ESACCI-LC-Ph2-PUGv2_2.0.pdf
FAO. 2014. Global Land Cover Share (GLC-SHARE) Beta-Release 1.0 Database. Rome, FAO.
www.fao.org/uploads/media/glc-share-doc.pdf
FAO. 2015a. Farmer’s compost handbook. Experiences in Latin America. Santiago, FAO.
www.fao.org/3/I3388E/i3388e.pdf
FAO. 2015b. Source of Administrative boundaries: The Global Administrative Unit Layers (GAUL)
dataset. Implemented by FAO within the CountrySTAT and Agricultural Market Information
System (AMIS) projects. Rome. Cited 21 July 2021. https://data.apps.fao.org/map/catalog/srv/
eng/catalog.search#/metadata/9c35ba10-5649-41c8-bdfc-eb78e9e65654
FAO. 2019. Measuring and modelling soil carbon stocks and stock changes in livestock production
systems: Guidelines for assessment (Version 1). Livestock Environmental Assessment and
Performance (LEAP) Partnership. Rome, FAO. www.fao.org/3/ca2934en/CA2934EN.pdf
FAO. 2020a. The Global Livestock Environmental Assessment Model–GLEAM. Rome, FAO. Cited
21 July 2021. www.fao.org/gleam/en/
FAO. 2020b. A protocol for measurement, monitoring, reporting and verification of soil organic
carbon in agricultural landscapes – GSOC-MRV Protocol. Rome, FAO. https://doi.org/10.4060/
cb0509en
FAO. 2021. The state of the world’s land and water resources for food and agriculture
– Systems at breaking point. Synthesis report 2021. Rome, FAO. Cited 27 July 2022.
https://doi.org/10.4060/cb7654en
FAO. 2022. Global Soil Organic Carbon Sequestration Potential Map – SOCseq v.1.1. Technical
report. Rome. https://doi.org/10.4060/cb9002en
FAO, IIASA, ISRIC, ISS-CAS & JRC. 2012. Harmonized World Soil Database. Rome, FAO. 43 pp.
www.fao.org/3/aq361e/aq361e.pdf)
FAO & ITPS. 2018. Global Soil Organic Carbon Map (GSOCmap). Technical Report. Rome, FAO.
www.fao.org/3/I8891EN/i8891en.pdf
FAOSTAT. 2016. FAOSTAT Database. Rome, FAO. Cited 27 July 2022. www.fao.org/faostat/
en/#home
Falloon, P., Smith, P., Coleman, K. & Marshall, S. 1998. Estimating the size of the inert
organic matter pool from total soil organic carbon content for use in the Rothamsted
carbon model. Soil Biology and Biochemistry, 30(8-9): 1207–1211. Cited 21 July 2021.
https://doi.org/10.1016/S0038-0717(97)00256-3
Garnett, T., Godde, C., Muller, A., Röös, E., Smith, P., de Boer, I.J.M., zu Ermgassen,
E., Herrero, M., van Middelaar, C., Schader, C. and van Zanten, H. 2017. Grazed and
Confused? Ruminating on cattle, grazing systems, methane, nitrous oxide, the soil carbon
sequestration question – and what it all means for greenhouse gas emissions. FCRN, University
References 49
of Oxford, England. Cited 5 May 2022. www.oxfordmartin.ox.ac.uk/downloads/reports/fcrn_

gnc_report.pdf
Gottschalk, P., Smith, J.U., Wattenbach, M., Bellarby, J., Stehfest, E., Arnell, N., Osborn,
T.J., Jones, C. & Smith, P. 2012. How will organic carbon stocks in mineral soils evolve
under future climate? Global projections using RothC for a range of climate change scenarios.
Biogeosciences, 9(8): 3151–3171. Cited 21 July 2021. https://doi.org/10.5194/bg-9-3151-2012
Grigal, D.F. & Berguson, W.E. 1998. Soil carbon changes associated with short-rotation
systems. Biomass and Bioenergy, 14(4): 371–377. Cited 21 July 2021. https://doi.org/10.1016/
S0961-9534(97)10073-3
Haberl, H., Erb, K.H., Krausmann, F., Gaube, V., Bondeau, A., Plutzar, C., Gingrich, S., Lucht,
W. & Fischer-Kowalski, M., 2007. Quantifying and mapping the human appropriation of net
primary production in earth’s terrestrial ecosystems. Proceedings of the National Academy of
Sciences, 104(31): 12942–12947. Cited 21 July 2021. https://doi.org/10.1073/pnas.0704243104
Hashimoto, S., Wattenbach, M., & Smith, P. 2011. Litter carbon inputs to the mineral
soil of Japanese Brown forest soils: comparing estimates from the RothC model with
estimates from MODIS. Journal of Forest Research, 16(1): 16–25. Cited 21 September 2021.
https://doi.org/10.1007/s10310-010-0209-6
Henderson, B., Gerber, P.J., Hilinski, T.E., Falcucci, A., Ojima, D.S., Salvatore, M. & Conant,
R.T. 2015. Greenhouse gas mitigation potential of the world’s grazing lands: Modeling soil
carbon and nitrogen fluxes of mitigation practices. Agriculture, Ecosystems & Environment,
207: 91–100. Cited 21 July 2021. https://doi.org/10.1016/j.agee.2015.03.029
Hengl, T., de Jesus, J.M., MacMillan, R.A., Batjes, N.H., Heuvelink, G.B.M., Ribeiro, E.,
Samuel-Rosa, A., Kempen, B., Leenaars, J.G.B., Walsh, M.G. & Ruiperez Gonzalez, M.
2014. SoilGrids1km — Global Soil Information Based on Automated Mapping. PLOS ONE,
9(8): e105992. Cited 21 September 2021. https://doi.org/10.1371/journal.pone.0105992
Hergoualc’h, K., Akiyama, H., Bernoux, M., Chirinda, N., Prado, A., Kasimir, A.,
MacDonald J.D., Ogle, S.M., Regina, K., van der Weerden, T.J. 2019. N2O Emissions
from Managed Soils, and CO2 Emissions from Lime and Urea Application. Volume 4 Chapter
11. In: Calvo Buendia, E., Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize, S.,
Osako, A., Pyrozhenko, Y., Shermanau, P. and Federici, S. 2019 Refinement to the 2006
IPCC Guidelines for National Greenhouse Gas Inventories, (eds). Published: IPCC, Switzerland.
www.ipcc-nggip.iges.or.jp/public/2019rf/pdf/4_Volume4/19R_V4_Ch11_Soils_N2O_CO2.pdf
Hugelius, G., Bockheim, J.G., Camill, P., Elberling, B., Grosse, G., Harden, J. W., Johnson,
K., Jorgenson, T., Koven, C.D., Kuhry, P., Michaelson, G., Mishra, U., Palmtag, J., Ping,
C.L., O’Donnell, J., Schirrmeister, L., Schuur, E.A.G., Sheng, Y., Smith, L.C., Strauss, J.
& Yu, Z. 2013. A new data set for estimating organic carbon storage to 3 m depth in soils of
the northern circumpolar permafrost region. Earth System Science Data, 5: 393–402. Cited 21
September 2021. https://doi.org/10.5194/essd-5-393-2013
IIASA & FAO. 2012. Global Agro-ecological Zones (GAEZ v3.0). IIASA, Laxenburg, Austria
and FAO, Rome, Italy. Cited 21 September 2021. www.gaez.iiasa.ac.at/docs/GAEZ_Model_
Documentation.pdf
IPCC. 2019: Climate Change and Land: an IPCC special report on climate change, desertification,
land degradation, sustainable land management, food security, and greenhouse gas fluxes
in terrestrial ecosystems [P.R. Shukla, J. Skea, E. Calvo Buendia, V. Masson-Delmotte, H.- O.
Pörtner, D. C. Roberts, P. Zhai, R. Slade, S. Connors, R. van Diemen, M. Ferrat, E. Haughey,

S. Luz, S. Neogi, M. Pathak, J. Petzold, J. Portugal Pereira, P. Vyas, E. Huntley, K. Kissick, M.
Belkacemi, J. Malley, (eds.)]. In press.
Jebari, A., Álvaro-Fuentes, J., Pardo, G., Almagro, M. & del Prado, A. 2020. Simulating soil
organic C dynamics in managed grasslands under humid temperate climatic conditions, SOIL
Discuss [preprint]. Cited 21 July 2021. https://doi.org/10.5194/soil-2020-76
Kabirizi, J., Mpairwe, D. & Mutetikka D. 2004. Testing forage legume technologies with
smallholder dairy farmers: a case study of Masaka district, Uganda. Uganda Journal of
Agricultural Sciences, 9: 906–913. Cited 21 September 2021. www.ajol.info/index.php/ujas/
article/view/135688
Kaltenegger, K, Erb, Kl-H, Matej, S. & Winiwarter, W. 2021. Gridded soil surface nitrogen
surplus on grazing and agricultural land: Impact of land use maps. Environmental Research
Communications, 3(5): 055003. Cited 21 September 2021. https://doi.org/10.1088/2515-
7620/abedd8
Kemmitt, S.J., Wright, D., Goulding, K.W.T. & Jones, D.L. 2006. pH regulation of carbon and
nitrogen dynamics in two agricultural soils. Soil Biology and Biochemistry, 38(5): 898–911.
Cited 21 September 2021. https://doi.org/10.1016/j.soilbio.2005.08.006
Lal, R. 2004. Soil carbon sequestration to mitigate climate change. Geoderma, 123: 1–22. Cited
21 April 2022. https://doi.org/ 10.1016/j.geoderma.2004.01.032
Liu, X., Sheng, H., Wang, Z., Ma, Z., Huang, X. & Li L. 2020. Does grazing exclusion improve
soil carbon and nitrogen stocks in alpine grasslands on the Qinghai-Tibetan Plateau? A meta-
analysis. Sustainability, 12(3): 977. Cited 6 May 2022. https://doi.org/10.3390/su12030977
Lorenz, K. & Lal, R. 2018. Carbon Sequestration in Agricultural Ecosystems. Springer Nature
Switzerland AG. Cited 21 July 2021. https://doi.org/10.1007/978-3-319-92318-5
MacLeod, M. J., Vellinga, T., Opio C., Falcucci, A., Tempio, G., B.Henderson, Makkar, H.,
Mottet, A., Robinson, T., Steinfeld, H. & Gerber, P.J. 2018. Invited review: A position on
the Global Livestock Environmental Assessment Model (GLEAM). Animal, 12(2): 383–397.
Cited 21 July 2021. https://doi.org/10.1017/S1751731117001847
Martin, M.P., Dimassi, B., Dobarco, M.R., Guenet, B., Arrouays, D., Angers, D.A., Blache,
F., Huard, F., Soussana, J.F. & Pellerin, S. 2021. Feasibility of the 4 per 1000 aspirational
target for soil carbon: A case study for France. Global Change Biology, 27(11): 2458–2477.
Cited 21 July 2021. https://doi.org/10.1111/gcb.15547
Meersmans, J., Martin, M.P., Lacarce, E., Orton, T.G., De Baets, S., Gourrat, M., Saby,
N.P.A., Wetterlind, J., Bispo, A., Quine, T.A. & Arrouays, D. 2013. Estimation of soil
carbon input in France: An inverse modelling approach. Pedosphere, 23: 422–436. Cited 21
July 2021. https://doi.org/10.1016/S1002-0160(13)60035-1
Merbold, L., Scholes, R.J., Acosta, M., Beck, J., Bombelli, A., Fiedler, B., Grieco, E., Helmschrot,
J., Hugo, W., Kasurinen, V., Arne Körtzinger, D.G., Leitner, S., López‑Ballesteros, A., Ndisi,
M., Nickless, A., Salmon, E., Saunders, M., Skjelvan, I., Vermeulen. A.T. & Kutsch W.E.
2021. Opportunities for an African greenhouse gas observation system. Regional Environmental
Change, 21(104): 1–12. Cited 14 June 2022. https://doi.org/10.1007/s10113-021-01823-w
Morais, T. G., Teixeira, R. F.M. & Domingos, D. 2019. Detailed global modelling of soil
organic carbon in cropland, grassland and forest soils. PLoS ONE, 14(9): e0222604. Cited 21
July 2021. https://doi.org/10.1371/journal.pone.0222604
References 51
Mottet, A., de Haan, C., Falcucci, A., Tempio, G., Opio, C., & Gerber, P. 2017. Livestock:
On our plates or eating at our table? A new analysis of the feed/food debate. Global Food
Security, 14, 1-8.
Neumann, M., Zhao, M., Kindermann, G. & Hasenauer, H. 2015. Comparing MODIS net
primary production estimates with terrestrial national forest inventory data in Austria. Remote
Sensing, 7(4): 3878-3906. Cited 20 September 2021. https://doi.org/10.3390/rs70403878
Opio, C., Gerber, P., Mottet, A., Falcucci, A., Tempio, G., MacLeod, M., Vellinga, T., Henderson,
B. & Steinfeld, H. 2013. Greenhouse gas emissions from ruminant supply chains – A global life
cycle assessment. Rome, FAO. Cited 21 September 2021. www.fao.org/3/i3461e/i3461e.pdf
Palmer, B., Macqueen, D.J. & Gutteridge, R.C. 1994. Calliandra calothyrsus – a multipurpose
tree legume for humid locations. In: R.C. Gutteridge and H.M. Shelton, eds. Forest Tree
Legumes in Tropical Agriculture. Wallingford, UK, CAB International.
Paterson, R.T. 1994. Use of Trees by Livestock, Calliandra. Chatham, UK, Natural Resource
Institute.
Paustian, K., Levine, E., Post, W.M. & Ryzhova, I.M. 1997. The use of models to integrate
information and understanding of soil C at the regional scale. Geoderma, 79(1-4): 227–260.
Cited 21 July, 2021. https://doi.org/10.1016/S0016-7061(97)00043-8
Petri, M., Batello, C., Villani, R. & Nachtergaele, F. 2010. Carbon status and carbon
sequestration potential in the world’s grasslands. Integrated Crop Management, 11: 11–31.
Cited 20 September 2021. www.fao.org/3/i1880e/i1880e.pdf
Pramod, J., Lal Lakaria, B., Vishwakarma, A.K., Wanjari, R.H., Mohanty, M, Sinha,
N.K., Somasundaram, J., Dheri, G.S., Dwivedi, A.K., Sharma, R.P., Singh, M., Dalal,
R.C., Biswas, A.K., Patra, A.K. & Chaudhari, S.K. 2021. Modeling the organic carbon
dynamics in long-term fertilizer experiments of India using the Rothamsted carbon model.
Ecological Modelling, 450(109562): 1–9. Cited 21 October 2021. https://doi.org/10.1016/j.
ecolmodel.2021.109562
Puche, N., Senapati, N., Flechard, C.R., Klumpp, K., Kirschbaum, M.U.F. & Chabbi, A.
2019. Modeling carbon and water fluxes of managed grasslands: Comparing flux variability
and net carbon budgets between grazed and mowed systems. Agronomy, 9(4): 183. Cited 21
July 2021. https://doi.org/10.3390/agronomy9040183
R Core Team. 2013. R: A language and environment for statistical computing. R Foundation for
Statistical Computing, Vienna, Austria. Cited 21 September 2021. http://www.R-project.org/
Ruane, A.C., Goldberg, R. & Chryssanthacopoulos, J. 2015. Climate forcing datasets
for agricultural modeling: Merged products for gap-filling and historical climate series
estimation. Agricultural and Forest Meteorology, 200: 233–248. Cited 21 July 2021.
https://doi.org/10.1016/j.agrformet.2014.09.016
Sardans, J., Rivas-Ubach, A. & Peñuelas, J. 2012. The C:N:P stoichiometry of organisms and
ecosystems in a changing world: A review and perspectives. Perspectives in Plant Ecology, Evolution
and Systematics, 14(1): 33–47. Cited 6 May 2022. https://doi.org/10.1016/j.ppees.2011.08.002
Sejian, V., Naqvi, S.M.K., Ezeji, T., Lakritz, J. & Lal, R., eds. 2012. Environmental stress and
amelioration in livestock production. 569 pp. Berlin, Germany, Springer.
Six, J., Conant, R.T., Paul, E.A. & Paustian, K. 2002. Stabilization mechanisms of soil organic
matter: implications for C-saturation of soils. Plant and Soil, 241(2): 155–176. Cited 21
October 2021. https://doi.org/10.1023/A:1016125726789
Smith, J.U. & Smith, P. 2007. Environmental Modelling: An Introduction. 256 pp. Oxford,
United Kingdom, Oxford University Press.
Smith, P., Martino, D., Cai, Z., Gwary, D., Janzen, H., Kumar, P., McCarl, B., Ogle, S., O’Mara,
F., Rice, C., Scholes, B., Sirotenko, O., Howden, M., McAllister, T., Pan, G., Romanenkov,
V., Schneider, U., Towprayoon, S., Wattenbach, M. & Smith, J. 2008. Greenhouse gas
mitigation in agriculture. Philosophical Transactions of the Royal Society B: Biological Sciences,
363: 789–813. Cited 21 September 2021. https://doi.org/10.1098/rstb.2007.2184
Sombroek, W.G., Nachtergaele, F.O. & Hebel. A. 1993. Amounts, dynamics and sequestering
of carbon in tropical and subtropical soils. Ambio, 22: 417–426. Cited 11 May 2022.
Sommer, R. & Bossio, D. 2014. Dynamics and climate change mitigation potential of soil
organic carbon sequestration. Journal of Environmental Management, 144: 83–87. Cited 21
July 2021. https://doi.org/10.1016/j.jenvman.2014.05.017
Soussana, J.F., Tallec, T. & Blanfort, V. 2010. Mitigating the greenhouse gas balance of
ruminant production systems through carbon sequestration in grasslands. Animal, 4(3): 334–
350. Cited 21 October 2021. https://doi.org/10.1017/S1751731109990784
Tessema, B., Sommer, R., Piikki, K., Söderström, M., Namirembe, S., Notenbaert, A.,
Tamene, L., Nyawira, S. & Paul, B. 2020. Potential for soil organic carbon sequestration in
grasslands in East African countries: A review. Grassland Science, 66(3): 135–144. Cited 21
July 2021. https://doi.org/10.1111/grs.12267
Tifafi, M., Guenet, B. & Hatte, C. 2018. Large differences in global and regional total soil carbon
stock estimates based on SoilGrids, HWSD, and NCSCD: Intercomparison and evaluation
based on field data from USA, England, Wales, and France. Global Biogeochemical Cycle,
32(1): 42–56. Cited 21 September 2021. https://doi.org/10.1002/2017GB005678
UN. 2020. Map of the World. In: UN. Cited 1 November 2021. www.un.org/geospatial/content/
map-world
Uwizeye, A., Reppin, S., Opio, C., Teno, G., Lopes, J., Dondini, M. & Langston Diagne, M.
2021. Boosting Koronivia in the livestock sector – Workshop report. Rome, FAO. Cited 21 July
2021. https://doi.org/10.4060/cb4348en
Wiese, L., Wollenberg, E., Alcántara-Shivapatham, V., Richards, M., Shelton, S., Hönle,
S.E., Heidecke, C., Madari, B.E. & Chen, C. 2021. Countries’ commitments to soil organic
carbon in Nationally Determined Contributions. Climate Policy, 21(8): 1005–1019. Cited 14
June 2022. https://doi.org/10.1080/14693062.2021.1969883
Xu, X.L., Liu, W. & Kiely, G. 2011. Modelling the change in soil organic carbon of grassland in
response to climate change: Effects of measured versus modelled carbon pools for initializing
the Rothamsted carbon model. Agriculture, Ecosystems & Environment, 140: 372381. Cited
21 July 2021. https://doi.org/10.1016/j.agee.2010.12.018
Yang, Y., Tilman, D., Furey, G. & Lehman, G. 2019. Soil carbon sequestration accelerated
by restoration of grassland biodiversity. Nature Communications, 10: 718. Cited 21 October
2021. https://doi.org/10.1038/s41467-019-08636-w
Zomer, R.J., Bossio, D.A., Sommer, R. & Verchot, L.V. 2017. Global sequestration potential
of increased organic carbon in cropland soils. Scientific Reports, 7(1): 1–8. Cited 21 July 2021.
https://doi.org/10.1038/s41598-017-15794-8
FAO TECHNICAL PAPER
FAO ANIMAL PRODUCTION AND HEALTH PAPER
1 Animal breeding: selected articles from the World Animal Review, 1977 (En, Fr, Es, Zh)
2 Eradication of hog cholera and African swine fever, 1976 (En, Fr, Es)
3 Insecticides and application equipment for tsetse control, 1977 (En, Fr)
4 New feed resources, 1977 (En/Fr/Es)
5 Bibliography of the criollo cattle of the Americas, 1977 (En/Es)
6 Mediterranean cattle and sheep in crossbreeding, 1977 (En, Fr)
7 The environmental impact of tsetse control operations, 1977 (En, Fr)
7 Rev.1 The environmental impact of tsetse control operations, 1980 (En, Fr)
8 Declining breeds of Mediterranean sheep, 1978 (En, Fr)
9 Slaughterhouse and slaughterslab design and construction, 1978 (En, Fr, Es)
10 Treating straw for animal feeding, 1978 (En, Fr, Es, Zh)
11 Packaging, storage and distribution of processed milk, 1978 (En)
12 Ruminant nutrition: selected articles from the World Animal Review, 1978 (En, Fr, Es, Zh)
13 Buffalo reproduction and artificial insemination, 1979 (En*)
14 The African trypanosomiases, 1979 (En, Fr)
15 Establishment of dairy training centres, 1979 (En)
16 Open yard housing for young cattle, 1981 (Ar, En, Fr, Es)
17 Prolific tropical sheep, 1980 (En, Fr, Es)
18 Feed from animal wastes: state of knowledge, 1980 (En, Zh)
19 East Coast fever and related tick‑borne diseases, 1980 (En)
20/1 Trypanotolerant livestock in West and Central Africa – Vol. 1. General study,
1980 (En, Fr)
20/2 Trypanotolerant livestock in West and Central Africa – Vol. 2. Country studies,
1980 (En, Fr)
20/3 Le bétail trypanotolérant en Afrique occidentale et centrale – Vol. 3. Bilan d’une
décennie, 1988 (Fr)
21 Guideline for dairy accounting, 1980 (En)
22 Recursos genéticos animales en América Latina, 1981 (Es)
23 Disease control in semen and embryos, 1981 (En, Fr, Es, Zh)
24 Animal genetic resources – conservation and management, 1981 (En, Zh)
25 Reproductive efficiency in cattle, 1982 (En, Fr, Es, Zh)
26 Camels and camel milk, 1982 (En)
27 Deer farming, 1982 (En)
28 Feed from animal wastes: feeding manual, 1982 (En, Zh)
29 Echinococcosis/hydatidosis surveillance, prevention and control: FAO/UNEP/WHO
guidelines, 1982 (En)
30 Sheep and goat breeds of India, 1982 (En)
31 Hormones in animal production, 1982 (En)
32 Crop residues and agro‑industrial by‑products in animal feeding, 1982 (En/Fr)
33 Haemorrhagic septicaemia, 1982 (En, Fr)
34 Breeding plans for ruminant livestock in the tropics, 1982 (En, Fr, Es)
35 Off‑tastes in raw and reconstituted milk, 1983 (Ar, En, Fr, Es)
36 Ticks and tick‑borne diseases: selected articles from the World Animal Review,
1983 (En, Fr, Es)
37 African animal trypanosomiasis: selected articles from the World Animal Review,
1983 (En, Fr)
38 Diagnosis and vaccination for the control of brucellosis in the Near East, 1982 (Ar, En)
39 Solar energy in small‑scale milk collection and processing, 1983 (En, Fr)
40 Intensive sheep production in the Near East, 1983 (Ar, En)
41 Integrating crops and livestock in West Africa, 1983 (En, Fr)
42 Animal energy in agriculture in Africa and Asia, 1984 (En/Fr, Es)
43 Olive by‑products for animal feed, 1985 (Ar, En, Fr, Es)
44/1 Animal genetic resources conservation by management, data banks and training,
1984 (En)
44/2 Animal genetic resources: cryogenic storage of germplasm and molecular engineering,
1984 (En)
45 Maintenance systems for the dairy plant, 1984 (En)
46 Livestock breeds of China, 1984 (En, Fr, Es)
47 Réfrigération du lait à la ferme et organisation des transports, 1985 (Fr)
48 La fromagerie et les variétés de fromages du bassin méditerranéen, 1985 (Fr)
49 Manual for the slaughter of small ruminants in developing countries, 1985 (En)
50/1 Better utilization of crop residues and by‑products in animal feeding:
research guidelines – 1. State of knowledge, 1985 (En)
50/2 Better utilization of crop residues and by‑products in animal feeding:
research guidelines – 2. A practical manual for research workers, 1986 (En)
51 Dried salted meats: charque and carne‑de‑sol, 1985 (En)
52 Small‑scale sausage production, 1985 (En)
53 Slaughterhouse cleaning and sanitation, 1985 (En)
54 Small ruminants in the Near East – Vol. I. Selected papers presented for the Expert
Consultation on Small Ruminant Research and Development in the Near East
(Tunis, 1985), 1987 (En)
55 Small ruminants in the Near East – Vol. II. Selected articles from World Animal Review
1972‑1986, 1987 (Ar, En)
56 Sheep and goats in Pakistan, 1985 (En)
57 The Awassi sheep with special reference to the improved dairy type, 1985 (En)
58 Small ruminant production in the developing countries, 1986 (En)
59/1 Animal genetic resources data banks –
1. Computer systems study for regional data banks, 1986 (En)
2. Descriptor lists for cattle, buffalo, pigs, sheep and goats, 1986 (En, Fr, Es)
3. Descriptor lists for poultry, 1986 (En, Fr, Es)
60 Sheep and goats in Turkey, 1986 (En)
61 The Przewalski horse and restoration to its natural habitat in Mongolia, 1986 (En)
62 Milk and dairy products: production and processing costs, 1988 (En, Fr, Es)
63 Proceedings of the FAO expert consultation on the substitution of imported concentrate
feeds in animal production systems in developing countries, 1987 (En, Zh)
64 Poultry management and diseases in the Near East, 1987 (Ar)
65 Animal genetic resources of the USSR, 1989 (En)
66 Animal genetic resources – strategies for improved use and conservation, 1987 (En)
67/1 Trypanotolerant cattle and livestock development in West and Central Africa –
Vol. I, 1987 (En)
67/2 Trypanotolerant cattle and livestock development in West and Central Africa –
Vol. II, 1987 (En)
68 Crossbreeding Bos indicus and Bos taurus for milk production in the tropics, 1987 (En)
69 Village milk processing, 1988 (En, Fr, Es)
70 Sheep and goat meat production in the humid tropics of West Africa, 1989 (En/Fr)
71 The development of village‑based sheep production in West Africa, 1988 (Ar, En, Fr, Es)
(Published as Training manual for extension workers, M/S5840E)
72 Sugarcane as feed, 1988 (En/Es)
73 Standard design for small‑scale modular slaughterhouses, 1988 (En)
74 Small ruminants in the Near East – Vol. III. North Africa, 1989 (En)
75 The eradication of ticks, 1989 (En/Es)
76 Ex situ cryoconservation of genomes and genes of endangered cattle breeds by means of
modern biotechnological methods, 1989 (En)
77 Training manual for embryo transfer in cattle, 1991 (En)
78 Milking, milk production hygiene and udder health, 1989 (En)
79 Manual of simple methods of meat preservation, 1990 (En)
80 Animal genetic resources – a global programme for sustainable development, 1990 (En)
81 Veterinary diagnostic bacteriology – a manual of laboratory procedures of selected
diseases of livestock, 1990 (En, Fr)
82 Reproduction in camels – a review, 1990 (En)
83 Training manual on artificial insemination in sheep and goats, 1991 (En, Fr)
84 Training manual for embryo transfer in water buffaloes, 1991 (En)
85 The technology of traditional milk products in developing countries, 1990 (En)
86 Feeding dairy cows in the tropics, 1991 (En)
87 Manual for the production of anthrax and blackleg vaccines, 1991 (En, Fr)
88 Small ruminant production and the small ruminant genetic resource in tropical Africa,
1991 (En)
89 Manual for the production of Marek’s disease, Gumboro disease and inactivated
Newcastle disease vaccines, 1991 (En, Fr)
90 Application of biotechnology to nutrition of animals in developing countries, 1991 (En, Fr)
91 Guidelines for slaughtering, meat cutting and further processing, 1991 (En, Fr)
92 Manual on meat cold store operation and management, 1991 (En, Es)
93 Utilization of renewable energy sources and energy‑saving technologies by small‑scale
milk plants and collection centres, 1992 (En)
94 Proceedings of the FAO expert consultation on the genetic aspects of trypanotolerance,
1992 (En)
95 Roots, tubers, plantains and bananas in animal feeding, 1992 (En)
96 Distribution and impact of helminth diseases of livestock in developing countries, 1992 (En)
97 Construction and operation of medium‑sized abattoirs in developing countries, 1992 (En)
98 Small‑scale poultry processing, 1992 (Ar, En)
99 In situ conservation of livestock and poultry, 1992 (En)
100 Programme for the control of African animal trypanosomiasis and related development,
1992 (En)
101 Genetic improvement of hair sheep in the tropics, 1992 (En)
102 Legume trees and other fodder trees as protein sources for livestock, 1992 (En)
103 Improving sheep reproduction in the Near East, 1992 (Ar)
104 The management of global animal genetic resources, 1992 (En)
105 Sustainable livestock production in the mountain agro-ecosystem of Nepal, 1992 (En)
106 Sustainable animal production from small farm systems in South-East Asia, 1993 (En)
107 Strategies for sustainable animal agriculture in developing countries, 1993 (En, Fr)
108 Evaluation of breeds and crosses of domestic animals, 1993 (En)
109 Bovine spongiform encephalopathy, 1993 (Ar, En)
110 L’amélioration génétique des bovins en Afrique de l’Ouest, 1993 (Fr)
111 L’utilización sostenible de hembras F1 en la producción del ganado lechero tropical,
1993 (Es)
112 Physiologie de la reproduction des bovins trypanotolérants, 1993 (Fr)
113 The technology of making cheese from camel milk (Camelus dromedarius), 2001 (En, Fr)
114 Food losses due to non-infectious and production diseases in developing countries,
1993 (En)
115 Manuel de formation pratique pour la transplantation embryonnaire chez la brebis et la
chèvre, 1993 (F S)
116 Quality control of veterinary vaccines in developing countries, 1993 (En)
117 L’hygiène dans l’industrie alimentaire, 1993 – Les produits et l’aplication de l’hygiène,
1993 (Fr)
118 Quality control testing of rinderpest cell culture vaccine, 1994 (En)
119 Manual on meat inspection for developing countries, 1994 (En)
120 Manual para la instalación del pequeño matadero modular de la FAO, 1994 (Es)
121 A systematic approach to tsetse and trypanosomiasis control, 1994 (En/Fr)
122 El capibara (Hydrochoerus hydrochaeris) – Estado actual de su producción, 1994 (Es)
123 Edible by-products of slaughter animals, 1995 (En, Es)
124 L’approvisionnement des villes africaines en lait et produits laitiers, 1995 (F)
125 Veterinary education, 1995 (En)
126 Tropical animal feeding – A manual for research workers, 1995 (En)
127 World livestock production systems – Current status, issues and trends, 1996 (En)
128 Quality control testing of contagious bovine pleuropneumonia live attenuated vaccine –
Standard operating procedures, 1996 (En, Fr)
129 The world without rinderpest, 1996 (En)
130 Manual de prácticas de manejo de alpacas y llamas, 1996 (Es)
131 Les perspectives de développement de la filière lait de chèvre dans le bassin
méditerranéen, 1996 (Fr)
132 Feeding pigs in the tropics, 1997 (En)
133 Prevention and control of transboundary animal diseases, 1997 (E)
134 Tratamiento y utilización de residuos de origen animal, pesquero y alimenticio en la
alimentación animal, 1997 (Es)
135 Roughage utilization in warm climates, 1997 (En, Fr)
136 Proceedings of the first Internet Conference on Salivarian Trypanosomes, 1997 (En)
137 Developing national emergency prevention systems for transboundary animal diseases,
1997 (En)
138 Producción de cuyes (Cavia porcellus), 1997 (Es)
139 Tree foliage in ruminant nutrition, 1997 (En)
140/1 Analisis de sistemas de producción animal – Tomo 1: Las bases conceptuales, 1997 (Es)
140/2 Analisis de sistemas de producción animal – Tomo 2: Las herramientas basicas, 1997 (Es)
141 Biological control of gastro-intestinal nematodes of ruminants using predacious fungi,
1998 (En)
142 Village chicken production systems in rural Africa – Household food security and gender
issues, 1998 (En)
143 Agroforestería para la producción animal en América Latina, 1999 (Es)
144 Ostrich production systems, 1999 (En)
145 New technologies in the fight against transboundary animal diseases, 1999 (En)
146 El burro como animal de trabajo – Manual de capacitación, 2000 (Es)
147 Mulberry for animal production, 2001 (En)
148 Los cerdos locales en los sistemas tradicionales de producción, 2001 (Es)
149 Animal production based on crop residues – Chinese experiences, 2001 (En, Zh)
150 Pastoralism in the new millennium, 2001 (En)
151 Livestock keeping in urban areas – A review of traditional technologies based on
literature and field experiences, 2001 (En)
152 Mixed crop-livestock farming – A review of traditional technologies based on literature
and field experiences, 2001 (En)
153 Improved animal health for poverty reduction and sustainable livelihoods, 2002 (En)
154 Goose production, 2002 (En, Fr)
155 Agroforestería para la producción animal en América Latina – II, 2003 (Es)
156 Guidelines for coordinated human and animal brucellosis surveillance, 2003 (En)
157 Resistencia a los antiparasitarios – Estado actual con énfasis en América Latina, 2003 (Es)
158 Employment generation through small-scale dairy marketing and processing, 2003 (En)
159 Good practices in planning and management of integrated commercial poultry
production in South Asia, 2003 (En)
160 Assessing quality and safety of animal feeds, 2004 (En, Zh)
161 FAO technology review: Newcastle disease, 2004 (En)
162 Uso de antimicrobianos en animales de consumo – Incidencia del desarrollo de
resistencias en la salud pública, 2004 (Es)
163 HIV infections and zoonoses, 2004 (En, Fr, Es)
164 Feed supplementation blocks – Urea-molasses multinutrient blocks: simple and effective
feed supplement technology for ruminant agriculture, 2007 (En)
165 Biosecurity for Highly Pathogenic Avian Influenza – Issues and options, 2008 (En, Fr, Ar, Vi)
166 International trade in wild birds, and related bird movements, in Latin America and the
Caribbean, 2009 (Ese Ene)
167 Livestock keepers – guardians of biodiversity, 2009 (En)
168 Adding value to livestock diversity – Marketing to promote local breeds and
improve livelihoods, 2010 (En, Fr, Es)
169 Good practices for biosecurity in the pig sector – Issues and options in developing and
transition countries, 2010 (En, Fr, Zh, Ru** Es**)
170 La salud pública veterinaria en situaciones de desastres naturales y provocados, 2010 (Es)
171 Approaches to controlling, preventing and eliminating H5N1 HPAI in endemic countries,
2011 (En, Ar)
172 Crop residue based densified total mixed ration – A user-friendly approach to utilise
food crop by-products for ruminant production, 2012 (En)
173 Balanced feeding for improving livestock productivity – Increase in milk production and
nutrient use efficiency and decrease in methane emission, 2012 (En)
174 Invisible Guardians - Women manage livestock diversity, 2012 (En)
175 Enhancing animal welfare and farmer income through strategic animal feeding –
Some case studies, 2013 (En)
176 Lessons from HPAI – A technical stocktaking of coutputs, outcomes, best practices
and lessons learned from the fight against highly pathogenic avian influenza in Asia
2005−2011, 2013 (En)
177 Mitigation of greenhouse gas emissions in livestock production – A review of technical
options for non-CO2 emissions, 2013 (En, Ese)
178 Африканская Чума Свиней в Российской Федерации (2007-2012), 2014 (Ru)
179 Probiotics in animal nutrition – Production, impact and regulation, 2016 (En)
180 Control of Contagious Bovine Pleuropneumonia – A policy for coordinated actions, 2018
(En, Zh**)
181 Exposure of humans or animals to SARS-CoV-2 from wild, livestock,companion and
aquatic animals. Qualitative exposure assessment, 2020 (En)
182 The economics of pastoralism in Argentina, Chad and Mongolia. Market participation
and multiple livelihood strategies in a shock-prone environment, 2020 (En)
183 Introduction and Spread of lumpy skin disease in South, East and Southeast Asia.
Qualitative Risk Assessment and Management, 2020 (En)
184 Animal nutrition strategies and options to reduce the use of antimicrobials in animal
production, 2021 (En, Ru)
185 Pastoralism – Making variability work, 2021 (En)
186 Qualitative risk assessment for African swine fever virus introduction – Caribbean, South,
Central and North Americas, 2022 (En)
Availability: January 2023

Ar – Arabic Multil – Multilingual
En – English * Out of print
Es – Spanish ** In preparation
Fr – French e E-publication
Pt – Portuguese (En, Fr, Es) = Separate editions in

Ru – Russian English, French and Spanish
Vi – Vietnamese (En/Fr/Es) = Trilingual edition
Zh – Chinese
The FAO Animal production and health papers are available through authorized FAO Sales Agents
or directly from Sales and Marketing Group, FAO, Viale delle Terme di Caracalla, 00153 Rome, Italy.
Dondini, M., Martin, M., De Camillis, C., Uwizeye, A., Soussana, J.-F., Robinson, T. & Steinfeld, H. 2023.
Global assessment of soil carbon in grasslands – From current stock estimates
to sequestration potential
FAO Animal Production and Health Paper No. 187. Rome, FAO. https://doi.org/10.4060/cc3981en
Corrigendum
23 February 2023
The following corrections were made to the PDF of the report after it went to print.
Page Location Text in printed PDF Text in corrected PDF/ Notes

9 Full page The global map excludes some countries The global map contains all the countries
Contact: publishing-submissions@fao.org
This report presents the estimation of the baseline soil organic carbon stocks
in global grasslands in the year 2010. It also summarises the assessment of
the carbon input levels needed to maintain current SOC stocks, and the
evaluation of the soil organic carbon sequestration potential of grasslands if
management practices known to improve soil organic carbon sequestration
are implemented worldwide
The results show the importance of the interaction between climate and
grassland management, with the latter playing a crucial role in the quality
and quantity of organic material entering the soil. The report provides
spatially explicit evidence on the state of grassland soils and can be used as a
baseline for future work to explore the impacts of livestock management on
soil organic carbon at regional, country and farm levels.
Grasslands are one of the major ecosystems of the world, covering close to
one-third of the Earth’s terrestrial surface. Extensively managed grasslands
are recognized globally for their high biodiversity, and together with other
rangelands, they often contribute to agricultural production through
livestock browsing on natural forage, leaves, soft shoots and shrubs. It is,
therefore, evident that assessing the current state of grassland systems,
and their potential to sequester carbon in the soil, is of key importance for
understanding the trade-offs between grassland services on food security,
biodiversity conservation and climate mitigation and offsets, and how current
grassland management could be improved to meet global climate targets.
ISBN 978-92-5-137550-1 ISSN 0254-6019
9 789251 375501
CC3981EN/1/02.23

Pastizales FAO 2023

Uploaded by

Copyright:

Available Formats

Pastizales FAO 2023

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pastizales FAO 2023

Uploaded by

Copyright:

Available Formats

ISSN 0254-6019

FAO ANIMAL PRODUCTION AND HEALTH / PAPER 187

Food and Agriculture Organization of the United Nations

ISSN 0254-6019 [Print]

Cover photo: © Tibor Végh via Unsplash

5. Conclusions and way forward 43

Abbreviations and acronyms

4p1000 4 per 1000 initiative

NPP net primary production

CHEMICAL ELEMENTS AND FORMULAE

SOIL CARBON STOCKS IN GRASSLANDS

ASSESSMENT OF CURRENT CARBON STOCK LEVELS

SOIL CARBON SEQUESTRATION POTENTIAL

©Ivan Bandura on Unsplash

2.1 FAO LEAP GUIDELINES

2.2 FRAMEWORK AND METHODOLOGY DEVELOPMENT

2.3 ASSESSING THE BASELINE SOIL CARBON STOCKS OF GRASSLAND

=( × 0.3) × 0.475 Eq. 2

Where AGDM is the above-ground dry matter.

Where BGDM is the below-ground dry matter.

2.4 ASSESSING CARBON INPUT LEVELS NEEDED TO MAINTAIN CURRENT

Cbal = (CRes + CExc) - C0 Eq. 5

2.5 ASSESSING SOIL ORGANIC CARBON SEQUESTRATION POTENTIAL IN

3.1 GLOBAL BASELINE SOIL ORGANIC CARBON STOCKS

Central & South America

West Asia & Northern Africa

©Ralph (Ravi) Kayden on Unsplash

Central & South America

West Asia & Northern Africa

Central & South America

West Asia & Northern Africa

3.2 ASSESSMENT OF CURRENT CARBON STOCK LEVELS

Central & South America

West Asia & Northern Africa

Central & South America

West Asia & Northern Africa

West Asia & Northern Africa

Central & South America

Central & South America

West Asia & Northern Africa

3.3 SOIL ORGANIC CARBON SEQUESTRATION POTENTIAL

West Asia & Northern Africa

Central & South America

potential SOC sequestered baseline SOC stocks SOC sequestration rate

4.1 BASELINE SOIL ORGANIC CARBON

The establishment of fodder gardens is a practice that results in high productivity

More specifically, in the publications by Tessema et al. (2020), eight different

Potential increase in soil carbon after the establishment of fodder gardens

United Republic of Tanzania

Potential soil organic carbon (SOC) change (tonnes C/ha)

SOC change (t C/ha)

4.2 SOIL ORGANIC CARBON BALANCE

4.3 GRASSLAND POTENTIAL TO SEQUESTER SOIL CARBON

4.4 SOURCES OF UNCERTAINTIES REGARDING THE BASELINE SOIL

- 50% of initial carbon inputs -28.8%

5. Conclusions and way forward

Soils contribute to the achievement of the UN SDGs through carbon sequestration.

©Gabriel Jimenez on Unsplash

management, such as the incorporation of animal manures, agroforestry and rotational

was used to establish a reliable, transparent and cost-effective mechanism to monitor,