Heinemann et al. (2024). No saturation of soil carbon under long-term extreme manure additions

Vol.: (0123456789)

Plant Soil (2025) 512:1367–1384

https://doi.org/10.1007/s11104-024-07146-z

RESEARCH ARTICLE

No saturation of soil carbon under long‑term extreme

manure additions

Henrike Heinemann · Axel Don ·

Christopher Poeplau · Ines Merbach ·

Thorsten Reinsch · Gerhard Welp · Cora Vos

Received: 6 November 2023 / Accepted: 10 December 2024 / Published online: 19 December 2024

Abstract

Background and aims

According to the carbon (C)

saturation concept, the capacity of soils to accumu-

late stabilized organic C is limited by the number of

binding sites on mineral surfaces. The concept and its

application are highly debated. Therefore, we aimed

at testing this theory using field experimental data.

Methods

Soils were sampled from four long-term

field experiments with different amounts of organic

fertilisation going up to extreme high C inputs

(20 Mg C ha

−1

) five times higher than in com-

mon agricultural practice. Soils were fractionated by

particle size to obtain sand-sized, coarse silt and fine

silt plus clay fractions.

Results

We found a linear relation between C input

and soil organic carbon stocks (SOC) even with

vast amounts of organic C inputs to the soil at three

experimental sites. Across all experiments, C stocks

in the sand-sized fraction increased on average by

146%, whereas C stocks in the fine silt plus clay frac-

tion (< 20 µm) increased by just 17% without distinct

saturation behaviour. The C sequestration efficiency

(amount of C retained as SOC per amount of C input)

tended to increase with initial SOC content which is

not in line with the saturation theory.

Conclusion

The experiments were subject to C

inputs via organic fertilisation that would and should

rarely be reached in agricultural practice due to nega-

tive side effects. Even under these artificial condi-

tions experiments did not show a distinct saturation

behaviour.

Initial SOC stocks or SOC in the mineral-associated

fraction did not appear to limit the potential of soils to

sequester additional SOC. It can be concluded that C

sequestration is mainly limited by the availability of

C inputs from biomass.

Responsible Editor: Jens-Arne Subke.

Supplementary Information

The online version

contains supplementary material available at

https:// doi.

org/ 10. 1007/ s11104- 024- 07146-z

H. Heinemann · A. Don (

) · C. Poeplau · C. Vos

Thünen Institute of Climate-Smart Agriculture,

Bundesallee 65, 38116 Brunswick, Germany

e-mail: axel.don@thuenen.de

I. Merbach

Helmholtz-Centre for Environmental Research – UFZ,

Experimental Station Bad Lauchstädt, Hallesche Straße

44, 06246 Bad Lauchstädt, Germany

T. Reinsch

Institute for Crop Science and Plant Breeding – Grass

and Forage Science/ Organic Agriculture, Christian-

Albrechts-University, Hermann-Rodewald-Straße 9,

24118 Kiel, Germany

G. Welp

University of Bonn - Institute of Crop Science

and Resource Conservation, Nussallee 13, 53115 Bonn,

Germany

Page 2

1368

Plant Soil (2025) 512:1367–1384

Vol:. (1234567890)

Keywords

Soil organic carbon · Organic

fertilisation · Size fractionation · Mineral surfaces ·

Storage capacity · Carbon sequestration

Introduction

Soils are the largest pool of organic carbon (C) in the

biosphere. Global soil organic carbon (SOC) stocks

hold more than half of all terrestrial C (Jobbágy and

Jackson

2000

) and thus SOC plays a critical role in

the global C cycle. The accumulation of C in soils

can help mitigate climate change and sustaining agri-

cultural productivity (Lal

2004

; Minasny et al.

2017

In order to differentiate between the suitability of

soils for C storage, several studies have applied what

is known as the carbon saturation theory to identify

soils with high C-sequestration potential (Angers

et al.

2011

; Carter et al.

2003

; Chen et al.

2019

; Wies-

meier et al.

2014

). This approach assumes that the

capacity of soils to store stabilised i.e., mineral asso-

ciated, SOC is limited by the proportion of fine silt

and clay particles (< 20 µm). In general, two fractions

are often separated: particulate organic carbon (POC)

and mineral-associated organic carbon (MAOC).

POC consists mainly of rather undecomposed, light-

weight fragments that are not protected from decom-

position. The simpler organic compounds of MAOC

are associated with minerals and hence protected

from decomposition (Schmidt et al.

2011

; Lavallee

et al.

2020

). From a positive correlation between the

amount of the fine fraction and the MAOC content

in a soil, Hassink (

1997

) inferred that the capacity of

soils to protect SOC from decomposition is limited

by their clay and fine silt content. This assumption

was supported by Six et al. (

2002

) and Chung et al.

(

2010

Chung et al. (

2010

) found a finite capacity to store

C in soil aggregate fractions, while only the particu-

late organic matter sequestered C infinitely. A recent

study by Cotrufo et al. (

2019

) strengthened the con-

cept of MAOC saturation and predicted an upper

limit of 47 g C kg

−1

soil for European soils. Further

studies support the saturation concept with various

experimental approaches (Brown et al.

2014

; Chung

et al.

; Stewart et al.

has also been suggested, that not only the fine fraction

may saturate, but also the bulk SOC (Stewart et al.

2007

). The saturation concept implies, that once a

certain threshold is reached, additional carbon is only

accumulating in a relatively labile form, with rela-

tively faster turnover (Hassink

1997

). Following this

logic, soils with a relatively low content of SOC in

the fine fraction are assumed to build up bulk SOC at

a high rate per unit of C added to the soil and there-

fore having a high sequestration potential (Dexter

et al.

2008

), because the mineral phase provides space

for this C input to become long-term stabilized. Thus,

the C-sequestration efficiency, a ratio between C input

to the soils and stored C, would be high in soils with a

low SOC content, but low when soils are close to sat-

uration (Georgiou et al.

2022

). However, this coher-

ence has been challenged by several studies that did

not find an asymptotic relationship between C input

and C accumulation in soils that are supposedly close

to the C saturation level (Matus

2021

; Orgill et al.

2017

). Georgiou et al. (

2022

) have recently shown

that the degree of MAOC saturation might negatively

influence SOC accrual, which would be explica-

ble by the fact that SOC tends to accumulate in fast

cycling pools once a saturation of the slower-cycling

MAOC pool is reached. Latest research has seriously

challenged the concept of limited mineral-associated

organic carbon accumulation (Begill et al.

2023

) and

has confirmed that clay-associated C storage can be

driven by building a thicker loading of organic C,

irrespective of limitations by clay content (Schweizer

et al.

2021

). Begill et al. (

2023

) used soil inventory

data to explore a theoretical upper limit of MAOC in

agricultural soils at large scale. However, this is only

indirect evidence and lack direct field experimental

evidence how increasing C input to the soil is trans-

formed into different SOC pools and mineralised to

. Due to contrasting results in the literature as

well as methodological limitations to comprehen-

sively test the saturation concept it remains unclear i)

if MAOC saturation exists, ii) at which point MAOC

saturates and how pedo-climatic conditions can influ-

ence this potential point, iii) if a potential MAOC

saturation would be relevant for overall SOC dynam-

ics i.e., the efficiency of C sequestration, or even bulk

SOC saturation.

When testing for saturation behaviour of a cer-

tain SOC pool or bulk SOC, distinction between

equilibrium SOC stock and SOC saturation need to

be clear. SOC equilibrium or steady state can occur

after a certain time under constant agronomic man-

agement and climatic conditions, when C input to

Page 3

1369

Plant Soil (2025) 512:1367–1384

Vol.: (0123456789)

the soil and output of C mainly via mineralisation

balance each other (Stewart et al.

2007

; West and

Six

2007

; Fig.

a). If SOC is not changing it can be

a result of SOC being in equilibrium, which is in

independent of C saturation. With changes in agri-

cultural management practice, C-input levels may

change and SOC content will therefore increase

or decrease until a new equilibrium SOC stock is

reached (Fig.

a). Soils with a SOC content that is

close to the management-specific equilibrium SOC

content will show slower changes in SOC than soils

that are far from this steady state. Therefore, SOC

contents will show asymptotic behaviour when

expressed over time (Six et al.

2002

; Stewart et al.

2007

). This behaviour is often misinterpreted as

soils reaching maximum storage capacity. However,

C saturation in a SOC pool would only be reached

when an increase in C input (e.g., via a change in

management practice) does not result in rising

SOC stocks in this pool. Thus, saturation must be

assessed by monitoring SOC changes expressed

over different C-input levels (Stewart et al.

2007

According to the saturation concept, there is a max-

imum level of stabilized SOC that will be reached

under conditions with large amounts of C inputs to

the soil (Fig.

b). This is associated with a decreas-

ing fraction of C inputs that will be retained as

SOC, eventually resulting in no change in SOC due

to additional C input. Without saturation, the SOC

stock will continue to increase linearly with rising

C inputs (Stewart et al.

2007

; West and Six

2007

According to the theory, saturation would occur in

the MAOC fraction, while POM as larger size frac-

tions may continue to increase with C input (Has-

sink

1997

; Six et al.

2002

). To detect SOC saturation

behaviour in soils and soil fractions, C inputs have to

be high enough and need to be applied over a long

time (Stewart et al.

2007

). Here we used long-term

field experiments to make sure that saturation could

be reached and to test the following research ques-

tions: (a) Does an asymptotic relationship occur

between C inputs and SOC stocks of different sized

fractions and how does this correspond with bulk

SOC stocks for high C inputs? (b) Does the fine silt

and clay fraction saturate while the larger sized frac-

tions continue to accumulate SOC with increasing C

inputs? An asymptotic relationship in the fine silt and

clay fraction with larger fractions accumulating SOC

would support the saturation hypothesis. The satura-

tion theory would be contradicted by a linear relation-

ship of bulk SOC or the fine silt and clay fraction with

increasing C inputs. A non-declining C-sequestration

efficiency across all fertilisation treatments would

furthermore contradict the concept of saturation.

Fig. 1

Dynamics of SOC or C fractions expressed over time (

) to illustrate steady-state behaviour, and expressed over C input to

the soil (

) to illustrate saturation behaviour

Page 4

1370

Plant Soil (2025) 512:1367–1384

Vol:. (1234567890)

Materials & methods

Study sites & soil sampling

Three German field sites with four long-term experi-

ments under different land uses were selected for this

study. The sites differed in their soil type, texture and

soil pH (Table

). Each experiment comprised a large

range of C inputs with organic fertilisation, as well as

varying crop rotations (Table

The Incremental Farmyard Manure Experiment

in Bad Lauchstädt in eastern Germany was laid out

in 1983 as a two-factorial block experiment in two

replications. The first factor is organic fertilisation in

four increments (0, 50, 100 and 200 Mg fresh mat-

ter ha

−1

) applied as farmyard manure (Table

giving a wide range of C inputs for fertilisation of 0,

4.7, 9.4 and 18.7 Mg C ha

−1

(Table

). The sec-

ond factor is crop rotation versus a mechanical bare

fallow. The crop rotation has been continuous silage

maize since 2015. The historical management prior to

maize was a crop rotation of potato, sugar beet and

silage maize, which is described in detail in Schulz

(

2016

The experimental site in Kiel is part of a long-term

grassland re-sowing experiment (Chen et al.

2016

;

Reinsch et al.

2018

). Experimental plots were estab-

lished in 2005 in a randomised design on grassland

seeded in 1994 at the Lindhof experimental farm of

Kiel University. The farm is located in northern Ger-

many close to the Baltic Coast and is managed in

accordance with the German organic farmers’ asso-

ciation Bioland, which prohibits the use of mineral

fertilisers or pesticides. Organic fertiliser is applied

as cattle slurry in three different nitrogen (N) fertili-

sation doses (0, 240 and 480 kg N ha

−1

), which

correspond to average fertilisation treatments of 0, 1.7

and 3.3 Mg C ha

−1

. Additional organic C inputs

have occurred with grassland litter or via root resi-

dues (Table

). In order to maintain the soil’s nutrient

Table 1

Site characteristics with mean annual temperature (MAT) and mean annual precipitation (MAP)

Data derived from Reinsch et al. (

2018

) and Scherer et al. (

2011

) except for soil texture and pH value

Site

Location degrees

minutes (WGS84)

MAT (°C) MAP (mm) Soil type (FAO

2014

) Soil pH

(CaCl

)

Soil texture
sand (%) silt (%) clay (%)

Bad Lauchstädt 51°24’N 11°53’E 8.9

481

Haplic Chernozem

6.9

Kiel

54°27’N 9°57’E

8.9

768

Eutric Cambisol

6.1

Meckenheim

50°32’N 6°59’E

9.3

594

Luvisol

5.9

Table 2

Land use, fertilisation treatments and corresponding organic C inputs, experimental duration, and soil disturbance at the

sampling sites
Site

Current land use

Fertiliser

Fertilisation

treatments

(Mg dry mat-

ter

−1

)

Organic C ferti-

liser inputs

(Mg C

−1

)

Duration

(years)

Soil disturbance

Bad Lauchstädt

crop rotation and

mechanical bare

fallow

farmyard manure

ploughing

12.6

4.7

25.2

9.4

50.4

18.7

Kiel

grassland

cattle slurry

non

1.7

3.3

Meckenheim

crop rotation

farmyard manure

ploughing

2.8

5.6

Page 5

1371

Plant Soil (2025) 512:1367–1384

Vol.: (0123456789)

status, all treatments are fertilised at irregular inter-

vals with rock phosphate and potassium-magnesium

sulphate (Loges et al.

2018

The long-term field experiment in Meckenheim in

western Germany was established in 1962 as a ran-

domised complete block design with four replica-

tions. The selected treatments are mineral fertiliser

and farmyard manure with three fertilisation treat-

ments of 0, 20 and 40 Mg C ha

−1

(Table

). Since

1997 sugar beet, wheat and barley have been grown in

rotation (Scherer et al.

2011

). The soil pH values of

the experimental sites distinguished by fertilisation

treatments are given in Supplementary Table 1.

Disturbed soil samples were collected from the

topsoil using a soil corer. The sampling of the crop-

lands was performed according to ploughing depth

at 0–25 cm in Bad Lauchstädt (sampled in 2020, in

the centre parts of the plots) and at 0–30 cm depth

in Meckenheim (sampled in 2006). In Kiel, samples

were taken at depths of 0 to 5 cm (sampled in 2018).

Since it is a grassland, SOC inputs are mainly depos-

ited in the upper horizon and SOC stocks decline

exponentially with increasing depth (Don et al.

2007

Thus, the upper horizon is especially interesting when

testing for C saturation.

Soil fractionation

The soil samples were separated into three size frac-

tions (> 63 µm, 63–20 µm and < 20 µm) based on a

revised method of van Wesemael et al. (

2019

). All

the samples were dried and sieved at ≤ 2 mm. A sub-

sample of 10 g of fine earth (≤ 2 mm) was mixed

with 100 ml of deionised water and shaken horizon-

tally for 20 min at 20 rpm. The suspension was then

washed through sieves of different sizes, and care-

fully rinsed with a rubber spatula and a spray bottle

filled with deionised water. First the suspension was

poured through a 63-µm sieve into a beaker until the

liquid that passed through the sieve contained no

visible particles. The material that remained on the

sieve formed the sand-sized fraction, according to the

German Soil Classification (Ad-hoc-Arbeitsgruppe

Boden der Staatlichen Geologischen Dienste und der

Bundesanstalt für Geowissenschaften und Rohstoffe

2005

). The material that passed the 63-µm sieve was

then poured through a 20-µm sieve until the water

was clear. The material that stayed on the 20-µm

sieve consisted of coarse silt. The fractions remaining

on the sieves were collected separately and dried at

60 °C. The liquid that passed the 20-µm sieve was

then centrifuged for 30 min at 3515 rpm and the clear

liquid was decanted. The remaining material (fine silt

and clay) was transferred into a beaker and dried at

60 °C.

After drying, all the fractions were weighed,

ground and analysed for their concentration of organic

carbon by combustion using an elemental analyser

(LECO, TruMac, St. Joseph, MI, USA). After par-

ticle size fractionation, the mean mass recovery was

96.7%, while the minimal mass recovery was 94.9%.

The mean and minimal C recoveries were 91.5% and

79.8% respectively.

Calculations & statistics

Data analysis was performed using R version 4.3.0.

To test for saturation behaviour versus non-saturation

behaviour, two different models were fitted for the

individual experiments and size fractions. Before fit-

ting a model, the data was inspected for heterosce-

dasticity visually. Heteroscedasticity was detected for

Kiel, bulk soil. To account for this, the variance struc-

ture was allowed to increase with higher C inputs for

both models.

A linear model (Eq.

) represented ‘no saturation

behaviour’. The SOC content was a linear function of

organic C input per year and hectare:

where:

a, b

= coefficients of the model and

where:

fertiliser

= C inputs from fertiliser application

(Mg C ha

−1

plant

= C inputs from aboveground and below-

ground plant material (non- harvested plant residues).

fertiliser

was calculated from the fertiliser application

(Mg dry matter ha

−1

) and its C content.

plant

was

calculated from the yield data of the main crops with

regional specific allocation functions from Jacobs et al.

(

2020

). Yield data from the last 10 years before sam-

pling were obtained for the Kiel site, while yield data

from the Bad Lauchstädt site covered the last 20 years.

Yield data for Meckenheim were not available and were

therefore obtained from a long-term experiment on the

Dikopshof estate located 20 km from Meckenheim

(1)

SOC

input

(2)

input

fertiliser

plant

Page 6

1372

Plant Soil (2025) 512:1367–1384

Vol:. (1234567890)

to estimate the yield level under different fertilisa-

tion treatments. Mean average yield per treatment per

experiment was used to estimate plant residual C inputs

(Supplementary Table 2).

‘Saturation behaviour’ was defined as described in

Eq.

, where the SOC content approached a certain

asymptotic value with increasing C inputs:

where:

a, b, c

= coefficients of the model.

The asymptotic model was fitted using the function

nlsLM

for sites with four different fertilisation treat-

ments (e.g. Bad Lauchstädt) and a generalised additive

model (

gam

) for sites with three different treatments.

This was done because the

nlsLM

function cannot be

applied to data sets with only three treatments, but it

would easier to interpret. The

gam

was fitted using the

gam

function in the ‘mgcv’ R package (Wood

2011

To compare the models, the Akaike Information

Criterion was used with a correction for small sample

sizes (AICc). The AICc values for the linear model

were compared with the asymptotic models (

gam

nlsLM

). The site and fraction were matched to the cat-

egory ‘no saturation behaviour’ when the linear model

had a lower AICc than the asymptotic model. Burn-

ham and Anderson (

2004

) outlined the following rules

for correctly interpreting differences in AICc values: a

model scoring less well has a) still substantial support

if the difference in AICc scores is smaller than two, b)

it has ’considerably less support’ if the score difference

ranges between four to seven and c) has ’essentially no

support’ if the score is higher than 10. If the linear and

asymptotic model showed the same AICc, the simplest

(e.g. linear) model was chosen and the site was assigned

to ‘no saturation behaviour’. To exclude the possibility

of the C content already having reached the saturation

plateau at the asymptote, the procedure included test-

ing for a slope > 0 in the linear model at the significance

level p < 0.05.

C stocks (Mg ha

−1

) were calculated from the thick-

ness and bulk density of the horizons according to

Ellert and Bettany (

1995

where:

conc

= C concentration (%).

𝜌

= bulk density (g cm

−3

T= thickness of soil layer (cm)

fraction

= mass portion of individual fractions.

(3)

SOC

(

−

input

)

(4)

Cstocks

conc

𝜌

fraction

The bulk density in Kiel (1.13 g cm

−3

) was

measured, the bulk density in Bad Lauchstädt

(1.40 g cm

−3

) was measured from Körschens and

Waldschmidt (

1995

). No bulk density data were avail-

able for Meckenheim. To calculate the bulk density

for the 0–30 cm depth (1.24 g cm

−3

), data from 21

cropland sites from the same region with the same

soil type, land use and texture were taken from the

First German Agricultural Soil Inventory – Core data-

set (Poeplau et al.

2020

). The C stocks were mass

corrected for changing proportions of mineral and

organic compounds of each fraction with increasing

C input. Thus, we accounted for the effect of different

bulk density due to changes in C content with manure

addition.

The C-sequestration efficiency of bulk soil per fer-

tilisation treatment was calculated as follows:

where:

SOC

= mean SOC stocks of bulk soil (Mg

−1

Cinput

= C inputs from fertiliser and aboveground

and belowground plant material (Mg ha

−1

ref

= foregone fertilisation treatment as reference.

The ratio between actual and expected SOC

(SOC:SOC

exp

) was calculated according to (Poeplau

and Don

2023

) where:

where: clay = clay content (%).

In addition, the maximum amount of C (g kg

−1

)

associated with particles < 20 µm was calculated

according to Hassink (

1997

) as follows:

where

particles

𝜇

= fine silt and clay content

(%).

Results

Total annual organic C input

Total annual organic C inputs differed greatly between

fertilisation treatments and the four experiments. The

(5)

−

sequestration efficiency

SOC

high

−

SOC

ref

Cinput

high

−

Cinput

ref

(6)

SOC

exp

clay

0.0288

13.674

(7)

C in fraction

𝜇

4.09

0.37

∗ %

particles

𝜇

Page 7

1373

Plant Soil (2025) 512:1367–1384

Vol.: (0123456789)

range of maximum total annual organic C input to

the soil was between 6.5 Mg ha

−1

in grassland

at Kiel and 20.2 Mg ha

−1

in cropland at Bad

Lauchstädt (Table

). In Bad Lauchstädt two differ-

ent experiments were sampled, one with a crop rota-

tion and one with continuous bare fallow. Thus, there

was no organic C input at all in the no fertilisation

treatment of the bare fallow at Bad Lauchstädt, but

still 18.7 Mg organic C input ha

−1

from organic

fertilisation with the highest treatment. Especially

under low organic C fertiliser input, crop residues

contributed up to 49% of total organic C input to

the soil, with maximum fertilisation at the grassland

site in Kiel (Table

). As fertiliser organic C inputs

were high at Bad Lauchstädt, residues only contrib-

uted between 7.3% and 25% of total annual organic C

input to the soil.

Changes in SOC content of bulk soil

The long-term fertilisation treatments in each of the

four experiments resulted in a wide range of SOC

contents. The SOC content under no organic fer-

tilisation ranged from 11 g kg

−1

(Meckenheim) to

30 g kg

−1

(Bad Lauchstädt – crop rotation). The

highest fertiliser application in each experiment led

to SOC contents of between 15 g kg

−1

at Mecken-

heim and 57 g kg

−1

at Bad Lauchstädt (crop rotation)

(Fig.

At three of four sites, a linear model was the best

to describe the relationship between C input to soil

and SOC content (Fig.

, Table

), based on the fit-

ting of the linear and asymptotic models. Only at Bad

Lauchstädt, the crop rotation showed a slightly lower

AICc score for the asymptotic model and thus pos-

sibly indicates a non-linear fit (Table

). As the AICc

value is almost 5 units lower, the asymptotic model

has considerable support (Burnham and Ander-

son

2004

). All models showed a significant positive

trend in SOC with increasing C input (Table

). At

Meckenheim, the asymptotic model even showed an

exponential trend, which was in direct contrast to

the assumed asymptotic behaviour with saturation

(Fig.

d). As SOC stocks were calculated from SOC

contents with equivalent soil mass (Ellert and Bettany

1995

), the models for SOC stocks were equal to the

models fitted through SOC content data (Supplemen-

tary Fig. 1).

The C-sequestration efficiency of the bulk soil

(Eq.

) did not generally increase with an increas-

ing fine silt and clay fraction (Fig.

a). Bad Lauch-

städt had the highest content of the fine silt and clay

and a storage efficiency ranging from 0.05 to 0.24,

while the C-sequestration efficiency at Kiel, the site

with the lowest fine silt and clay content, ranged

from 0.09 to 0.12. Bad Lauchstädt showed the widest

range of C-sequestration efficiency, but the bare fal-

low was generally less efficient than the crop rotation.

Table 3

Amounts of

derived organic C from

different sources and total

annual organic C input per

treatment

Site

Land use/

management

Organic C

fertiliser input

levels

(Mg ha

−1

)

Mean annual organic C

input levels from residues

(Mg ha

−1

)

Total annual

organic C input

levels

(Mg ha

−1

)

Bad Lauchstädt crop rotation 0

1.4

4.7

1.6

6.3

9.4

1.5

10.9

18.7

1.5

20.2

bare fallow

4.7

9.4

18.7

Kiel

grassland

2.9

1.7

3.1

4.8

3.3

3.2

6.5

Meckenheim

crop rotation 0

1.1

2.8

1.8

4.6

5.6

2.0

7.6

Page 8

1374

Plant Soil (2025) 512:1367–1384

Vol:. (1234567890)

Meckenheim showed the lowest C-sequestration effi-

ciency while at the same time the initial SOC con-

tent in the fine silt and clay fraction was the lowest

of all experiments (Fig.

b). Kiel showed the high-

est initial SOC:SOC

exp

(Eq.

) of all experiments

(Fig.

c) and highest initial SOC content in the fine

silt and clay fraction (Fig.

b). At the same time the

C sequestration efficiency in Kiel was with around

10% similar as in Bad Lauchstädt and higher as in

Meckenheim.

The C-sequestration efficiency of bulk soil

increased at Bad Lauchstädt under both types of

management in relation to the none fertilisation treat-

ment, for the first fertilisation level to 0.24 under crop

Fig. 2

SOC content (g kg

−1

) of bulk soil at each sampling site with fitted linear, asymptotic and generalised additive models (GAM)

Table 4

AICc of the linear (lin), asymptotic (asymp) and generalised additive model (GAM) fitted to the SOC stocks at each sam-

pling site

Slope of the linear model with * showing significant differences from zero at

< 0.05

Site

Management

lin

GAM

asymp

Indicates

saturation

Linear fit

AICc

slope

Bad Lauchstädt

crop rotation

782.10

777.31

yes

5.82 *

bare fallow

661.70

663.65

3.29 *

Kiel

grassland

71.74

69.11

1.58 *

Meckenheim

crop rotation

819.81

802.42

2.19 *

Page 9

1375

Plant Soil (2025) 512:1367–1384

Vol.: (0123456789)

rotation and 0.15 under bare fallow (Fig.

). Under

the highest fertilisation level, the C-sequestration

efficiency of the crop rotation declined to 0.11 and

the bare fallow declined to 0.06. Thus, the crop rota-

tion stored C more efficiently than the bare fallow

although the initial bulk SOC content was higher in

the crop rotation (Fig.

d). At Kiel, the C-sequestra-

tion efficiency declined in relation to the foregone

fertilisation level from 0.12 to 0.09 with increasing

bulk SOC contents. In contrast, Meckenheim showed

an increase in C-sequestration efficiency from 0.03 to

0.07 with increasing bulk SOC contents.

Changes in size fractions with increasing C inputs

At all three sites, C stocks in the sand-sized frac-

tion increased the most with increasing C input. At

Kiel, the sand-sized fraction stored three times more

C than the smaller fractions, irrespective of the fer-

tilisation level (Supplementary Fig. 2). Based on the

AICc values, the linear regression was a better fit

than the asymptotic regression for the fine silt and

clay fraction in all four experiments (Table

). The

sand-sized fraction was the best described with a lin-

ear model at one experimental site as the AICc value

was lower. For the sand-sized and coarse silt fraction

at Meckenheim a nonlinear model had a lower AICc,

but instead of asymptotic behaviour it showed expo-

nential growth (Fig.

d). At the cropland site in Bad

Lauchstädt the sand size fraction showed a non-linear

behaviour with an asymptotic model being the best.

All other fractions at all sites were best described by

linear models without indication of saturation.

To test whether C fractions were increasing with

C input, we tested if the slopes of linear regres-

sions differed significantly from zero. C stocks of

Fig. 3

C-sequestration efficiency of bulk soil defined as the

ration between cumulative C input to the soil over the experi-

mental duration and the SOC stock change. The SOC:SOCexp

is an indicator for the SOC status of the soils taking into

account the effect of clay (Poeplau and Don

2023

)

Page 10

1376

Plant Soil (2025) 512:1367–1384

Vol:. (1234567890)

the sand-sized fraction increased significantly with

increasing fertiliser C input in all experiments,

except at Kiel. The coarse silt fraction showed a

significant slope of C content in all four experi-

ments. C stocks in the fine silt and clay fraction

increased significantly in Meckenheim and in Bad

Lauchstädt (bare fallow and crop rotation) but not

in Kiel (Table

C stored in the fine silt and clay fraction

Soils in the four experiments with organic fertilisa-

tion were potentially closer to a hypothetic satura-

tion than agricultural soils under common practices

due to excessive C inputs via fertilisation. Figure

shows a comparison of the actual stored C in the

fine silt plus clay fraction with storage capac-

ity calculated as described in Eq.

in accordance

with Hassink (

1997

). At Bad Lauchstädt, the high-

est hypothetical storage capacity was calculated

according to Hassink (

1997

) due to the highest fine

silt and clay content. Both crop rotation and bare

fallow exceeded the calculated storage capacity

even without organic C inputs (Fig.

a, b). At Kiel,

the C storage capacity was calculated to be 12 g C

−1

soil and was the lowest of all the sites. This

storage capacity was exceeded by almost five times

(Fig.

c). The storage capacity of the fine silt and

clay fraction was not reached in Meckenheim only.

Discussion

No saturation of bulk SOC

No saturation behaviour of bulk SOC was detected

at all three sites in this study. For one experiment

(Bad Lauchstädt – crop rotation) it was less clear.

In this case, the AIC was only slightly lower for

the asymptotic model than for the linear model.

According to Burnham and Anderson (

2004

) the

difference of AICc scores is too small to completely

omit the ‘weaker’ model but it has ‘considerably

less support’ than the better fitting one. In the case

of Bad Lauchstädt – crop rotation this means, that

there could be a tendency to saturation but it is not

distinctive. Still, even with using extreme agricul-

tural measures such as the heavy organic fertilisa-

tion we could not detect a systematically decreasing

proportion of C input that was retained in the soil.

In order to be able to detect C saturation effects,

long-term high C inputs are required. Was there

sufficiently high C input in the selected long-term

field experiments for C-saturation behaviour to be

determined? Mean inputs of organic C from crop

residues and organic fertilisers to agricultural soils

(arable land and grassland) in Germany are 3.7 Mg

C ha

−1

(Jacobs et al.

2020

). In the long-term

field experiments chosen for the present study, C

inputs exceeded this average C input rate up to 5.5-

fold. Such high C inputs would rarely be applied

Table 5

AICc of the linear

(lin), asymptotic (asymp)

and generalised additive

model (GAM) fitted to

C stocks of different size

fractions for each sampling

site

Slopes of the linear model

with * indicate a significant

difference from zero

(p < 0.05)

Site

Land use

Fraction

lin

GAM asymp Indicates

saturation

Linear fit

AICc

AICc AICc

Slope

Bad Lauchstädt crop rotation Fine silt & clay

95.10

95.10 no

0.47 *

Coarse silt

105.93

106.05 no

1.32 *

Sand

100.45

97.71 yes

1.94 *

bare fallow

Fine silt & clay

57.49

63.42 no

0.40*

Coarse silt

57.47

62.23 no

0.98*

Sand

56.04

60.25 no

1.28 *

Kiel

grassland

Fine silt & clay

3.63 3.85

0.07

Coarse silt

16.75 16.75

0.28 *

Sand

60.12 60.72

1.06

Meckenheim

crop rotation Fine silt & clay

80.55 80.55

0.43 *

Coarse silt

88.09 87.85

0.79 *

Sand

91.84 90.73

0.77 *

Page 11

1377

Plant Soil (2025) 512:1367–1384

Vol.: (0123456789)

in agricultural reality. This leads to the conclusion

that bulk SOC saturation is unlikely to occur under

common agricultural land use. Furthermore, our

findings are in line with other studies such as Orgill

et al. (

2017

), who found no limited C-storage capac-

ity in SOC with increasing C inputs in a laboratory

incubation experiment. Even though their C input

was equal to 40.6 Mg C ha

−1

and their mineral soils

had moderate to high mean initial SOC contents of

32 to 63 g kg

−1

, C input and SOC were linearly cor-

related. Thus, under controlled lab conditions with

an extreme C input treatment soil saturation behav-

iour was not detectable. Kong et al. (

2005

) found no

saturation behaviour of bulk SOC in a Californian

long-term (10 years) experiment featuring 10 dif-

ferent cropping systems with C inputs ranging from

0.8 to 9.0 Mg C ha

−1

either.

The findings at Bad Lauchstädt are in contrast to

those reported by Franko and Schulz (

2020

) who

found C saturation behaviour at the bare fallow of

Bad Lauchstädt. The different outcomes might be

driven by the sampling design that could explain

the different findings: Due to the absence of buffer

strips between each plot, there might be a carryo-

ver of organic fertilisers or soil. Whereas we sam-

pled only the centre of the plots, Franko and Schulz

(

2020

) sampled the whole plots potentially including

edge effects. Thus, the carry over might have a higher

influence when sampling the whole plots rather than

sampling in the middle of the plots.

It is assumed that saturation only occurs in the fine

silt and clay fraction, while particulate, non-protected

matter may increase linearly with increasing C input

(Castellano et al.

2015

; Cotrufo et al.

2019

; Stewart

Fig. 4

Soil C stocks of different size fractions with fitted linear, asymptotic or generalised additive models (GAM)

Page 12

1378

Plant Soil (2025) 512:1367–1384

Vol:. (1234567890)

et al.

2008

2012

). However, if the fine silt and clay

fraction would saturate at high C inputs, no linear

relation between C input and bulk SOC could be

obtained given the current kinetic conceptualisation

of the fractions is approximately correct. Only if the

fine fraction would decrease its turnover, a linear rela-

tionship might be maintained between total SOC and

C input when the fine fraction would saturate. How-

ever, this hypothetical decrease in the fine fraction C

turnover would need to be exactly to such an extent

that it compensates the missing build-up of mineral

associated C, which is unlikely the case.

In the present study, the fraction of C input that

is retained as SOC (C-sequestration efficiency) was

largely independent of the amount of manure that

was added and also independent of the status of the

soil in terms of SOC stocks and mineral associated C

(Fig.

). The SOC:SOC

exp

is a new indicator for the

SOC level in relation to a clay determined average

SOC content (SOC

exp

) (Poeplau and Don

2023

). It

ranged from 0.95 (Meckenheim) to 3.22 (Kiel), with

Bad Lauchstädt in between (1.93 bare fallow and 2.08

crop rotation) for the unfertilised plots respectively.

This indicates that even SOC contents in unfertilised

plots without organic amendments would be classi-

fied as good or very good according to Poeplau and

Don (

2023

) indicating high SOC levels as compared

to average German agricultural soils. With increasing

C input, SOC:SOC

exp

ratios rose up to 3.40 (Kiel),

2.55 (Bad Lauchstädt—crop rotation), 2.23 (Bad

Lauchstädt – bare fallow) and 1.02 (Meckenheim).

Furthermore, we found the lowest C-sequestration

Fig. 5

C concentrations in fine silt + clay fraction (< 20 µm) of the fertilisation treatment. Dashed line indicates the C-storage capac-

ity calculated according to Hassink (

1997

)

Page 13

1379

Plant Soil (2025) 512:1367–1384

Vol.: (0123456789)

efficiency in the soil with the lowest SOC:SOC

exp

ratio (Fig.

c), which contradicts the notion that soils

with a low SOC might sequester SOC more efficiently

(Georgiou et al.

2022

The crop rotation at Bad Lauchstädt was generally

more efficient than the bare fallow in converting C

input into bulk SOC (Fig.

). Thus, the additional C

input from plant residues did not reduce the efficiency

as it would be expected under C saturation condi-

tions. Root C inputs are known to be a main source

of stabilized SOC (Poeplau et al.

2021

). Thus, the

additional C inputs from belowground plant material

could be sequestered more efficiently than the manure

amendments, leading to greater total C-sequestration

efficiency in the crop rotation. This becomes appar-

ent by taking a closer look at the C-sequestration

efficiency (Fig.

): The grassland site Kiel showed a

higher C-sequestration efficiency than the cropland

site Meckenheim even though the clay content was

lower. But this behaviour could of course occur due

to a time effect (Fig.

) as the experiment in Kiel is

younger than the experiment in Meckenheim. This

indicates that the grassland might not be in a steady

state and could therefore show a higher C-sequestra-

tion efficiency.

Generally, permanent grasslands are known to

have higher SOC contents than arable land (Poeplau

and Don

2013

) because of high root C inputs and low

soil disturbance, and thus are expected to be more

easily C saturated (Hassink

1997

; Guillaume et al.

2022

). The additional SOC stored in grasslands is

thought to be non-complexed, unprotected organic

matter and thus to be sensitive to soil disturbance

(Carter et al.

2003

; Dexter et al.

2008

) because partic-

ulate, non-protected organic matter has a higher turn-

over than mineral-associated organic matter (Buy-

anovsky et al.

1994

; Poeplau et al.

2018

). Thus, the

high storage efficiency and high SOC stocks at Kiel

could be unprotected C accumulating not in the fine

silt and clay fraction. However, this was not the case,

as we discuss in the next chapter.

Differing responses of the size fractions

Hassink (

1997

) proposed a saturation line in depend-

ence on the clay content of soils and further studies

supported or developed this approach (Cotrufo et al.

; Dexter et al.

; Guillaume et al.

Dexter et al. (

) calculated a new saturation line

relating SOC stocks to clay content. They concluded

that the bulk SOC content of pasture mostly exceeds

the saturation line, while most arable soils fall below

the saturation line. The additional stored SOC in

grasslands is thought to be non-complexed, unpro-

tected organic matter and thus sensitive to soil distur-

bance. Our findings are not in line with this because

the silt and clay fraction exceeded the calculated

C-storage capacity of 1 g OC per 10 g of clay at all

experiments except at Meckenheim. Therefore, three

experiments stored more C in the fine silt and clay

fraction than the mineral surfaces should have been

able to bind. Guillaume et al. (

2022

) suggested an

even lower maximal value of C stored in the mineral-

associated OC (43 mg C g

−1

silt and clay) for perma-

nent grasslands. This threshold is exceeded in the fine

silt and clay fraction of the grassland at Kiel under

all fertilisation treatments (Fig.

c). In contrast to

the perception by Guillaume et al. (

2022

) that grass-

lands are C saturated in the fine fraction, the present

study and Fornara et al. (

2016

) found linear increases

in SOC stocks with increasing C input to the soil in

intensively fertilised grasslands. This suggests that

long-term C sequestration in permanent grasslands is

not limited by the fine silt and clay content and high-

lights the potential for carbon sequestration also in

land use types with relatively high SOC stocks.

Whilst Cotrufo et al. (

2019

) also found an upper

limit of mineral-associated OC storage of approxi-

mately 47 g kg

−1

bulk soil, this was recently ques-

tioned by Begill et al. (

2023

) based on a compre-

hensive data set of the German Agricultural Soil

Inventory. The authors found a linear correlation

between SOC and MAOC up to 12% SOC. Also,

Matus (

2021

) and Georgiou et al. (

2022

) found

MAOC contents of more than 80 g kg

−1

. This argues

against the notion of C saturation due to limited

mineral surfaces. It also shows that the experiments

in this study are not near the upper limits that were

recently proposed (Supplementary Fig. 1). This indi-

cates that even under high C inputs that are far beyond

common agricultural practice, saturation of MAOC is

not relevant.

A linear relationship between the SOC content

of the fine silt and clay fraction and C input to the

soil is often explained by a large saturation defi-

cit at which no non-linear saturation behaviour is

expected (Feng et al.

2014

; Kong et al.

2005

). For

example, Feng et al. (

2014

) did not detect saturation

Page 14

1380

Plant Soil (2025) 512:1367–1384

Vol:. (1234567890)

behaviour of fine silt and clay at four Chinese sites

and one Canadian agricultural site with different

crop rotations where initial C contents ranged from

2 to 35 g C kg

−1

. The authors argued that the inten-

sively cultivated soils in China and Canada exhib-

ited a large C-saturation deficit due to management

with low C inputs and a large storage potential due

to high clay contents (Feng et al.

2014

). C concen-

trations in the fine silt and clay fraction in our study

ranged from 13 to 45 g C kg

−1

which is at least

partly above proposed saturation lines.

The C-sequestration efficiency of the crop rota-

tion was higher than that of the bare fallow, which

suggests that the sequestration efficiency might also

be driven by the quality of C input. Differences due

to the tillage regime can be excluded because soil

management was the same in both treatments. The

quality of C input differs between sources, e.g., root

C input and straw. It is known that root C has a resi-

dence time in soils two to three times longer than

that of other crop residues or manure-derived C

inputs (Kätterer et al.

2012

; Menichetti et al.

2015

)

and is therefore more effective at maintaining and

building up SOC. This interpretation is underlined

by the fact that the C-storage efficiency in Kiel was

similar high even though this site stored more initial

C in the fine silt and clay fraction (Fig.

The storage of C in soils is influenced not only

by the quality of C inputs, but also by the decom-

position processes that depend on the quality and

dynamics of SOC) as well as the role of decom-

posers. The size of the microbial biomass pool, on

the other hand, may have a negative effect on SOC

size due to its positive impact on carbon decompo-

sition (Wieder et al.

2013

). Modelling results sug-

gest that microbially derived necromass C could

contribute anywhere from 10 to 80% to the forma-

tion of SOC (Fan et al.

2021

; Liang et al.

2019

Additionally, the composition of OC compounds

plays a role in the long-term storage of SOC. This is

because smaller, hydrophobic, and nitrogenous OC

compounds, derived from microbes, exhibit prefer-

ential sorption (Lehmann and Kleber

2015

; Barré

et al.

2018

). Therefore, decomposers can impact the

storage of SOC by altering the composition of OC

compounds (Kiem and Kögel-Knabner

2002

; Wies-

meier et al.

2019

). This study explores the results

of biological processes, but was not able to measure

them. Further research is necessary to capture the

seasonal dynamics while considering the long-term

trend in the experiments.

Several findings and concepts are proposing other

C-storage mechanisms than sorption to mineral sur-

faces (Kleber et al.

; Schweizer et al.

Vogel et al.

). The concept of organo-mineral

interactions was promoted by Kleber et al. (

2007

According to this model, C sorbs onto mineral sur-

faces but also onto organic matter in a zonal sequence.

Thus, under increasing C input, the mineral surface

coverage with OC will not necessarily increase, but

the OC layer will grow thicker. The protection of C is

greatest near to mineral surfaces in the ‘contact zone’,

where strong organo-mineral associations form (Kle-

ber et al.

2007

). Taking this model into account, the

C-storage capacity of soils is at least partly depend-

ent on the binding sites of mineral surfaces and there

should be saturation behaviour with decreasing C sta-

bilisation with increasing C inputs (Kögel-Knabner

et al.

2008

). In contrast, findings from Vogel et al.

(

2014

) showed that additional C input binds mainly to

already existing OC clusters, while mineral surfaces

are not completely occupied. In a study along a clay

gradient of cropland sites it was shown that clay-asso-

ciated C storage is driven by thicker accumulation of

organic C (Schweizer et al.

2021

). Thus, according

to these findings the capacity of soil to accumulate C

seems to be largely independent of the mineral sur-

face area (Schweizer et al.

2018

2021

; Vogel et al.

2014

). These nano-scale studies support our finding

that silt and clay mineral surfaces are not imposing

limitations for C stabilisation in agricultural soils.

Implications for C sequestration

The sequestration of SOC in soils can be an impor-

tant mean to mitigate climate change (Chenu et al.

2019

). Sequestration is the retention of atmos-

pheric C in the soil as part of soil organic matter

(Don et al.

2023

). To mitigate climate change, long-

term C sequestration is preferable. Thus, C should

be stored in stabile soil fractions such as the fine

silt and clay fraction (Six et al.

2002

). However,

in contrast to this our experiments sequestered C

with a high efficiency even though the additional C

input was not stored in the fine silt and clay frac-

tion (Fig.

). This study does not claim to be rep-

resentative for Germany and should be treated as a

case study. Therefore, caution should be used when

Page 15

1381

Plant Soil (2025) 512:1367–1384

Vol.: (0123456789)

trying to extrapolate these results. Furthermore, the

experiments were subject to C inputs via organic

fertilisation that would rarely be reached in agricul-

tural reality (Jacobs et al.

2020

). Organic fertilisa-

tion is used for agricultural nutrient management,

and SOC effects may thus be a by-product. Nutri-

ent applications need to be restricted and efficient

in order to avoid negative side effects. Increasing

manure application may increase environmental

costs due to nitrate leaching and may also result in

increased N

O emissions (Wang et al.

2019

). Thus,

C sequestration via organic fertilisation is restricted

in practice by these potential environmental costs

rather than the soil´s capacity to stabilise SOC.

With increasing C input, more C accumulated in

the coarse silt and sand-sized fraction. Coarse silt

can contribute to the stabilisation of organic mat-

ter by entrapping organic C within silt-sized aggre-

gates, where it is physically protected (Virto et al.

2008

). The sand-sized fraction consists of partially

decomposed plant residues. This particulate organic

matter is not physically stabilised (Gregorich et al.

2006

). Thus, C in the sand-sized fraction may easily

be affected by soil management and climate changes

(Carter et al.

2003

). However, the field experiments

that we investigated were between 15 and 44 years

old and showed a stable C retention also in the

coarse fraction. At least as transitional stage until

land use and management related steady state con-

ditions are reached also for the C pool distribution.

Overall, this data set suggests that long-term

C sequestration in soils is not limited by the abil-

ity of the fine silt and clay content to stabilise SOC

and highlights the potential for C sequestration in

different land use types even when SOC stocks are

already high. C sequestration in soils seems thus

possible in all soils, regardless of their actual SOC

content or their SOC in the fine silt and clay frac-

tion. Limited is the availability and quality of bio-

mass C which depends on ecosystem properties,

management and productivity. However, this study

does not aim to evaluate organic fertilisers. Instead,

it focuses on the principles of carbon saturation.

Acknowledgements

This project has received funding from

the European Unions´ Horizon 2020 research and innovation

programme under grant agreement No. 862695 EJP SOIL.

Funding

Open Access funding enabled and organized by

Projekt DEAL.

Data availability

The data of this manuscript are available at

https:// doi. org/ 10. 5281/ zenodo. 14444 507

Open Access

This article is licensed under a Creative Com-

mons Attribution 4.0 International License, which permits

use, sharing, adaptation, distribution and reproduction in any

medium or format, as long as you give appropriate credit to the

original author(s) and the source, provide a link to the Crea-

tive Commons licence, and indicate if changes were made. The

images or other third party material in this article are included

in the article’s Creative Commons licence, unless indicated

otherwise in a credit line to the material. If material is not

included in the article’s Creative Commons licence and your

intended use is not permitted by statutory regulation or exceeds

the permitted use, you will need to obtain permission directly

from the copyright holder. To view a copy of this licence, visit

http://creativecommons.org/licenses/by/4.0/

References

Ad-hoc-Arbeitsgruppe Boden der Staatlichen Geologischen

Dienste und der Bundesanstalt für Geowissenschaften und

Rohstoffe (2005): Bodenkundliche Kartieranleitung. With

assistance of Herbert Sponagel. 5., verb. und erw. Aufl.

Stuttgart, Hannover: Schweizerbart i. Komm; Bundesanst.

für Geowiss. und Rohstoffe.

Angers DA, Arrouays D, Saby NPA, Walter C (2011) Estimat-

ing and mapping the carbon saturation deficit of French

agricultural topsoils. Soil Use Manag 27(4):448–452.

https:// doi. org/ 10. 1111/j. 1475- 2743. 2011. 00366.x

Barré P, Quénéa K, Vidal A, Cécillon L, Christensen BT, Kät-

terer T et al (2018) Microbial and plant-derived com-

pounds both contribute to persistent soil organic carbon

in temperate soils. Biogeochemistry 140(1):81–92.

https://

doi. org/ 10. 1007/ s10533- 018- 0475-5

Begill N, Don A, Poeplau C (2023) No detectable upper limit

of mineral-associated organic carbon in temperate agri-

cultural soils. Global Chang Biol.

https:// doi. org/ 10. 1111/

gcb. 16804

Brown KH, Bach EM, Drijber RA, Hofmockel KS, Jeske ES,

Sawyer JE, Castellano MJ (2014) A long-term nitrogen

fertilizer gradient has little effect on soil organic matter in

a high-intensity maize production system. Global Change

Biol 20(4):1339–1350.

https:// doi. org/ 10. 1111/ gcb. 12519

Burnham KP, Anderson DR (2004) Multimodel inference.

Sociol Methods Res 33(2):261–304.

https:// doi. org/ 10.

1177/ 00491 24104 268644

Buyanovsky GA, Aslam M, Wagner GH (1994) Carbon

turnover in soil physical fractions. Soil Sci Soc Am J

58(4):1167–1173.

https:// doi. org/ 10. 2136/ sssaj 1994.

03615 99500 58000 40023x

Carter MR, Angers DA, Gregorich EG, Bolinder MA (2003)

Characterizing organic matter retention for surface soils

in eastern Canada using density and particle size frac-

tions. Can J Soil Sci 83(1):11–23.

https:// doi. org/ 10. 4141/

S01- 087

Castellano MJ, Mueller KE, Olk DC, Sawyer JE, Six J

(2015) Integrating plant litter quality, soil organic matter

Page 16

1382

Plant Soil (2025) 512:1367–1384

Vol:. (1234567890)

stabilization, and the carbon saturation concept. Global

Chang Biol 21(9):3200–3209.

https:// doi. org/ 10. 1111/

gcb. 12982

Chen S-M, Lin S, Loges R, Reinsch T, Hasler M, Taube F

(2016) Independence of seasonal patterns of root func-

tional traits and rooting strategy of a grass-clover sward

from sward age and slurry application. Grass Forage Sci

71(4):607–621.

https:// doi. org/ 10. 1111/ gfs. 12222

Chen S, Arrouays D, Angers DA, Martin MP, Walter C (2019)

Soil carbon stocks under different land uses and the appli-

cability of the soil carbon saturation concept. Soil Tillage

Res 188:53–58.

https:// doi. org/ 10. 1016/j. still. 2018. 11. 001

Chenu C, Angers DA, Barré P, Derrien D, Arrouays D, Bales-

dent J (2019) Increasing organic stocks in agricultural

soils: Knowledge gaps and potential innovations. Soil Till-

age Res 188:41–52.

https:// doi. org/ 10. 1016/j. still. 2018. 04.

011

Chung H, Grove JH, Six J (2008) Indications for soil carbon

saturation in a temperate agroecosystem. Soil Sci Soc Am

J 72(4):1132–1139.

https:// doi. org/ 10. 2136/ sssaj 2007.

0265

Chung H, Ngo KJ, Plante A, Six J (2010) Evidence for carbon

saturation in a highly structured and organic-matter-rich

soil. Soil Sci Soc Am J 74(1):130–138.

https:// doi. org/ 10.

2136/ sssaj 2009. 0097

Cotrufo MF, Ranalli MG, Haddix ML, Six J, Lugato E (2019)

Soil carbon storage informed by particulate and mineral-

associated organic matter. Nat Geosci 12(12):989–994.

https:// doi. org/ 10. 1038/ s41561- 019- 0484-6

Dexter AR, Richard G, Arrouays D, Czyż EA, Jolivet C, Duval

O (2008) Complexed organic matter controls soil physical

properties. Geoderma 144(3–4):620–627.

https:// doi. org/

10. 1016/j. geode rma. 2008. 01. 022

Don A, Schumacher J, Scherer-Lorenzen M, Scholten T,

Schulze E-D (2007) Spatial and vertical variation of soil

carbon at two grassland sites — Implications for meas-

uring soil carbon stocks. Geoderma 141(3–4):272–282.

https:// doi. org/ 10. 1016/j. geode rma. 2007. 06. 003

Don A, Seidel F, Leifeld J, Kätterer T, Martin M, Pellerin S

et al (2023): Carbon sequestration in soils and climate

change mitigation-definitions and pitfalls. Global Chang

Biol e16983.

https:// doi. org/ 10. 1111/ gcb. 16983

Du Z-L, Wu W-L, Zhang Q-Z, Guo Y-B, Meng F-Q (2014)

Long-term manure amendments enhance soil aggrega-

tion and carbon saturation of stable pools in North China

Plain. J Integr Agric 13(10):2276–2285.

https:// doi. org/

10. 1016/ S2095- 3119(14) 60823-6

Ellert BH, Bettany JR (1995) Calculation of organic matter and

nutrients stored in soils under contrasting management

regimes. Can J Soil Sci 75(4):529–538.

https:// doi. org/ 10.

4141/ cjss95- 075

Fan X, Gao D, Zhao C, Wang C, Qu Y, Zhang J, Bai E (2021)

Improved model simulation of soil carbon cycling by

representing the microbially derived organic carbon

pool. ISME J 15(8):2248–2263.

https:// doi. org/ 10. 1038/

s41396- 021- 00914-0

FAO (2014) World reference base for soil resources 2014.

International soil classification system for naming soils

and creating legends for soil maps. Rome: FAO (World

soil resources reports, 106). Available online at

https://

www. fao. org/3/ i3794 en/ I3794 en. pdf

, checked on 2 Feb

2022

Feng W, Xu M, Fan M, Malhi SS, Schoenau JJ, Six J, Plante

AF (2014) Testing for soil carbon saturation behavior in

agricultural soils receiving long-term manure amend-

ments. Can J Soil Sci 94(3):281–294.

https:// doi. org/ 10.

4141/ cjss2 013- 012

Fornara DA, Wasson E-A, Christie P, Watson CJ (2016) Long-

term nutrient fertilization and the carbon balance of per-

manent grassland: any evidence for sustainable intensifi-

cation? n Biogeosciences 13(17):4975–4984.

https:// doi.

org/ 10. 5194/ bg- 13- 4975- 2016

Franko U, Schulz E (2020) Carbon accumulation in a bare fal-

low Chernozem soil with high carbon input rates. Eur J

Soil Sci 72(1):265–273.

https:// doi. org/ 10. 1111/ ejss.

12937

Georgiou K, Jackson RB, Vindušková O, Abramoff RZ,

Ahlström A, Feng W et al (2022) Global stocks and

capacity of mineral-associated soil organic carbon.

Nat Commun 13(1):3797.

https:// doi. org/ 10. 1038/

s41467- 022- 31540-9

Gregorich EG, Beare MH, McKim UF, Skjemstad JO (2006)

Chemical and biological characteristics of physi-

cally uncomplexed organic matter. Soil Sci Soc Am J

70(3):975–985.

https:// doi. org/ 10. 2136/ sssaj 2005. 0116

Guillaume T, Makowski D, Libohova Z, Bragazza L, Sallaku F,

Sinaj S (2022) Soil organic carbon saturation in cropland-

grassland systems: Storage potential and soil quality. Geo-

derma 406:115529.

https:// doi. org/ 10. 1016/j. geode rma.

2021. 115529

Hassink J (1997) The capacity of soils to preserve organic C

and N by their association with clay and silt particles.

Plant Soil 191(1):77–87.

https:// doi. org/ 10. 1023/A: 10042

13929 699

Jacobs A, Poeplau C, Weiser C, Fahrion-Nitschke A,

Don A (2020) Exports and inputs of organic carbon

on agricultural soils in Germany. Nutr Cycl Agro-

ecosyst 118(3):249–271.

https:// doi. org/ 10. 1007/

s10705- 020- 10087-5

Jobbágy EG, Jackson RB (2000) The vertical distribution of

soil organic carbon and its relation to climate and vegeta-

tion. Ecol Appl 10(2):423–436.

https:// doi. org/ 10. 1890/

1051- 0761(2000) 010[0423: TVDOSO] 2.0. CO;2

Kätterer T, Bolinder MA, Berglund K, Kirchmann H (2012)

Strategies for carbon sequestration in agricultural soils

in northern Europe. Acta Agric Scand, A-Anim Sci

62(4):181–198.

https:// doi. org/ 10. 1080/ 09064 702. 2013.

779316

Kiem R, Kögel-Knabner I (2002) Refractory organic carbon

in particle-size fractions of arable soils II: organic carbon

in relation to mineral surface area and iron oxides in frac-

tions <6 μm. Org Geochem 33(12):1699–1713.

https://

doi. org/ 10. 1016/ S0146- 6380(02) 00112-2

Kleber M, Sollins P, Sutton R (2007) A conceptual model of

organo-mineral interactions in soils: self-assembly of

organic molecular fragments into zonal structures on

Page 17

1383

Plant Soil (2025) 512:1367–1384

Vol.: (0123456789)

mineral surfaces. Biogeochemistry 85(1):9–24.

https:// doi.

org/ 10. 1007/ s10533- 007- 9103-5

Kögel-Knabner I, Guggenberger G, Kleber M, Kandeler

E, Kalbitz K, Scheu S et al (2008) Organo-mineral

associations in temperate soils: Integrating biology,

mineralogy, and organic matter chemistry. J Plant Nutr

Soil Sci 171(1):61–82.

https:// doi. org/ 10. 1002/ jpln. 20070

0048

Körschens M, Waldschmidt U (1995) Ein beitrag zur quanti-

fizierung der beziehungen zwischen humusgehalt und

bodenphysikalischen eigenschaften. Arch. Acker- Pfl.

Boden. 39:165–173

Kong AYY, Six J, Bryant DC, Denison RF, van Kessel C

(2005) The relationship between carbon input, aggrega-

tion, and soil organic carbon stabilization in sustainable

cropping systems. Soil Sci Soc Am J 69(4):1078–1085.

https:// doi. org/ 10. 2136/ sssaj 2004. 0215

Lal R (2004) Soil carbon sequestration impacts on global cli-

mate change and food security. In Science (New York,

N.Y.) 304 (5677), pp. 1623–1627.

https:// doi. org/ 10. 1126/

scien ce. 10973 96

Lavallee JM, Soong JL, Cotrufo MF (2020) Conceptualizing

soil organic matter into particulate and mineral-associated

forms to address global change in the 21st century. Glob

Change Biol 26(1):261–273.

https:// doi. org/ 10. 1111/ gcb.

14859

Lehmann J, Kleber M (2015) The contentious nature of soil

organic matter. Nature 528(7580):60–68.

https:// doi. org/

10. 1038/ natur e16069

Liang C, Amelung W, Lehmann J, Kästner M (2019) Quanti-

tative assessment of microbial necromass contribution

to soil organic matter. Global Change Biol 25(11):3578–

3590.

https:// doi. org/ 10. 1111/ gcb. 14781

Loges R, Bunne I, Reinsch T, Malisch C, Kluß C, Herrmann A,

Taube F (2018) Forage production in rotational systems

generates similar yields compared to maize monocultures

but improves soil carbon stocks. Eur J Agron 97:11–19.

https:// doi. org/ 10. 1016/j. eja. 2018. 04. 010

Matus FJ (2021) Fine silt and clay content is the main factor

defining maximal C and N accumulations in soils: a meta-

analysis. Sci Rep 11(1):6438.

https:// doi. org/ 10. 1038/

s41598- 021- 84821-6

Menichetti L, Ekblad A, Kätterer T (2015) Contribution of

roots and amendments to soil carbon accumulation within

the soil profile in a long-term field experiment in Swe-

den. Agric Ecosyst Environ 200:79–87.

https:// doi. org/ 10.

1016/j. agee. 2014. 11. 003

Minasny B, Malone BP, McBratney AB, Angers DA, Arrouays

D, Chambers A et al (2017) Soil carbon 4 per mille. Geo-

derma 292:59–86.

https:// doi. org/ 10. 1016/j. geode rma.

2017. 01. 002

Orgill SE, Condon JR, Kirkby CA, Orchard BA, Conyers MK,

Greene RSB, Murphy BW (2017) Soil with high organic

carbon concentration continues to sequester carbon with

increasing carbon inputs. Geoderma 285:151–163.

https://

doi. org/ 10. 1016/j. geode rma. 2016. 09. 033

Poeplau C, Don A (2013) Sensitivity of soil organic carbon

stocks and fractions to different land-use changes across

Europe. Geoderma 192:189–201.

https:// doi. org/ 10.

1016/j. geode rma. 2012. 08. 003

Poeplau C, Don A (2023) A simple soil organic carbon level

metric beyond the organic carbon-to-clay ratio. Soil Use

Manage 39(3):1057–1067.

https:// doi. org/ 10. 1111/ sum.

12921

Poeplau C, Don A, Six J, Kaiser M, Benbi D, Chenu C et al

(2018) Isolating organic carbon fractions with varying

turnover rates in temperate agricultural soils – a compre-

hensive method comparison. Soil Biol Biochem 125:10–

26.

https:// doi. org/ 10. 1016/j. soilb io. 2018. 06. 025

Poeplau C, Don A, Schneider F (2021) Roots are key to

increasing the mean residence time of organic carbon

entering temperate agricultural soils. Global Chang Biol

27(19):4921–4934.

https:// doi. org/ 10. 1111/ gcb. 15787

Poeplau C, Don A, Flessa H (2020) Erste Bodenzustandserhe-

bung Landwirtschaft – Kerndatensatz: Johann Heinrich

von Thünen-Institut

Reinsch T, Loges R, Kluß C, Taube F (2018) Effect of grass-

land ploughing and reseeding on CO2 emissions and

soil carbon stocks. Agric Ecosyst Environ 265:374–383.

https:// doi. org/ 10. 1016/j. agee. 2018. 06. 020

Scherer HW, Metker DJ, Welp G (2011) Effect of long-term

organic amendments on chemical and microbial properties

of a luvisol. Plant Soil Environ 57(11):513–518.

https://

doi. org/ 10. 17221/ 3283- PSE

Schmidt MWI, Torn MS, Abiven S, Dittmar T, Guggenberger

G, Janssens IA et al (2011) Persistence of soil organic

matter as an ecosystem property. Nature 478(7367):49–

56.

https:// doi. org/ 10. 1038/ natur e10386

Schulz E (2016) Incremental Farmyard Manure Experiment

- Helmholtz-Zentrum für Umweltforschung UFZ. Avail-

able online at

https:// www. ufz. de/ index. php? de= 37008

updated on 4/14/2016, checked on 2/18/2022.

Schweizer SA, Hoeschen C, Schlüter S, Kögel-Knabner

I, Mueller CW (2018) Rapid soil formation after

glacial retreat shaped by spatial patterns of organic

matter accrual in microaggregates. Global Chang Biol

24(4):1637–1650.

https:// doi. org/ 10. 1111/ gcb. 14014

Schweizer SA, Mueller CW, Höschen C, Ivanov P, Kögel-

Knabner I (2021) The role of clay content and mineral

surface area for soil organic carbon storage in an ara-

ble toposequence. Biogeochemistry 156(3):401–420.

https:// doi. org/ 10. 1007/ s10533- 021- 00850-3

Six J, Conant RT, Paul EA, Paustian K (2002) Stabilization

mechanisms of soil organic matter: Implications for

C-saturation of soils. Plant Soil 241(2):155–176.

https://

doi. org/ 10. 1023/A: 10161 25726 789

Stewart CE, Paustian K, Conant RT, Plante AF, Six J (2007)

Soil carbon saturation: concept, evidence and evalua-

tion. Biogeochemistry 86(1):19–31.

https:// doi. org/ 10.

1007/ s10533- 007- 9140-0

Stewart CE, Paustian K, Conant RT, Plante AF, Six J (2008)

Soil carbon saturation: evaluation and corroboration by

long-term incubations. Soil Biol Biochem 40(7):1741–

1750.

https:// doi. org/ 10. 1016/j. soilb io. 2008. 02. 014

Stewart CE, Follett RF, Wallace J, Pruessner EG (2012) Impact

of biosolids and tillage on soil organic matter fractions:

implications of carbon saturation for conservation man-

agement in the Virginia Coastal Plain. Soil Sci Soc Am J

76(4):1257–1267.

https:// doi. org/ 10. 2136/ sssaj 2011. 0165

Page 18

1384

Plant Soil (2025) 512:1367–1384

Vol:. (1234567890)

van Wesemael B, Chartin C, Wiesmeier M, von Lützow M,

Hobley E, Carnol M et al (2019) An indicator for organic

matter dynamics in temperate agricultural soils. Agric

Ecosyst Environ 274:62–75.

https:// doi. org/ 10. 1016/j.

agee. 2019. 01. 005

Virto I, Barré P, Chenu C (2008) Microaggregation and organic

matter storage at the silt-size scale. Geoderma 146(1–

2):326–335.

https:// doi. org/ 10. 1016/j. geode rma. 2008. 05.

021

Vogel C, Mueller CW, Höschen C, Buegger F, Heister K,

Schulz S et al (2014) Submicron structures provide prefer-

ential spots for carbon and nitrogen sequestration in soils.

Nat Commun 5:2947.

https:// doi. org/ 10. 1038/ ncomm

s3947

Wang Y, Ying H, Yin Y, Zheng H, Cui Z (2019) Estimating

soil nitrate leaching of nitrogen fertilizer from global

meta-analysis. Sci Total Environ 657:96–102.

https:// doi.

org/ 10. 1016/j. scito tenv. 2018. 12. 029

West TO, Six J (2007) Considering the influence of sequestra-

tion duration and carbon saturation on estimates of soil

carbon capacity. Clim Chang 80(1–2):25–41.

https:// doi.

org/ 10. 1007/ s10584- 006- 9173-8

Wieder WR, Bonan GB, Allison SD (2013) Global soil car-

bon projections are improved by modelling microbial

processes. Nat Clim Chang 3(10):909–912.

https:// doi.

org/ 10. 1038/ nclim ate19 51

Wiesmeier M, Hübner R, Spörlein P, Geuß U, Hangen E, Reis-

chl A et al (2014) Carbon sequestration potential of soils

in southeast Germany derived from stable soil organic

carbon saturation. Global Chang Biol 20(2):653–665.

https:// doi. org/ 10. 1111/ gcb. 12384

Wiesmeier M, Urbanski L, Hobley E, Lang B, von Lützow M,

Marin-Spiotta E et al (2019) Soil organic carbon storage

as a key function of soils - a review of drivers and indica-

tors at various scales. Geoderma 333:149–162.

https:// doi.

org/ 10. 1016/j. geode rma. 2018. 07. 026

Wood SN (2011) Fast stable restricted maximum likelihood

and marginal likelihood estimation of semiparametric

generalized linear models. J Royal Stat Soc Series B: Stat

Methodol 73(1):3–36

Publisher’s Note

Springer Nature remains neutral with regard

to jurisdictional claims in published maps and institutional

affiliations.

Page 1

Page 2

Page 3

Page 4

Page 5

Page 6

Page 7

Page 8

Page 9

Page 10

Page 11

Page 12

Page 13

Page 14

Page 15

Page 16

Page 17

Page 18

Document Outline