Field Crops Research 299 (2023) 108971

Contents lists available at ScienceDirect

Field Crops Research

journal homepage: www.elsevier.com/locate/fcr

Higher lime rates for greater nitrogen recovery: A long-term no-till

experiment labeled with 15N
João William Bossolani a, Carlos Alexandre Costa Crusciol a, *, Eduardo Mariano a, b,
Luiz Gustavo Moretti a, José Roberto Portugal a, Mariley Fonseca a, Letusa Momesso a,
Andressa Selestina Dalla Côrt a, Vladimir Eliodoro Costa c, Heitor Cantarella d
a
São Paulo State University (UNESP), College of Agricultural Sciences (FCA), Department of Crop Science, 18610–034 Botucatu, São Paulo, Brazil
b
University of São Paulo (USP), Center for Nuclear Energy in Agriculture (CENA), 13416-000 Piracicaba, São Paulo, Brazil
c
São Paulo State University (UNESP), Institute of Biosciences (IBB), Center for Stable Isotopes “Prof. Dr. Carlos Ducatti” (CIE), 18618-689 Botucatu, São Paulo, Brazil
d
Agronomic Institute of Campinas (IAC), Soils and Environmental Resources Center, Campinas, SP, Brazil

A R T I C L E I N F O A B S T R A C T

Keywords: Context or problem: Soil acidity limits crop growth and yield all over the world. Low grain yields is usually
Liming associated with poor soil fertility; however, little attention has been given to the nitrogen-based fertilizer use
Maize/soybean rotation efficiency in soils managed with lime.
15
N-labeled fertilizer
Objective: Given the current scenario of uncertainties regarding the availability and prices of fertilizers, our study
Nitrogen losses
aimed to understand how maize intercropped with ruzigrass and soybean plants develop in long-term soils
managed with lime rates, and what the fate of the 15N–labeled ammonium sulfate [(15NH4)2SO4] applied in the
soil-plant system.
Methods: The treatments consisted of four dolomitic lime rates applied to the soil surface [control, half the
recommended lime rate (½ RLR), full recommended lime rate (1 RLR) and double the recommended lime rate (2
RLR)].
Results: The higher lime rate (2 RLR) improved fertility, carbon and nitrogen stocks in the soil profile, and grain
and/or stover production of maize, ruzigrass and soybean. As a consequence, maize and ruzigrass recovered a
high amount of 15N-fertilizer. On the other hand, soybean recovered less 15N-fertilizer, regardless of treatment,
but a greater amount was found in acidic soils. At the end of the maize and soybean growth cycles, our results
showed that in 2 RLR-amended soil, the 15N unrecovered was 71% lower than control. Finally, our results
suggested that the use of low lime rates (½ RLR) may increase the 15N losses potential to deep layers, whereas low
amounts of 15N were found in the subsoil when higher lime rates were applied.
Conclusions and implications: Soil acidity management through higher lime rates leads, over time, to increased soil
fertility, resulting in a favorable environment for plant growth and the use of nitrogen fertilizers. In this way, it is
possible to obtain a more productive and less costly agricultural system, and with less potential to pollute the
environment.

1. Introduction et al., 2017a; Bossolani et al., 2021b). In addition to correcting soil

acidity, liming improves soil fertility by supplying calcium (Ca2+) and
Lime application is an important practice for ameliorating soil magnesium (Mg2+), reducing toxic aluminum (Al3+) levels, and
acidity (Meng et al., 2019; Šiaudinis et al., 2020), a common issue in increasing soil organic matter (SOM) content over time (Briedis et al.,
many tropical regions around the world (Li et al., 2019; Patra et al., 2012; Bossolani et al., 2022a). As a consequence of soil improvements,
2021). Standard lime application practices have been developed for the crop root system can growth to deep layers, leading to higher uptake
long-term conservation systems in which there is no soil disturbance, of soil resources (water and nutrients) (Crusciol et al., 2019; Bossolani
such as no-tillage systems (NTSs) (Tiritan et al., 2016; Carmeis Filho et al., 2021b, 2022a), and greater fertilizer use efficiency (Fageria and

* Corresponding author.
E-mail address: [email protected] (C.A.C. Crusciol).

https://doi.org/10.1016/j.fcr.2023.108971
Received 1 February 2023; Received in revised form 8 May 2023; Accepted 11 May 2023
Available online 24 May 2023
0378-4290/© 2023 Elsevier B.V. All rights reserved.
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

Nascente, 2014; Crusciol et al., 2022b). Table 1

For maize (Zea mays L) and soybean (Glycine max) crops, nitrogen Geographic coordinates, climate and physical-chemical attributes, and crop
(N) is the most important nutrient (Bender et al., 2013, 2015). Biological management of the experimental field area. Botucatu, Brazil.
nitrogen fixation (BNF) provides practically all necessary N for soybean Site Description Value Unit
(Freitas et al., 2022). However, for maize, N fertilizers are required, Geographical coordinates
which significantly impacts production costs (Lu et al., 2021), particu­ Latitude 22◦ 83′ 3′ S
◦

larly in the current scenario of rising fertilizer prices (Schnitkey et al., Longitude 48◦ 42′ 6′ W
◦

2022). Furthermore, the complex dynamics of N in the soil–plant system Sea level 765 m
Climate attributes
can cause losses by volatilization, denitrification and leaching of nitrate a
Climate classification Mesothermal climate Cwa
(NO-3) in the soil profile (Zhou et al., 2021). Interestingly, leaching ac­ Annual precipitationb ~1360 mm
counts for the majority of N losses (Tamagno et al., 2022). Nevertheless, Air temperature (minimum) 15.3 ◦
C
N-fertilizer not absorbed by crops and present in soil surface layers Air temperature (maximum) 26.1 ◦
C
(biologically active layers), can be lost by denitrification (Bossolani Initial soil physical attributesc (0–20 cm)
Soil typed Typic Haplorthox
et al., 2020a). Agricultural practices that improve soil chemical char­
–
Clay 347 g kg− 1
acteristics can favor crop root growth, thereby allowing the exploitation Silt 108 g kg− 1
of a greater volume of soil and reducing NO-3 leaching and denitrification Sand 545 g kg− 1
(Caires et al., 2016). Bulk density 1.19 g cm− 3
Initial soil chemical attributesf (0–20 cm)
In NTSs, soil chemical attributes at deeper soil layers can be
pH (CaCl2) 4.2 –
improved by applying higher rates of lime on the soil surface (Carmeis Soil organic C (SOC) 12.2 g kg− 1
Filho et al., 2017a; Bossolani et al., 2022a), which can be enhanced by Phosphorus–available (P resin) 9.2 mg kg− 1

including tropical forage grasses in the system, particularly intercropped 2+

Calcium (Ca resin) 14.0
2+ 1
with maize (Ceccon et al., 2013; Costa et al., 2021). Tropical forage Magnesium (Mg resin) 5.0 mmolc kg−
1
Potassium (K+ resin) 1.2 mmolc kg−
grasses, such as ruzigrass, present abundant and aggressive root growth, 1
Total acidity (H+Al) (at pH 7.0) 37.0 mmolc kg−
occupying a large volume of soil (Baptistella et al., 2020). When Aluminum saturation (AS) 65.0 %
decomposing, these roots form biopores in the soil (Rosolem et al., Base saturation (BS) 35.0 %
2017), facilitating the translocation of suspended particles from the lime Cation exchange capacity (CEC pH 57.0 %
to deeper soil layers (Tiritan et al., 2016; Bossolani et al., 2020b). Under 7.0)
Crop management Maize Ruzigrass Soybean
these conditions, a soil acidity correction front forms in deeper layers Sowing Mar. 2019 Mar. 2019 Nov. 2019
down to 1.0 m, favoring greater growth of the subsequent crop root Cultivar Hybrid Common TMG 7062 IPRO
system, and thus increasing the possibility of NO-3 absorption (Calonego P3707VYH
and Rosolem, 2010), and reducing their losses to the environment Row spacing (m) 0.45 – 0.45
Plant density (plants ha− 1) 65.000 10 kg seeds 280.0000
(Rosolem et al., 2017).
Base fertilization (kg ha− 1) 28 N; 98 P2O5; – 0 N; 70 P2O5; 70
In the present study, we hypothesized that the surface application of 56 K2O K2O
double the recommended lime rate under long-term NTS based on maize Top dressing (N; kg ha− 1) 100 – –
intercropped with ruzigrass followed by soybean would: i) improve Harvest/desiccation Jul. 2019 Oct. 2019 Mar. 2020
chemical attributes in the soil profile; ii) increase biomass production; a
Alvares et al. (2013). bUnicamp (2020). cDonagema et al. (2017). dSoil
iii) increase the recovery of 15N-fertilizer by maize, ruzigrass and soy­ Survey Staff (2014). fPrior to establishment of the study (2002), the initial soil
bean crops; and iv) reduce 15N loss by potential leaching. To test these properties were determined at a depth of 0–20 cm according to van Raij et al.
hypotheses, we evaluated the effects of surface application of different (2001)
lime rates in a long-term experiment on soil fertility down to a depth of
100 cm; soil C and N stocks; maize, ruzigrass and soybean aboveground 2.2. Experimental design and field management
biomass production; the recovery of 15N fertilizer by the crops; and the
stratified soil 15N down to 100 cm depth. A randomized complete block design involving four different treat­
ments with four replicates each was used. Each plot was 57 m2 (9.0 ×
2. Material and methods 6.3 m). The treatments were (i) control (no liming); (ii) half the rec­
ommended lime rate (½ RLR); (iii) recommended lime rate (1 RLR) and
2.1. Site description and crop management (iv) twice the recommended lime rate (2 RLR). Over the 18-year
experimental period, the treatments were applied four times (2002,
This study used a long-term (18 years) field experiment [registered 2004, 2010, and 2016). Reapplications were based on the results of
by the Global Long Term Agricultural Experiments Network (GLTEN), annual assessments of base saturation. The lime rate was defined
Rothamsted Research, UK; https://www.glten.org/experiments/62] following the recommendations of van Raij et al. (1997) for fertilization
established in Botucatu, São Paulo State, Brazil. This experiment was and liming for the state of São Paulo, Brazil, considering base saturation
based on surface applications of lime in an agricultural system managed (BS) and cation exchange capacity (CEC). Over the time, liming occurred
under long-term no-till. All geographical, climate and soil attributes are in 2002 (beginning of the experiment; 1 RLR = 2.7 Mg ha− 1), 2004 (1
summarized in Table 1, and the climatic conditions during the experi­ RLR = 2.0 Mg ha− 1), 2010 (1 RLR = 2.0 Mg ha− 1), and 2016 (1 RLR =
mental period are illustrated in Fig. 1A. 6.5 Mg ha− 1). The residual effects of liming was characterized in 2019,
The chronological details of the crop management are detailed in the third year after the last soil amendment reapplication (2016). The
Fig. 1A and B and Table 1. The maize grain crop was intercropped (same cropping history from 2002 to 2020 and the details of previous treat­
row) with ruzigrass (Urochloa ruziziensis). The maize stover (stalk, ment applications (including the lime rates applied in each year) are
leaves, sheaths, tassel, core cob, and straw cob) was left in the field. given in Table S1.
After the maize harvest, live ruzigrass remained until October 2019,
when it was chemically terminated using glyphosate (2.5 kg ha–1 a.i.).
Soybean was then sown over the ruzigrass residue. For all crops, phy­ 2.3. Soil sampling and chemical analyses
tosanitary treatments were performed as necessary and recommended
for maize and soybean. To measure soil fertility status, composite soil samples (n = 8) were
taken from the 0–5, 5–10, 10–20, 20–40, 40–60, 60–80, and 80–100 cm

2
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

15
Fig. 1. (A) Weather conditions during the experimental period and (B) schematic representation of the plot, N–labeled fertilizer microplot, and sampling area.

layers of each plot using a soil push probe with an inner diameter of were DM is the total dry matter accumulated on aboveground of maize
50 mm. Sampling was performed in October 2019, before ruzigrass or ruzigrass; Nac is the total N accumulated on total DM; and Nap is the
desiccation (3 years after the last application of lime rates and 17 years applied N dose (28 kg ha− 1 at sowing + 100 kg ha− 1 at topdressing).
after the onset of the experiment). The soil was air-dried and ground to
pass through a 2 mm sieve for chemical analysis according to (van Raij 2.5. Establishment of 15
N microplots
et al., 2001). Total organic carbon (C) and total N were analyzed by an
elemental analyzer (LECO-TruSpect CHNS) using 0.2 g of soil. The Unconfined microplots (1.8 × 1.2 m) were set up in each treatment
samples were tested for the presence of CO2- 3 -C according to the method during the maize + ruzigrass season (Fig. 1 C). All agricultural man­
described by (Nelson and Sommers, 1996). As no CO2- 3 -C was found in agement in the microplots matched those of the main plots. Each
the soil (last liming occurred in 2016 and soil samplings occurred in microplot consisted of four rows of five maize plants each (~4.2 plants
2019), we assume that all C is present in organic forms and therefore m–1). 15N-labeled ammonium sulfate [(15NH4)2SO4] fertilizer with an
designate this main reservoir as organic C (OC). Then, when C and N abundance of 6.31 atom % 15N excess (Sigma–Aldrich Inc., St. Louis,
values were multiplied by the soil bulk density at each soil depth, MO, USA) was applied as topdressing (100 kg ha–1 N) at the V5 maize
originated the C and N stocks. Total C and N stocks were calculated by stage. The microplots received only 15N-labeled fertilizer, whereas the
summing all soil layers. main plots received unlabeled ammonium sulfate. After lab processing,
the maize stover was returned to the microplot to ensure the cycling of
2.4. Non-isotopic nitrogen use index the 15N present in the plant residues. After the maize harvest, the re­
covery of residual 15N-labeled fertilizer by soybean in the microplot was
The non-isotopic N use index were calculated only for maize and assessed.
ruzigrass crops. Soybean performs biological N fixation, which can lead
to misinterpretations of the conventional calculations used here. Nitro­ 2.6. Sampling procedure of 15
N-labeled material and isotopic analyses
gen usage index (NUI) and N uptake efficiency (NUpE) were calculated
according to the following equations: At maize physiological maturity (R6 stage), six maize plants were
NUI
( ) /
kg DMkg− 1 Nac = total aboveground DM total Nac (1) sampled from each microplot. The maize plants were partitioned into
vegetative fractions (stalk, leaves, sheaths, tassel, core cob, and straw
( ) / cob) and grains. All vegetative fractions from each microplot were
NUpE kgNac kg− 1
Nap = Nac N dose (2)
mixed (herein defined as stover), chopped with a forage grinder, and
oven-dried at 65 ◦ C to constant weight to obtain the dry weight. The

3
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

same procedure was used to dry the grain fraction. Subsamples of the and/or soil); 15N recovery is the percentage of fertilizer N recovery; NFR
dried stover and dried grain were ground in a Wiley mill (0.50 mm is the N fertilizer rate applied (kg ha–1); 15N unrecoveredFS and 15N
sieve). The remaining stover was returned to the microplots. unrecoveredSS are the percentages of N fertilizer unaccounted for (i.e.,
The ruzigrass biomass was sampled in October 2019, before chemical potential losses) after maize + ruzigrass (15N-labeled fertilizer applied
desiccation. In each microplot, an area of 0.25 m2 was collected (at in the maize + ruzigrass season) and after soybean (15N-labeled fertilizer
ground level). The ruzigrass biomass samples were oven-dried, and a residual in the second season), respectively; 15N recoveryTFS is the total N
subsample of the dry plant material was ground in a Wiley mill. Similar recovery (%; sum of maize stover and grain, ruzigrass, and soil) in the
to maize, the remaining biomass was returned to the microplots. At the maize + ruzigrass season; 15N remaining is the amount of 15N available in
beginning of soybean physiological maturity [R7 phenological stage; the system before soybean was grown; 15N recoverymaize grain is the
(Fehr and Caviness, 1977); March 2020], before leaf senescence, six amount of 15N exported in maize grain; 15N recoveryTFS is the total N
soybean plants were sampled and separated into vegetative fractions recovery (%; sum of soybean stover and grain and soil) in the second
[stem, leaves (including petiole), and pods] and grains. All vegetative season.
fractions were pooled and termed soybean stover. The same drying and Isotopic analysis was performed at the Stable Isotopes Center at São
grinding procedures were carried out. Plants from each main plot (maize Paulo State University – UNESP, Brazil. Dry and homogenized soil
stover and grains, ruzigrass, and soybean stover and grains) were sub­ samples were weighed to a mass of 30–35 mg in 5 × 8 mm tin capsules
jected to the same procedure above to assess the natural 15N abundance. (PN 24006400, Thermo Scientific, Germany). The capsules were
All milled plant tissue was used to determine the total N concentration analyzed in a continuous flow isotope ratio mass spectrometry system
and 15N measurements. (CF-IRMS) in which an IRMS (Delta V, Thermo Scientific, Germany) was
Soil was sampled using a core sampler at seven depths (0–5, 5–10, coupled to an elemental analyzer (Flash EA, Thermo Scientific, Ger­
10–20, 20–40, 40–60, 60–80, and 80–100 cm). Six soil samples were many) through a gas interface (ConFlo IV, Thermo Scientific, Germany).
collected per microplot and combined into one sample per depth per The IRMS determined the isotopic ratio of N (15N/14N) expressed as the
microplot. Three of the soil samples were collected from the rows of relative difference in the isotopic ratio (δ15N) in ‰ according to the
maize that received 15N-labeled fertilizer, and the other three samples following equation (Coplen, 2011):
were collected between the maize rows. The soil samples were oven-
R(15N/14N)sample
dried, ground in a ball mill and passed through a 100 mesh sieve δ15 N = − 1 (9)
(0.15 mm sieve). These soil samples were used to measure the total N R(15N/14N)Air
concentration, 15N and natural 15N abundance in the soil. To estimate The results were normalized via two-point anchoring (Paul et al.,
the soil N accumulation for each soil depth and treatment, the soil bulk 2007) using the IAEA-N-1 and IAEA-311 standards. The CF-IRMS stan­
density was assessed using the volumetric ring method (Blake and dard uncertainty for δ15N is ± 0.15‰ and ± 56.64‰ for IAEA-N-1 and
Hartge, 1986) during the ruzigrass season. All plant tissue (maize, IAEA-311, respectively.
ruzigrass, and soybean) and soil samples were analyzed for total N
concentration and 15N abundance using an automatic elemental
analyzer (Flash EA, Thermo Scientific, Germany) interfaced with an 2.8. Statistical analysis
isotope ratio mass spectrometer (CF–IRMS, Delta V, Thermo Scientific,
Germany). Our dataset were tested for normality [Anderson–Darling test;
(Nelson, 1998)] and homoscedasticity [Levene’s test; (Levene, 1960)].
Subsequently, the means were subjected to analysis of individual vari­
15
2.7. N calculations ance (one-way ANOVA) by the F test (p ≤ 0.05). Lime rates were
included as fixed effects. Blocks, growing seasons and depth (soil vari­
A range of variables were calculated to determine 15N recovery in the ables) were considered as random factors. When significant, means were
soil–plant system, including the amount of N derived from fertilizer compared using the modified t test [Fisher’s protected least significant
(Ndff), 15N recovery by crops, soil 15N retention in each soil layer and difference (LSD), at p ≤ 0.05].
across all soil depths (down to 100 cm), and unrecovered 15N. The first
season considered was maize + ruzigrass, while soybean was the second 3. Results
season. 15N recovery was determined according to the following
equations: 3.1. Soil chemical analysis
( )
Ndff kgha− 1 = (a/b ) × TN (3)
The lasting effects of surface liming at 48 months after the last lime
Ndfs
(
kgha− 1
)
= TN − Ndff (4) reapplication (2016) maintained soil pH ≥ 5.0 to a depth of 20 cm in 2
RLR-amended soil, whereas at lower lime rates, soil pH ≥ 5.0 occurred
15N recovery (%) = (Ndff /NFR) × 100 (5) only to a depth of 10 cm (Table 2). In addition, soil managed with 2 RLR
presented the highest soil pH values to a depth of 100 cm, ranging from
15N unrecoveredFS (%) = 100 − 15N recoveryTFS (6) 6 (0–5 cm) to 4.06 (80–100 cm), whereas the pH range with depth was
4.08 (0–5 cm) to 3.75 (80–100 cm) in the control treatment.
15Nremaining(%) = 15N recoveryTFS − 15N recoverymaize grain
Over time, the application of 2 RLR also increased CEC to a depth of
(7) 20 cm, whereas BS was higher in all soil layers to a depth of 100 cm,
with a range of 72.8% (0–5 cm) to 6.55% (80–100 cm). By contrast, in
15N unrecoveredSS (%) = 15N remaining − 15N recoverySS the control treatment, BS ranged from 15.3% (0–5 cm) to 0.63%
(8) (80–100 cm). In general, the higher the lime rate, the higher the BS in all
soil layers.
where Ndff is the N derived from fertilizer; a and b are the 15N enrich­ Even in deep layers, the lasting effects of the two highest lime rates (1
ment (atom % 15N excess) in the plant (maize/ruzigrass/soybean) RLR and 2 RLR) reduced the Al3+ concentration. Down to a depth of
fractions (stover and grain) or soil and in the substrate (fertilizer), 10 cm, the C stock was highest in 2 RLR-amended soil (20.3 Mg ha− 1),
respectively (a natural abundance of 0.368 atom % 15N was considered but at deeper soil layers, the C stock did not differ between 1 RLR and 2
in the calculations); TN represents the N accumulation (kg ha–1) in the RLR. Interestingly, the N stock down to a depth of 20 cm was highest at 2
plant fractions or soil; Ndfos is the N derived from other sources (BNF RLR (4.05 Mg ha− 1), followed by 1 RLR (3.53 Mg ha− 1); however, at

4
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

Table 2 aboveground dry matter production during the first (maize + ruzigrass)
Soil attributes according to the lime rate [control (no liming), half the recom­ and second (soybean) growing seasons (Fig. 2; Table S1). During the
mended lime rate (½ RLR), recommended lime rate (1 RLR) and twice the rec­ maize + ruzigrass season, soil managed with 2 RLR produced 20.57 Mg
ommended lime rate (2 RLR)] in seven stratified soil layers to a depth of 100 cm. ha− 1 of aboveground dry matter (stover = 6.24 Mg ha− 1; grain = 6.25
Soil Lime pH C N BS CEC Al3+ Mg ha− 1; ruzigrass = 8.08 Mg ha− 1), an increase of ~75% compared
depth rates stock stock with the control treatment (total dry matter = 11.76 Mg ha− 1; stover =
(cm) (CaCl2) (Mg ha− 1) % (mmolc kg− 1)
4.11 Mg ha− 1; grain = 2.98 Mg ha− 1; ruzigrass = 4.67 Mg ha− 1)
0–5 Control 4.08 d †
4.88c 0.84c 15.3 92 b 4.45 a (Fig. 2A). Moreover, applying 2 RLR increased total dry matter by
d ~53.3% and 13.3% compared with ½ RLR and 1 RLR, respectively.
½ RLR 5.32c 6.80 b 0.92c 40.9c 106 1.16 b
ab
Similar results were obtained for soybean (Fig. 2C). Specifically, the
1 RLR 5.66 b 7.92 b 1.09 b 59.0 108 0.15c control treatment produced ~4.35 Mg ha− 1 of aboveground dry matter
b ab (stover = 2.4 Mg ha− 1; grain = 1.95 Mg ha− 1), ~28.9%, 59.5% and 80%
2 RLR 6.00 a 10.8 a 1.37 a 72.8 126 a 0.16c less than ½ RLR (stover = 2.95 Mg ha− 1; grain = 2.6 Mg ha− 1), 1 RLR
a
(stover = 3.35 Mg ha− 1; grain = 3.59 Mg ha− 1), and 2 RLR (stover =
5–10 Control 3.68c 4.10 d 0.72 b 13.0 99c 6.68 a
d 3.61 Mg ha− 1; grain = 4.23 Mg ha− 1). Considering the grain yield
½ RLR 5.01 bc 5.71c 0.82 36.7c 110 b 1.34 b (grains at 13% of humidity), the highest lime rate increased the grain
ab yield of maize from 3.43 (control) to 7.18 Mg ha− 1 (2 RLR), and soybean
1 RLR 5.25 b 6.81 b 0.86 a 55.6 108 b 0.26c by 2.24 (control) to 4.86 Mg ha− 1 (2 RLR) (Fig. 2B and D).
b
2 RLR 5.40 a 9.50 a 0.94 a 70.4 125 a 0.24c
a 3.3. Nitrogen use efficiency during the maize + ruzigrass season
10–20 Control 4.01c 10.4 b 1.09c 6.78 103 a 14.9 a
d The rates of N use by maize and ruzigrass were also positively
½ RLR 4.66 bc 12.4 1.39 b 22.3c 99 a 2.65 b
influenced by liming (Fig. 3; Table S2). The N usage index of maize and
ab
1 RLR 4.83 ab 13.6 a 1.58 38.1 100 a 1.42c ruzigrass were higher at 2 RLR (maize = 1.0 kg DM kg− 1 Nac; ruzigrass =
ab b 0.4 kg DM kg− 1 Nac) and 1 RLR (maize = 0.91 kg DM kg− 1 Nac; ruzigrass
2 RLR 5.01 a 14.1 a 1.74 a 51.6 104 a 1.06c = 0.35 kg DM kg− 1 Nac) than in the other treatments (average maize =
a 0.71 kg DM kg− 1 Nac; average ruzigrass = 0.22 kg DM kg− 1 Nac)
20–40 Control 3.98c 22.8 b 2.34 a 4.82 124 b 15.5 a
d
(Fig. 3A, B). The N uptake efficiency of maize and ruzigrass also
½ RLR 4.11bc 26.1 a 2.40 a 15.9c 120 b 14.3 b increased with increasing lime rate (Fig. 3C, D). The N uptake efficiency
1 RLR 4.30 ab 27.2 a 2.53 a 25.8 146 a 3.87c of maize and ruzigrass plants was highest at 2 RLR (maize = 1.55 kg Nac
b kg− 1 Nap; ruzigrass = 1.62 kg Nac kg− 1 Nap), followed by 1 RLR (maize =
2 RLR 4.53 a 28.6 a 2.49 a 41.5 143 a 2.85c
1.35 kg Nac kg− 1 Nap; ruzigrass = 1.45 kg Nac kg− 1 Nap), ½ RLR (maize =
a
40–60 Control 3.83 b 20.6 b 2.47 b 2.16 164 a 16.7 a 0.94 kg Nac kg− 1 Nap; ruzigrass = 1.11 kg Nac kg− 1 Nap), and the control
d (maize = 0.73 kg Nac kg− 1 Nap; ruzigrass = 1.07 kg Nac kg− 1 Nap).
½ RLR 4.00 ab 22.7 a 2.91 a 7.14c 160 a 14.1 b
1 RLR 4.03 ab 23.7 a 2.30 b 11.1 162 a 10.3c 15
3.4. Fate of N fertilizer in the soil–plant system
b
2 RLR 4.14 a 24.3 a 2.31 b 17.5 157 a 9.40c
15
a N fertilizer recovery by the crops was positively affected by the
60–80 Control 3.76c 20.7 a 2.22 a 0.89 186 a 20.5 a lime rate (Fig. 4). Here, the values presented as percentages are equiv­
d alent to the values in kg ha− 1, as the applied dose of N was 100 kg ha− 1.
½ RLR 3.91 bc 21.4 a 2.61 a 3.24c 186 a 17.2
During the first season (maize + ruzigrass), 15N recovery was highest at
ab
1 RLR 4.02 ab 22.3 a 2.28 a 5.33 189 a 14.7 2 RLR (stover = 8.2%; grain = 29.6%; ruzigrass = 30.1%), followed by 1
b bc RLR (stover = 7.2%; grain = 26.4; ruzigrass = 25.9%), ½ RLR (stover =
2 RLR 4.13 a 22.7 a 1.59 b 9.15 187 a 11.3c 6.1%; grain = 16.1%; ruzigrass = 24.5%), and the control (stover =
a 5.9%; grain = 10.3%; ruzigrass = 22.9%) (Fig. 4A). The treatments had
80–100 Control 3.75 b 17.6 b 1.93 0.63 181 a 20.9 a
ab d
the strongest effect on grain 15N recovery. Considering all treatments,
½ RLR 3.84 ab 18.8 2.14 a 2.20c 187 a 18.2 a approximately 75% of the 15N from fertilizer found in maize shoots was
ab exported by the grains. The 15N found in maize grains (fraction exported
1 RLR 3.97 a 19.5 a 1.86 b 3.65 181 a 14.1 b by harvest) was 187%, 156% and 56% higher at 2 RLR (29.6%) than in
b
the control, ½ RLR and 1 RLR, respectively. Soil 15N retention (to a depth
2 RLR 4.06 a 19.7 a 1.37c 6.55 184 a 11.7 b
a of 100 cm) was higher at ½ RLR and 1 RLR (average = 21.9%) than in
the control and 2 RLR (average = 18%). Importantly, unrecovered 15N
† Different lowercase letters for each soil layer indicate significant differences
was highest in the control treatment (43.1%), followed by ½ RLR
between treatments by Fisher’s protected LSD test at p ≤ 0.05.
(32.2%), 1 RLR (17.8%) and 2 RLR (13.9%); even though 15N export was
highest at 1 RLR and 2 RLR, the amount of 15N remaining at the end of
depths greater than 40 cm, the N stock was highest at ½ RLR (7.66 Mg the first season was ~7% higher in these treatments (average = 52.6%)
ha− 1) and in the control (6.62 Mg ha− 1), followed by 1 RLR and 2 RLR than in the control and at ½ RLR (average = 49.2%) (Fig. 4B). Inter­
(average = 5.86 Mg ha− 1). At a depth of 100 cm, the total C stock estingly, most of the total N found in the shoots of maize and ruzigrass
increased with increasing lime rate as follows: 2 RLR (130 Mg ha− 1), 1 came from to the soil (76–80%) rather than N fertilizer (20–24%)
RLR (121 Mg ha− 1), ½ RLR (114 Mg ha− 1), and control (101 Mg ha− 1) (Fig. 5A and B). Regardless of which fraction (N fertilizer or soil) the N is
(Fig. S1A). By contrast, the total N stock at 100 cm was highest at ½ RLR sourced from, 2 RLR provided the highest N accumulation in maize and
(13.2 Mg ha− 1) and lowest in the control (11.6 Mg ha− 1) (Fig. S1B). ruzigrass plants.
Unlike maize, the recovery of 15N fertilizer by soybean was low
3.2. Total dry matter production and grain yield (Fig. 4C). On average, soybean stover and grains recovered 2.1% of the
15
N applied to the system. Interestingly, 15N recovery by soybean was
Surface-applied lime resulted in pronounced differences in total highest in the control (stover = 0.82%; grain = 2.11%), which, although

5
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

Fig. 2. Aboveground (stover + grain) dry matter yield in the (A) first (maize + ruzigrass) and (B) second (soybean) growing seasons in response to lime rate [control
(no liming), half the recommended lime rate (½ RLR), recommended lime rate (1 RLR) and twice the recommended lime rate (2 RLR)]. Different lowercase or capital
letters indicate significant differences between treatments by Fisher’s protected LSD test at p ≤ 0.05. Error bars express the standard error of the mean (n = 4).

low, was on average ~58% higher than in the other treatments. The (1.79%).
amount of 15N retained in the soil (down to 100 cm depth) did not differ
among the treatments with liming (average = 37.36%) and was ~13.5% 4. Discussion
higher in these treatments than in the control treatment (32.9%).
Additionally, the amount of unrecovered 15N fertilizer during the soy­ 4.1. Soil profile fertility and biomass production
bean season did not differ between treatments (average = 14.3%). After
the soybean harvest, the amount of 15N remaining in the control treat­ Surface application of lime without soil disturbance is a viable long-
ment (33.7%) was lower than the average for the liming treatments term practice to reduce subsoil acidity and increase soil profile fertility
(37.9%) (Fig. 4D). Considering the entire agricultural year (first + sec­ in tropical agricultural systems managed under no-till, but the magni­
ond seasons), total unrecovered 15N was lowest in the treatments that tude of the effect varies depending on the rate of lime application. Here,
received liming (Fig. 4E). Unrecovered 15N was highest in the control we determined the long-term impact on subsoil fertility of four appli­
treatment (53.9%), whereas 15N loss was lowest in 2 RLR (30.3%), fol­ cations (2002, 2004, 2010 and 2016) of lime rates over 17 years. Even
lowed by 1 RLR (34.5%) and ½ RLR (45.4%). 36 months after the last lime reapplication, the highest lime rate (2 RLR)
The N accumulation in the soybean shoots (stover + grain) showed was associated with pH > 5.0 to a depth of 20 cm, the soil layer where
that ~98% of the total N came from other sources (BNF + soil), whereas the concentration of crop roots is generally greatest (Rellán-Álvarez
only ~2% came from the N fertilizer (Fig. 5C). Soybean plants estab­ et al., 2016; Rosolem et al., 2017). Compared with 1 RLR, 2 RLR also
lished in acid soil (control treatment) were more dependent on soil N increased soil pH to a depth of 80 cm. Increased pH directly impacts the
(even if in low proportion) than in more fertile soils. On the other hand, concentration of free Al3+ in the soil layers. According to Bossolani et al.
N derived from other sources increased by 50% using 1/2 RLR, 92% on 1 (2022), high lime rates reduce Al3+ and boost the growth and distri­
RLR and by 125% on 2 RLR treatments. bution of the root system in the soil profile, leading to increased
acquisition of water and nutrients from the soil. Al3+ toxicity is one of
the main factors negatively affecting root growth (Parker et al., 1988;
15
3.5. N fertilizer retention in the soil profile Reis et al., 2018). The increase in soil pH [which induced deprotonation
of acidic groups and increased negative charges (increase in CEC)]
15
N retention in the soil profile after the first and second seasons (Limousin and Tessier, 2007) combined with the supply of Ca2+ and
varied with the liming rate (Fig. 6). Long-term application of higher lime Mg2+ by sedimentary dolomitic lime (calcium and magnesium carbon­
rates (1 RLR and 2 RLR) increased 15N retention in the uppermost soil ates) (Bossolani et al., 2020b) contributed to greater retention of Ca2+
layers (0–20 cm; average = 12.9%), but in layers below 60 cm, 15N and Mg2+ and, consequently, to increased BS along the soil profile.
fertilizer retention was highest at ½ RLR (5%) and lowest at 2 RLR Notably, there was a strong effect of the applied lime rates on BS vari­
(2.1%) (Fig. 6A). After the soybean harvest, 15N fertilizer retention was ations at soil layers below 20 cm.
highest in the 0–5 cm layer (average = 9.67%), and 15N fertilizer As a consequence of the increase in soil fertility at higher lime
retention in this layer decreased in the order 2 RLR (11.71%) > 1 RLR application rates, aboveground biomass (straw and grain) production by
(11.15%) > ½ RLR (8.74%) > control (7.08%) (Fig. 6B). At a depth of maize, ruzigrass and soybean was increased. Numerous studies have
5–20 cm, 15N fertilizer retention was higher at 1 RLR and 2 RLR (9.5%) reported increased biomass and grain production in lime-amended soils
than the average of ½ RLR and the control (5.6%). Interestingly, in the (Joris et al., 2016; Carmeis Filho et al., 2017a; Bossolani et al., 2018;
deepest layer (80–100 cm), 15N fertilizer retention was highest at ½ RLR Anderson et al., 2020; Crusciol et al., 2022a), primarily due to improved
(5.74%), followed by the control (4.95%), 2 RLR (2.47%), and 1 RLR

6
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

Fig. 3. Usage index (A = maize; B = ruzigrass) and uptake efficiency (C = maize; D = ruzigrass) of nitrogen in response to lime rate [control (no liming), half the
recommended lime rate (½ RLR), recommended lime rate (1 RLR) and twice the recommended lime rate (2 RLR)]. Different lowercase or capital letters indicate
significant differences between treatments by Fisher’s protected LSD test at p ≤ 0.05. Error bars express the standard error of the mean (n = 4).

soil profile fertility, lower Al3+ levels and increased root growth (greater increasing pH by liming helps increase the rate of mineralization of SOM
exploitation of soil resources). Interestingly, we also observed an in­ (Carmeis Filho et al., 2017b; Holland et al., 2018), resulting in greater N
crease in the N use index of maize and ruzigrass (N usage index and N release to the soil. However, SOM mineralization alone does not explain
uptake efficiency) in the present study, which, together with the the reduction in the N stock in the soil profile in lime-amended soil;
increased availability of nutrients in the soil profile, may have positively nutrient export also plays a direct role. The higher the grain yield, the
impacted plant growth. higher the export of nutrients, including N (Jones et al., 2013). In gen­
Here, our results showed that high lime rates are essential for eral, maize export (nutrient removal from area by crop harvest)
building up fertility in the soil profile over the years. Tropical regions are approximately 15–18 kg N Mg− 1 of grain produced (Bender et al.,
characterized by low rainfall during autumn/winter cultivation (maize 2013), whereas soybean, which are largely benefited from BNF, exports
season) and recurrent dry spells during spring/summer cultivation ~45–60 kg N Mg− 1 of grain produced (Bender et al., 2015; Esper Neto
(soybean season) (Cunningham, 2020); therefore, improving the pro­ et al., 2021). Nutrient export may explain the higher N stock in the soil
duction environment is essential for maize and soybean to reach high profile at ½ RLR (which had the lowest grain yield with the exception of
grain yields and production stability (Carmeis Filho et al., 2017a; Bos­ the control) compared with 1 RLR and 2 RLR. Nutrient export can
solani et al., 2021b, 2022a). directly influence the soil N stock. The N stock was low in the control
In our study, we observed high input of organic residues into soil treatment due to the lower biomass C input by the crops over the years
managed with higher lime rates. C input from biomass and its rela­ and the lower N fertilizer use efficiency due to low absorption and the
tionship with SOC stocks are important indicators of the influence of high rate of denitrification in acidic soils (Jones et al., 2013), which
lime on C sequestration (Briedis et al., 2012; Carmeis Filho et al., 2017b; caused a large part of the N be lost to the environment (leaching and/or
Inagaki et al., 2017) and soil quality (Inagaki et al., 2016; Bossolani denitrification) (Zhou et al., 2021; Tamagno et al., 2022).
et al., 2021a). Crop residues are the main source of SOM (Briedis et al.,
2012). Both the quantity and quality of the crop biomass (crop rotation; 4.2. 15
N fate in the soil–plant system
Table S1) that accumulated between 2002 and 2019 impacted the
accumulation of C in the soil profile, as evidenced by the increase in the Increasing soil fertility by applying lime altered 15N fate in the
C stock with increasing lime rate. soil–plant system. Maize and ruzigrass plants grown in lime-amended
The N stock is influenced by many factors (Song et al., 2018). For soils, particularly when 2 RLR was applied, presented higher 15N fer­
example, after mineralization by soil microbes, SOM becomes a poten­ tilizer recovery in the aboveground biomass (stover of maize and ruzi­
tial source of mineral N for crops (Bertol et al., 2022). Interestingly, grass, and maize grain). The N fertilizer recovered by crops is exported

7
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

Fig. 4. (A, C) 15N recovery in each compart­

ment (plant, soil or unrecovered) in the (A) first
(maize + ruzigrass) and (C) second (soybean)
growing seasons; (B, D) 15N fertilizer remaining
after the (B) first and (D) second growing sea­
sons; (E) total 15N unrecovered (first + second
growing seasons) in response to lime rate
[control (no liming), half the recommended
lime rate (½ RLR), recommended lime rate (1
RLR) and twice the recommended lime rate (2
RLR)]. Different lowercase or capital letters
indicate significant differences between treat­
ments by Fisher’s protected LSD test at
p ≤ 0.05. Error bars express the standard error
of the mean (n = 4).

Fig. 5. Unlabeled nitrogen derived from fertil­

izer and from other sources (BNF and/or soil) in
ruzigrass (A), maize (B) and soybean (C) shoots
in response to lime rate [control (no liming),
half the recommended lime rate (½ RLR), rec­
ommended lime rate (1 RLR) and twice the
recommended lime rate (2 RLR)]. Different
lowercase or capital letters indicate significant
differences between treatments by Fisher’s
protected LSD test at p ≤ 0.05. Error bars ex­
press the standard error of the mean (n = 4).

in the grain harvest or returned to the system by nutrient cycling from (Fageria and Nascente, 2014). In our study, the N fertilizer efficiency
crop residues (Chen et al., 2014), as supported by our results for 15N was one of the main factors for maize to achieve higher grain yield.
remaining in the system. To increase N fertilizer use efficiency in agri­ Interestingly, Bossolani et al. (2022) reported that the number of grains
cultural systems, most of the applied fertilizer must be absorbed and per plant and the 100-grains weight were the main production compo­
assimilated by the crops and converted into biomass (Fageria and nents affected by higher lime rates, which directly impacted the final
Moreira, 2011). The combination of soil acidity correction and soil maize grain yield. Ammonium-based fertilizers undergo nitrification
-
fertility improvement increases soil exploitation by plant roots (Bosso­ (conversion of NH+ 4 to NO3), which is enhanced in acidity-corrected soils
lani et al., 2022a), leading to greater fertilizer absorption capacity (Karaivazoglou et al., 2007). Although nitrification is more likely to

8
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

Fig. 6. 15N retention in seven stratified soil

layers to a depth of 100 cm after the (A) first
(maize + ruzigrass) and (B) second (soybean)
growing seasons in response to lime rate [con­
trol (no liming), half the recommended lime
rate (½ RLR), recommended lime rate (1 RLR)
and twice the recommended lime rate (2 RLR)].
Different lowercase or capital letters indicate
significant differences between treatments by
Fisher’s protected LSD test at p ≤ 0.05. Error
bars express the standard error of the mean
(n = 4).

increase in soils that are managed with high lime rates (Holland et al., it comes to a leguminous plant with high BNF efficiency, such as soy­
2018; Bossolani et al., 2020a), increasing the tendency for N losses bean. Second, most of the N present in soybean plant tissues (~90%)
through leaching and denitrification (Tamagno et al., 2022), such soils comes from BNF (Freitas et al., 2022), which is improved by liming
are also more fertile and boost crop root growth (Rellán-Álvarez et al., (Alves et al., 2021). Our results also showed that only 2% of the total N
2016). Under these conditions, most of the N present in the soil (NH+ 4 or in soybean shoots was derived from fertilizer, whereas 98% was derived
NO-3) is quickly absorbed by the plants, reducing their propensity to lose from other sources (presumed to come predominantly from BNF, and to
N to the environment (Caires et al., 2016; Rosolem et al., 2017). The a lesser extent from SOM mineralization). Liming increases soybean root
results of the analysis of unrecovered 15N fertilizer during the maize growth (Bossolani et al., 2021b) and, in turn, the area available for
+ ruzigrass season are consistent with these effects. Soils managed with nodulation by Bradyrhizobium sp. (Alves et al., 2021). The higher
2 RLR and, to a lesser extent, 1 RLR presented the lowest potential N availability of nutrients in soils managed with lime [e.g., P, Ca, Mg,
losses, whereas unrecovered 15N was highest in the control (no (Costa et al., 2018)] also increases the efficiency of nodules in con­
amendments applied) and at ½ RLR. verting atmospheric N2 into ureides. The combination of these effects
Nitrogen crop demand can be met by supplying inorganic N (fertil­ may have reduced the dependence of soybean on fertilizer-derived 15N.
izers) and/or through N mineralization from SOM (SOM-N) (Chen et al., Several studies have shown that the abundance (Andrade et al., 2002)
2014). Based on these assumptions, it is important to emphasize that the and activity (Zhalnina et al., 2013) of Rhizobia species are lower in
increase in biomass production of maize and ruzigrass plants was not highly acidic soil, leading to lower BNF capacity of soybean.
totally dependent on the N derived from the fertilizer. Most of the N Interestingly, after the soybean harvest, 15N fertilizer remaining was
accumulated in stover and grains was derived from the soil, especially highest in soil amended with 2 RLR, followed by 1 RLR, ½ RLR, and the
from the SOM-N fraction (Dourado-Neto et al., 2010). These authors control. This pattern is consistent with the pattern of unrecovered 15N
confirmed that N from the net N mineralization of SOM was the domi­ fertilizer during the agricultural year (maize + ruzigrass/soybean sea­
nant source of N in the crops. In this way, N fertilizer functions as a sons). When 15N recovery by crops is greater, fertilizer loss to the
readily available source for plant uptake, while SOM-derived N is environment is reduced. 15N fertilizer losses were ~71% higher in the
released throughout the crop cycle (Mwafulirwa et al., 2017). Therefore, control (highest loss, 54%) than at 2 RLR (lowest loss, 31%).
long-term management with lime is a important tool that increases the
crop N use efficiency from mineral N fertilizers (Crusciol et al., 2022a), 4.3. Stratified 15
N retention in the soil profile
and increases the increment of SOM in the system by increasing the
production of biomass by the roots and shoots of cultivated plants In addition to the amount of 15N fertilizer remaining in the soil
(Bossolani et al., 2022a; b), as supported by our results of SOC contents. (down to a depth of 100 cm), its distribution in the soil profile is an
As a consequence, SOM will become a natural reservoir of N and other important factor in better understanding the dynamics of N in soil
elements in the soil, in addition to several other benefits linked to soil managed with liming. According to our results, as the lime rate and,
quality (Nunes et al., 2018). consequently, soil fertility increased, maize and ruzigrass, both highly
Crop rotation of maize (intercropped or not with forage grasses) and N-demanding cereals (Omara et al., 2019), became more able to recover
soybean is the most common model in Brazil (Ceccon et al., 2013). Here, high amounts of 15N fertilizer (based on data from 15N recovery) and
we showed that managing soil with 2 RLR led to higher production of store them in aboveground crop residues. After nutrient cycling, this 15N
biomass (stover and grain) and greater recovery of 15N fertilizer by crops is available in the uppermost soil layers, as supported by our 15N
during the maize + ruzigrass season (and consequently, less unrecov­ retention results. In addition, due to the improvement in the soil profile
ered 15N); by contrast, 15N recovery by soybean was highest in the (i.e., higher availability of nutrients and lower levels of Al3+), root
control and at ½ RLR. There are two main explanations for the opposing development can also increase (Rosolem et al., 2017), enabling the ab­
results for these crops. First, because 15N fertilizer recovery was higher sorption of high amounts of N (mostly as NO-3) (Yu et al., 2015) and
during the first season (part exported by grain, part retained in crop reducing the N concentration in intermediate soil layers (5–20 cm).
residues), a smaller amount of fertilizer remained in the soil to be Beginning with the 20–40 cm layer (in which 15N retention did not differ
absorbed by soybean in succession. Nitrogen fertilizer recovery by the significantly among the treatments), the amount of 15N fertilizer in
successor crop is usually low (Smith and Chalk, 2018), especially when deeper layers tended to be the highest in soil managed with ½ RLR. The

9
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

same trend was observed after soybean cultivation in the control and ½ Writing – review & editing. Andressa Selestina Dalla Côrt: Visualiza­
RLR treatments. These results have an important implication for tropical tion, Writing – review & editing. Vladimir Eliodoro Costa: Isotopic
agricultural systems: applying low lime rates may increase N leaching analysis, Visualization, Writing – review & editing. Heitor Cantarella:
(probably as NO-3). The surface application of lime at low rates (e.g., ½ Visualization, Validation, Writing – review & editing.
RLR) increases soil pH in the surface layers (where N fertilizer is applied
and the release of N forms is greatest after nutrient cycling) and, Declaration of Competing Interest
consequently, soil nitrification rates (Beeckman et al., 2018; Holland
et al., 2018) in the same manner as higher lime rates. Liming has been The authors declare that they have no known competing financial
shown to potentiate soil N transformation by microorganisms (Bossolani interests or personal relationships that could have appeared to influence
-
et al., 2020a), thereby regulating the rates of conversion of NH+ 4 to NO3 the work reported in this paper.
by nitrification. However, in the present study, soil fertility at depths
below 10 cm was lower when ½ RLR was applied. As a consequence, the Data availability
NO-3 generated by nitrification in the uppermost soil layers was absorbed
in smaller amounts than in soils corrected with higher lime rates, given Data will be made available on request.
that root growth and distribution are better in the fertile soils, as
observed by Bossolani et al. (2022) in a study of the root development of Acknowledgments
soybean and maize in the same experimental area.
During crop development, as long as there are no limitations on root This study was supported by the São Paulo Research Foundation
development and/or water restriction, the NO–3 generated by nitrifica­ (FAPESP; grant numbers 2018/11063–7 and 2019/12764–1) and the
tion is taken up by crops; conversely, when N uptake is low, there is a National Council for Scientific and Technological Development (CNPq;
risk of NO−3 leaching (Holland et al., 2018). Plants established in soils Universal Research Project: 421637/2018–8). In addition, CACC
corrected with 1 RLR and, in particular, 2 RLR produced higher amounts received a research productivity fellowship (PQ) from the CNPq
of biomass (grain and/or straw) and recovered greater amounts of 15N (310535/2021–2).
fertilizer than plants grown in less fertile soils (control and ½ RLR). This
indicates that a higher amount of 15N fertilizer was lost to the envi­ Appendix A. Supporting information
ronment (supported by unrecovered 15N data) and that part of the N
fertilizer may have been leached (high amounts of 15N in deep layers) Supplementary data associated with this article can be found in the
and/or denitrified (Zhou et al., 2021). Although the supply of N is online version at doi:10.1016/j.fcr.2023.108971.
greater in less fertile soils due to lower crop uptake, the remaining N is
liable to loss by leaching and denitrification, processes that are harmful References
to the environment and indicate low sustainability of the agricultural
system. Despite the strong evidence validating the hypotheses raised Alves, L.A., Ambrosini, V.G., Denardin, L.G., de, O., Flores, J.P.M., Martins, A.P.,
Filippi, D., Bremm, C., Carvalho, P.C., de, F., Farias, G.D., Ciampitti, I.A., Tiecher, T.,
here, more studies must be carried out to better understand the dy­ 2021. Biological N2 fixation by soybeans grown with or without liming on acid soils
namics of N in soils treated with low lime rates. in a no-till integrated crop-livestock system. Soil Tillage Res. 209.
Anderson, G.C., Pathan, S., Easton, J., Hall, D.J.M., Sharma, R., 2020. Short- And long-
term effects of lime and gypsum applications on acid soils in a water-limited
5. Conclusions environment: 1. Grain yield response and nutrient concentration. Agronomy 10,
1–15.
Lime application to the soil surface under a no-till system amelio­ Andrade, D.S., Murphy, P.J., Giller, K.E., 2002. Effects of liming and legume/cereal
cropping on populations of indigenous rhizobia in an acid Brazilian Oxisol. Soil Biol.
rated subsoil acidity by increasing soil pH, base saturation and reducing Biochem. 34, 477–485.
Al3+ toxicity. The effects were proportional to the lime rates used (2 RLR Baptistella, J.L.C., de Andrade, S.A.L., Favarin, J.L., Mazzafera, P., 2020. Urochloa in
> 1 RLR > ½ RLR > control). In addition, higher lime rates increased C tropical agroecosystems. Front. Sustain. Food Syst. 4, 119.
Beeckman, F., Motte, H., Beeckman, T., 2018. Nitrification in agricultural soils: impact,
stock on soil profile, even at 1 m depth. Maize intercropped with ruzi­
actors and mitigation. Curr. Opin. Biotechnol. 50, 166–173. https://doi.org/
grass and soybean grown in fertile soils by the application of high lime 10.1016/j.copbio.2018.01.014.
rates presented higher aboveground biomass production (grains and/or Bender, R.R., Haegele, J.W., Ruffo, M.L., Below, F.E., 2013. Nutrient uptake,
straw), which led to an increase in the recovery of 15N-labeled fertilizer, partitioning, and remobilization in modern, transgenic insect-protected maize
hybrids. Agron. J. 105, 161–170.
mainly by maize and by ruzigrass. Soybean recovered a small amount of Bender, R.R., Haegele, J.W., Below, F.E., 2015. Nutrient uptake, partitioning, and
15
N, regardless of treatment, but larger amounts were found in plants remobilization in modern soybean varieties. Agron. J. 107, 563–573.
established in acidic soils (control and ½ RLR). The final balance of Bertol, F.D., Martins, A.P., Denardin, L.G. de O., Kunrath, T.R., Souza Filho, W. de,
Goulart, M.W., Carvalho, P.C., de, F., Anghinoni, I., 2022. Liming and grazing
potential 15N-fertilizer losses (total 15N unrecovered) showed that the intensities effects on soil mineral nitrogen throughout the pasture cycle in a
higher the recovery of 15N-fertilizer, the smaller the amount that can be subtropical integrated crop-livestock system. Rev. Bras. De. Ciência do Solo 46.
lost to the environment. Interestingly, this study showed that large Blake, G.R., Hartge, K.H., 1986. Bulk density. Methods Soil Anal.: Part 1 Phys. Mineral.
Methods 5, 363–375.
amounts of 15N-fertilizer were found in deep soil layers when ½ RLR was Bossolani, J.W., Lazarini, E., Santos, F.L., Sanches, I.R., Meneghette, H.H.A., Parra, L.F.,
applied, indicating that low lime rates, or even its absence, can increase Souza, L.G.M., 2018. Surface Reapplication of Lime and Gypsum on Maize Cultivated
losses (leaching) of unabsorbed N-fertilizer by the crops, resulting in low Sole and Intercropped with Urochloa. Commun. Soil Sci. Plant Anal. 49, 1855–1868.
https://doi.org/10.1080/00103624.2018.1475565.
sustainability of agricultural systems. Bossolani, J.W., Crusciol, C.A.C., Merloti, L.F., Moretti, L.G., Costa, N.R., Tsai, S.M.,
Kuramae, E.E., 2020a. Long-term lime and gypsum amendment increase nitrogen
CRediT authorship contribution statement fixation and decrease nitrification and denitrification gene abundances in the
rhizosphere and soil in a tropical no-till intercropping system (At: Geoderma 375,
114476 https://doi.org/10.1016/j.geoderma.2020.114476.
João William Bossolani: Conceptualization, Investigation, Data Bossolani, J.W., Santos, F.L., Meneghette, H.H.A., Sanches, I.R., Moretti, L.G., Parra, L.F.,
curation, Formal analysis, Writing – original draft. Carlos Alexandre Lazarini, E., 2020b. Soybean in Crop Rotation with Maize and Palisade Grass
Intercropping Enhances the Long-term Effects of Surface Liming in No-till System.
Costa Crusciol: Project administration, Funding acquisition, Supervi­
J. Soil Sci. Plant Nutr. 20, 1–12.
sion. Eduardo Mariano: Formal analysis, Visualization, Writing – re­ Bossolani, J.W., Crusciol, C.A.C., Leite, M.F.A., Merloti, L.F., Moretti, L.G., Pascoaloto, I.
view & editing. Luiz Gustavo Moretti: Investigation, Validation, M., Kuramae, E.E., 2021a. Modulation of the soil microbiome by long-term Ca-based
Writing – review & editing. José Roberto Portugal: Investigation, soil amendments boosts soil organic carbon and physicochemical quality in a
tropical no-till crop rotation system. Soil Biol. Biochem. 156, 108188.
Validation, Writing – review & editing. Mariley Fonseca: Investigation, Bossolani, J.W., Crusciol, C.A.C., Portugal, J.R., Moretti, L.G., Garcia, A., Rodrigues, V.
Validation, Writing – review & editing. Letusa Momesso: Visualization, A., Fonseca, M., de, C., da, Bernart, L., Vilela, R.G., Mendonça, L.P., Reis, A.R.,

10
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

2021b. Long-term liming improves soil fertility and soybean root growth, reflecting Jones, D.L., Cross, P., Withers, P.J.A., DeLuca, T.H., Robinson, D.A., Quilliam, R.S.,
improvements in leaf gas exchange and grain yield. Eur. J. Agron. 128, 126308. Harris, I.M., Chadwick, D.R., Edwards-Jones, G., 2013. Nutrient stripping: the global
Bossolani, J.W., Crusciol, C.A.C., Momesso, L., Portugal, J.R., Moretti, L.G., Garcia, A., disparity between food security and soil nutrient stocks. J. Appl. Ecol. 50, 851–862.
Fonseca, M. de C. da, Rodrigues, V.A., Calonego, J.C., Andre, 2022a. Surface liming Joris, H.A.W., Caires, E.F., Scharr, D.A., Bini, Â.R., Haliski, A., 2016. Liming in the
triggers improvements in subsoil fertility and root distribution to boost maize crop conversion from degraded pastureland to a no-till cropping system in Southern
physiology, yield and revenue (At: Plant Soil 472, 1–23. https://doi.org/10.1007/ Brazil. Soil Tillage Res. 162, 68–77.
s11104-022-05432-2. Karaivazoglou, N.A., Tsotsolis, N.C., Tsadilas, C.D., 2007. Influence of liming and form of
Bossolani, J.W., Crusciol, C.A.C., Moretti, L.G., Garcia, A., Portugal, J.R., Bernart, L., nitrogen fertilizer on nutrient uptake, growth, yield, and quality of Virginia (flue-
Vilela, R.G., Caires, E.F., Amado, T.J.C., Calonego, J.C., dos Reis, A.R., 2022b. cured) tobacco. Field Crops Res. 100, 52–60.
Improving soil fertility with lime and phosphogypsum enhances soybean yield and Levene, H., 1960. Robust tests for equality of variances. In: Olkin, I., Ghurye, S.G.,
physiological characteristics. Agron. Sustain. Dev. 42. Hoeffding, W., Madow, W.G., Mann, H.B. (Eds.), Contributions to probability and
Briedis, C., Sá, J.C., de, M., Caires, E.F., Navarro, J., de, F., Inagaki, T.M., Boer, A., statistics: Essays in …. Stanford University Press, pp. 278–292.
Neto, C.Q., Ferreira, A., de, O., Canalli, L.B., Santos, J.B. dos, 2012. Soil organic Li, Y., Cui, S., Chang, S.X., Zhang, Q., 2019. Liming effects on soil pH and crop yield
matter pools and carbon-protection mechanisms in aggregate classes influenced by depend on lime material type, application method and rate, and crop species: a
surface liming in a no-till system (At: Geoderma 170, 80–88. https://doi.org/ global meta-analysis. J. Soils Sediment. 19, 1393–1406.
10.1016/j.geoderma.2011.10.011. Limousin, G., Tessier, D., 2007. Effects of no-tillage on chemical gradients and topsoil
Caires, E.F., Zardo Filho, R., Barth, G., Joris, H.A.W., 2016. Optimizing Nitrogen Use acidification. Soil Tillage Res. 92, 167–174.
Efficiency for No-Till Corn Production by Improving Root Growth and Capturing Lu, J., Hu, T., Zhang, B., Wang, L., Yang, S., Fan, J., Yan, S., Zhang, F., 2021. Nitrogen
NO3-N in Subsoil. Pedosphere 26, 474–485. fertilizer management effects on soil nitrate leaching, grain yield and economic
Calonego, J.C., Rosolem, C.A., 2010. Soybean root growth and yield in rotation with benefit of summer maize in Northwest China. Agric. Water Manag. 247, 106739.
cover crops under chiseling and no-till. Eur. J. Agron. 33, 242–249. Meng, C., Tian, D., Zeng, H., Li, Z., Yi, C., Niu, S., 2019. Global soil acidification impacts
Carmeis Filho, A.C.A., Crusciol, C.A.C., Castilhos, A.M., 2017a. Liming demand and plant on belowground processes. Environ. Res. Lett. 14, 074003.
growth improvements for an Oxisol under long-term no-till cropping. J. Agric. Sci. Mwafulirwa, L.D., Baggs, E.M., Russell, J., Morley, N., Sim, A., Paterson, E., 2017.
155, 1093–1112. Combined effects of rhizodeposit C and crop residues on SOM priming, residue
Carmeis Filho, A.C.A., Penn, C.J., Crusciol, C.A.C., Calonego, J.C., 2017b. Lime and mineralization and N supply in soil (At:). Soil Biol. Biochem. 113, 35–44. https://
phosphogypsum impacts on soil organic matter pools in a tropical Oxisol under long- doi.org/10.1016/j.soilbio.2017.05.026.
term no-till conditions (At: Agric., Ecosyst. Environ. 241, 11–23. https://doi.org/ Nelson, D.W., Sommers, L.E., 1996. Total carbon, organic carbon, and organic matter.
10.1016/j.agee.2017.02.027. Methods Soil Anal.: Part 3 Chem. Methods 5, 961–1010.
Ceccon, G., Staut, L.A., Sagrilo, E., Machado, L.A.Z., Nunes, D.P., Alves, V.B., 2013. Nelson, L.S., 1998. The Anderson-Darling test for normality. J. Qual. Technol. 30, 298.
Legumes and forage species sole or intercropped with corn in soybean-corn Nunes, M.R., van Es, H.M., Schindelbeck, R., Ristow, A.J., Ryan, M., 2018. No-till and
succession in midwestern Brazil. Rev. Bras. De. Ciência do Solo 37, 204–212. cropping system diversification improve soil health and crop yield (At). Geoderma
Chen, B., Liu, E., Tian, Q., Yan, C., Zhang, Y., 2014. Soil nitrogen dynamics and crop 328, 30–43. https://doi.org/10.1016/j.geoderma.2018.04.031.
residues. A review. Agron. Sustain. Dev. 34, 429–442. Omara, P., Aula, L., Oyebiyi, F., Raun, W.R., 2019. World cereal nitrogen use efficiency
Costa, C.H.M., Carmeis Filho, A.C.A., Crusciol, C.A.C., Soratto, R.P., Guimarães, T.M., trends: review and current knowledge. Agrosystems. Geosci. Environ. 2, 1–8.
2018. Intensive annual crop production and root development in a tropical acid soil Parker, D.R., Kinraide, T.B., Zelazny, L.W., 1988. Aluminum speciation and phytotoxicity
under long-term no-till and soil-amendment management. Crop Pasture Sci. 69, in dilute hydroxy-aluminum solutions. Soil Sci. Soc. Am. J. 52, 438–444.
488–505. Patra, A., Sharma, V.K., Nath, D.J., Ghosh, A., Purakayastha, T.J., Barman, M., Kumar, S.,
Costa, N.R., Andreotti, M., Crusciol, C.A.C., Pariz, C.M., Bossolani, J.W., Pascoaloto, I.M., Chobhe, K.A., Anil, A.S., Rekwar, R.K., 2021. Impact of soil acidity influenced by
da Rocha Lima, C.G., dos Santos Batista Bonini, C., de Castilhos, A.M., Calonego, J. long-term integrated use of enriched compost, biofertilizers, and fertilizer on soil
C., 2021. Soybean yield and nutrition after tropical forage grasses. Nutr. Cycl. microbial activity and biomass in rice under acidic soil. J. Soil Sci. Plant Nutr. 21,
Agroecosyst. 121, 31–49. 756–767.
Crusciol, C.A.C., Marques, R.R., Carmeis Filho, A.C.A., Soratto, R.P., Costa, C.H.M., van Raij, B., Cantarella, H., Quaggio, J.A. & Furlani, A.M.C., 1997, Recomendações da
Ferrari Neto, J., Castro, G.S.A., Pariz, C.M., Castilhos, A.M., Franzluebbers, A.J., adubação e calagem para o Estado de São Paulo. Instituto Agronômico/Fundação
2019. Lime and gypsum combination improves crop and forage yields and estimated IAC Campinas.
meat production and revenue in a variable charge tropical soil. Nutr. Cycl. van Raij, B., Quaggio, J.A., Cantarella, H., Abreu, C.A., 2001. Os métodos de análise
Agroecosyst. 115, 347–372. química do sistema IAC de análise de solo no contexto nacional. Análise Quím. Para.
Crusciol, C.A.C., Bossolani, J.W., Portugal, J.R., Moretti, L.G., Momesso, L., Campos, M., Aval. da Fertil. De. solos Trop. 5–39.
Costa, N.R., Volf, M.R., Calonego, J.C., Rosolem, C.A., 2022a. Exploring the Reis, A.R. dos, Lisboa, L.A.M., Reis, H.P.G., Barcelos, J.P. de Q., Santos, E.F., Santini, J.M.
synergism between surface liming and nitrogen fertilization in no-till system. Agron. K., Venâncio Meyer-Sand, B.R., Putti, F.F., Galindo, F.S., Kaneko, F.H., Barbosa, J.Z.,
J. 1–16. Paixão, A.P., Junior, E.F., de Figueiredo, P.A.M., Lavres, J., 2018. Depicting the
Crusciol, C.A.C., Bossolani, J.W., Portugal, J.R., Moretti, L.G., Momesso, L., de physiological and ultrastructural responses of soybean plants to Al stress conditions.
Campos, M., Costa, N.R., Volf, M.R., Calonego, J.C., Rosolem, C.A., 2022b. Exploring Plant Physiol. Biochem. 130, 377–390. https://doi.org/10.1016/j.
the synergism between surface liming and nitrogen fertilization in no-till system. plaphy.2018.07.028.
Agron. J. 113, 1–10. Rellán-Álvarez, R., Lobet, G., Dinneny, J.R., 2016. Environmental Control of Root System
Cunningham, C., 2020. Characterization of dry spells in Southeastern Brazil during the Biology. Annu. Rev. Plant Biol. 67, 619–642.
monsoon season. Int. J. Climatol. 40, 4609–4621. Rosolem, C.A., Ritz, K., Cantarella, H., Galdos, M.V., Hawkesford, M.J., Whalley, W.R.,
Dourado-Neto, D., Powlson, D., Bakar, R.A., Bacchi, O.O.S., Basanta, M.V., Cong, P., thi, Mooney, S.J., 2017. Enhanced Plant Rooting and Crop System Management for Improved
Keerthisinghe, G., Ismaili, M., Rahman, S.M., Reichardt, K., Safwat, M.S.A., N Use Efficiency. 1st ed. Elsevier Inc,. https://doi.org/10.1016/bs.
Sangakkara, R., Timm, L.C., Wang, J.Y., Zagal, E., van Kessel, C., 2010. Multiseason agron.2017.07.002.
Recoveries of Organic and Inorganic Nitrogen-15 in Tropical Cropping Systems. Soil Schnitkey, G., Paulson, N., Zulauf, C., Swanson, K., Colussi, J., Baltz, J., 2022. Nitrogen
Sci. Soc. Am. J. 74, 139–152. Fertil. Prices. Supply Light Ukr. -Russ. Confl. Farm. Dly. 12.
Esper Neto, M., Lara, L.M., Maciel de Oliveira, S., Santos, R.F. dos, Braccini, A.L., Šiaudinis, G., Jasinskas, A., Karčauskienė, D., Repšienė, R., 2020. The effect of liming and
Inoue, T.T., Batista, M.A., 2021. Nutrient Removal by Grain in Modern Soybean nitrogen application on common osier and black poplar biomass productivity and
Varieties. Front. Plant Sci. 12, 1–14. determination of biofuel quality indicators. Renew. Energy 152, 1035–1040.
Fageria, N.K., Moreira, A., 2011. The Role of Mineral Nutrition on Root Growth of Crop Smith, C.J., Chalk, P.M., 2018. The residual value of fertiliser N in crop sequences: An
Plants. 1st ed. Elsevier Inc, (At: https://doi.org/10.1016/B978-0-12-385531- appraisal of 60 years of research using 15N tracer (At:). Field Crops Res. 217, 66–74.
2.00004-9.). https://doi.org/10.1016/j.fcr.2017.12.006.
Fageria, N.K. & Nascente, A.S., 2014, Management of soil acidity of South American soils Song, X.-D., Yang, F., Ju, B., Li, D.-C., Zhao, Y.-G., Yang, J.-L., Zhang, G.-L., 2018. The
for sustainable crop production. Elsevier. (At: https://doi.org/10.1016/B978–0- influence of the conversion of grassland to cropland on changes in soil organic
12–802139-2.00006–8.). carbon and total nitrogen stocks in the Songnen Plain of Northeast China. Catena
Fehr, W.R. & Caviness, C.E. 1977. Stages of soybean development. 171, 588–601.
Freitas, V.F., de, Cerezini, P., Hungria, M., Nogueira, M.A., 2022. Strategies to deal with Tamagno, S., Eagle, A.J., McLellan, E.L., van Kessel, C., Linquist, B.A., Ladha, J.K.,
drought-stress in biological nitrogen fixation in soybean. Appl. Soil Ecol. 172, Pittelkow, C.M., 2022. Quantifying N leaching losses as a function of N balance: A
104352 https://doi.org/10.1016/j.apsoil.2021.104352. path to sustainable food supply chains. Agric., Ecosyst. Environ. 324, 107714.
Holland, J.E., Bennett, A.E., Newton, A.C., White, P.J., McKenzie, B.M., George, T.S., Tiritan, C.S., Büll, L.T., Crusciol, C.A.C., Carmeis Filho, A.C.A., Fernandes, D.M.,
Pakeman, R.J., Bailey, J.S., Fornara, D.A., Hayes, R.C., 2018. Liming impacts on Nascente, A.S., 2016. Tillage system and lime application in a tropical region: Soil
soils, crops and biodiversity in the UK: A review. Sci. Total Environ. 610–611, chemical fertility and corn yield in succession to degraded pastures (At:). Soil Tillage
316–332 (At: https://doi.org/10.1016/j.scitotenv.2017.08.020.). Res. 155, 437–447. https://doi.org/10.1016/j.still.2015.06.012.
Inagaki, T.M., de Moraes Sá, J.C., Caires, E.F., Gonçalves, D.R.P., 2016. Lime and gypsum
application increases biological activity, carbon pools, and agronomic productivity
in highly weathered soil. Agric., Ecosyst. Environ. 231, 156–165. https://doi.org/
10.1016/j.agee.2016.06.034.
Inagaki, T.M., de Moraes Sá, J.C., Caires, E.F., Gonçalves, D.R.P., 2017. Why does carbon
increase in highly weathered soil under no-till upon lime and gypsum use? Sci. Total
Environ. 599–600, 523–532. https://doi.org/10.1016/j.scitotenv.2017.04.234.

11
J.W. Bossolani et al. Field Crops Research 299 (2023) 108971

Yu, P., Li, X., White, P.J., Li, C., 2015. A large and deep root system underlies high nitrososphaera and bradyrhizobium are inversely correlated and related to
nitrogen-use efficiency in maize production. PLoS ONE 10, 1–17. agricultural practices in long-term field experiments. Front. Microbiol. 4, 1–13.
Zhalnina, K., De Quadros, P.D., Gano, K.A., Davis-Richardson, A., Fagen, J.R., Brown, C. Zhou, M., Ying, S., Chen, J., Jiang, P., Teng, Y., 2021. Effects of biochar-based fertilizer
T., Giongo, A., Drew, J.C., Sayavedra-Soto, L.A., Arp, D.J., Camargo, F.A.O., on nitrogen use efficiency and nitrogen losses via leaching and ammonia
Daroub, S.H., Clark, I.M., McGrath, S.P., Hirsch, P.R., Triplett, E.W., 2013. Ca. volatilization from an open vegetable field. Environ. Sci. Pollut. Res. 28,
65188–65199.

12