Professional Documents
Culture Documents
DOI: 10.1002/agj2.20324
ARTICLE
S o i l Fe r t i l i t y & C ro p N u t r i t i o n
remains high even after decades of fertilization (Roy et al., which was soluble in a 2% citric acid solution, considering
2017). a ground phosphate (<0.063 mm) to extraction solution of
This work sought to evaluate the effects of different soil P 1:100. The TSP contained 20.8% total P, 92% of which was
correction and maintenance fertilization strategies on the soluble in a 2% citric acid solution.
response of a soybean–corn rotation system, cultivated in Before installing the experiment, samples were taken
an Oxisol in central Brazil managed under a no-tillage sys- for soil analysis in 12 plots, four from each of the three
tem (NT). blocks, with 20 subsamples per plot in the 0- to 20-cm layer
(Table 2). Soil chemical deficiencies, except for P, were
then interpreted and corrected for according to Sousa and
2 MATERIALS AND METHODS Lobato (2004) with the application of dolomitic lime (65%
relative total neutralizing power, RTNP) to raise cation
2.1 Characterization of the exchange saturation to 50%, 75 kg ha−1 of K in the form of
experimental area potassium chloride, 75 kg ha−1 of S in the form of agricul-
tural gypsum and 100 kg ha−1 of FTE BR-12 as a micronu-
The experiment was conducted in the experimental area trient source (Table 3).
of Embrapa Cerrados, in Planaltina, DF, Brazil (15◦ 36ť S, The soil was plowed and harrowed to incorporate the
47◦ 42ť W). The climate is classified as Cwa according to fertilizers in the total experimental area to 20-cm deep at
the Köppen classification (Alvares, Stape, Sentelhas, De the end of 1998. Before planting the first crop in 1999, TSP
Moraes Gonçalves, & Sparovek, 2013), with annual precip- and RRP (105 kg ha−1 of P) were applied to plots according
itation and temperature means of 1570 mm and 21.3 ◦ C, to the treatments that would receive corrective P fertiliza-
respectively. The elevation is 1,014 m, with smooth undu- tion, then proceeding with another harrowing operation.
lating relief and the natural vegetation is that of Cerrado. After this last soil preparation, the experiment was then
Soil is characterized as a clayey Oxisol (Rhodic Haplus- conducted under no-tillage (NT).
tox) (USDA-NRCS, 2003), with 54% clay, whose mineralog- The annual maintenance dose of 35 kg of total P ha−1
ical composition in the diagnostic horizon is dominated by was band-applied to the crop furrow or broadcast, depend-
kaolinite, gibsite, and hematite. ing on the treatments. The furrows were opened with a
no-till planter to allow for manual application of fertilizers
at about 5-cm deep. In treatments that received broadcast
2.2 Experimental design and application, fertilizers were uniformly distributed by hand
management in the respective plots after planting all treatments.
Corn (Zea mays L.) was sown manually, using two seeds
The experiment was installed in June 1999, where the per position, to guarantee the germination of at least one
effects of P sources and application methods were evalu- plant. Plants were thinned when necessary to guarantee a
ated on a soybean−corn rotation system, with one main plant population of 70,000 plants ha−1 . Soybean [Glycine
crop per year (during summer) followed by a cover crop max (L.) Merr.] was sown using a portable machine, to
in winter (millet). Before the first crop, three initial con- establish 17 to 25 plants per meter, according to the vari-
ditions of P in soil were defined: corrective P fertilization ety used. For planting millet [Pennisetum glaucum (L.)
with triple superphosphate (TSP) or reactive rock phos- R.Br.], a traditional no-till planter was used, with a spac-
phate (RRP) (both at the rate of 105 kg ha−1 of total P) ing of 0.20 m between rows, configured to distribute 20 kg
and a control which had no P application (natural condi- seeds ha−1 .
tion). From 1999 to 2014, maintenance P applications were As for maintenance fertilization with other nutrients,
conducted with 35 kg P ha−1 in each main crop as TSP, 67 kg K ha−1 were applied in the form of potassium chlo-
Gafsa RRP, or a mixture of both in equal P parts, plus ride to every summer crop. To soybean, Co and Mo were
controls without these maintenance applications. Main- added as seed dressing. No application of N was nec-
tenance phosphate fertilizers were evaluated under two essary in the soybean crop due to inoculation of seeds
application strategies: broadcast spreading of the fertil- with Bradyrhizobium. However, in corn 30 kg of N ha−1
izer on the soil surface or band applying it in the crop were applied to the sowing furrow in addition to two top-
row furrow, about 5 cm below the seeds. Thus, the exper- dressing applications of 60 kg ha−1 each, always using urea
imental design was a complete factorial (3 × 3 × 2 + con- as source. In order to amend the soil profile, 1 t ha−1 of agri-
trols) arranged in randomized blocks, with three replicates cultural gypsum (20% Ca, 15% S, 0.2% P) was applied annu-
(Table 1). ally between the fifth and seventh crops, totaling 3 t ha−1 ,
The RRP fertilizer particles were mostly between 0.5 and as indicated for soil correction up to 80-cm deep (Sousa &
2.8 mm in diameter and contained 12.3% total P, 44% of Lobato, 2004) (Table 3).
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4306 ZANCANARO DE OLIVEIRA et al.
TA B L E 2 Chemical and texture characteristics of the soil prior to liming and fertilization of the experimental area in the 0- to 20-cm layer
K+ Ca+2 Mg+2 Al+3
a b b c c c d
pH H2 O P H + Al CEC
−1 −1
mg kg cmolc kg
4.5 1.2 0.1 0.2 0.2 1.4 8.1 8.6
OM V Clay Silt Coarse sand Fine sand
g kg−1 %
2.8 5.6 54 5 12 29
Note. OM: organic matter by the Walkley−Black procedure; V: cation exchange saturation; CEC: cation exchange capacity.
a
1:2.5 soil solution ratio.
b
Mehlich-1.
c
KCl 1 mol L−1 .
d
calcium acetate 1 mol L−1 in pH 7.0.
Agricultural gypsum reapplications for S supply were was 0.45 m for soybean, 0.75 cm for corn, and 0.20 m for
as follows: 20 kg ha−1 of S in 2006 and 15 kg S ha−1 yr−1 millet. Supplementary sprinkler irrigation was performed
from 2007 onward. Lime reapplications occurred in 2006 when tensiometer readings installed at a depth of 20 cm
(975 kg ha−1 to reach 50% cation exchange saturation) and in plots indicated a value greater than 45 kPa, aiming at
in 2014 (2,158 kg ha−1 ). maintaining the productive potential.
The first three main crops were soybean (1999/2000, In order to better evaluate and discuss the effects of
2000/2001, and 2001/2002), followed by a rotation between treatments, in some cases yields were calculated relative
corn and soybean, always with millet as a winter cover to the treatment most widely adopted by farmers in the
crop, until the 16th and last crop (corn). Experimental plots Cerrado region, which is demonstrated in Treatment 10
measured 49.5 mš (11 by 4.5 m), and the inter-row distance (Table 1): use of a soluble P source in corrective and
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZANCANARO DE OLIVEIRA et al. 4307
maintenance fertilization, band applying the fertilizer in better considers the localized effects of band-applying P
maintenance applications. Therefore, crop yields for the fertilizers.
different treatments were expressed as a percentage of the The STP was determined using a Mehlich-1 (0.05 M
reference treatment. HCl + 0.025 M H2 SO4 ) and Bray-1 (HCl 0.025 M + NH4 F
0.03 M) (Bray & Kurtz, 1945) solutions, and the content was
assessed by colorimetric analysis (Murphy & Riley, 1962).
2.3 Soil sampling and analysis
Soil samples were taken with a Dutch auger in the 0- to 2.4 Residual phosphorus stocks
20-cm layer. These samplings were carried out after har-
vesting the following crops: 2nd to 6th and 11th to 16th, The amounts of P applied to the soil before the first har-
in treatments fertilized exclusively with TSP or RRP. For vest can be separated into four groups: the control with-
plots with broadcast P fertilization, 20 subsamples were out any application of P (Treatment 1); those without initial
taken randomly in the plot to make up the composite sam- P correction but with annual maintenance inputs accord-
ple. For plots with band-applied P fertilizer, sampling was ing to the different sources and modes of application, char-
performed according to Nicolodi, Anghinoni, and Salet acterizing a gradual corrective fertilization (Treatments 2–
(2002). This method involves sampling from seven points 7); the controls with application of corrective fertilization
across a crop row: one right over the row, and three equally using TSP or RRP, but without maintenance fertilization
spaced to each side, to the center of the corn inter-row. (Treatments 8 and 15, respectively); and finally treatments
Soil from each of the seven positions in a given layer was with both corrective and maintenance fertilization (Treat-
then mixed to form one subsample per layer. This proce- ments 9–14 and 16–21, or treatments under total correc-
dure was repeated in three crop rows to form three sub- tive fertilization). Thus, before the first crop 0, 35, 105, and
samples per plot, which were mixed to form the plot com- 140 kg P ha−1 , respectively, were applied to these treatment
posite sample. This methodology was adopted because it sets (Table 1).
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4308 ZANCANARO DE OLIVEIRA et al.
In each harvest after the second, the residual P stocks in The model identity analysis was used to verify the pos-
soil before each cultivation were calculated from the differ- sibility of creating a unique model for each phosphate
ence between all P inputs via fertilizers or soil condition- source, which involved the two different crops grown in
ers and export by the harvested grains up to that moment the experiment (soybean and corn). Similarly, the possi-
(P input–output budget). For this, the P content in the bility of creating a model for each crop, involving the two
grains was evaluated by wet digestion with HNO3 + HClO4 P sources was also tested. Through the likelihood ratio
(3:1, v:v) (Embrapa, 1999), and then multiplied by the grain test with approximation by the F statistic (Regazzi & Silva,
yield to obtain the equivalent total of exported P per ha in 2010), it was determined if the parameters βk are the same
the grains. for each set of observations under comparison.
The regressions between the Mehlich-1 P and residual
P in soil were adjusted to the Freundlich model using the
2.5 Statistical analysis “freundlichanalysis” command of the “PUPAIM” package
present in the R statistical software version 3.5.2 (R Core
Data was subjected to ANOVA considering the fixed effects Team, 2018). Regression between Bray-1 P and residual P
model and in case of significant differences, the means was adjusted to the linear model through the “lm” com-
were compared using the Tukey test (P < .05). Controls mand. All other statistical analyses mentioned above were
for each soil correction conditions were treated as addi- also carried out in the R software.
tional treatments.
The following model was considered for the complete
factorial (without control): 3 RESULTS
𝑌𝑖𝑗𝑘𝑙 = μ + 𝐵𝑙 + 𝐶𝑖 + 𝑀𝑗 + 𝐴𝑘 + 𝐶𝑀𝑖𝑗 + 𝐶𝐴𝑖𝑘 + 𝑀𝐴𝑗𝑘 + ε𝑖𝑗𝑘𝑙 3.1 Annual and total grain yield
where μ = general average, B = block (l = 1, 2, 3), C = soil Figure 1 shows crop yields obtained over the 16 experimen-
correction (j = 1, 2, 3), M = maintenance source (j = 1, 2, 3), tal years, indicating a high yield potential, compatible with
A = mode of application (k = 1,2), ε = experimental error. high technology systems. In the last soybean (15th) and
Assumptions of homoscedasticity and normality of the corn (16th) crops, for example, the average yield of treat-
residues were verified by the Levene and Shapiro–Wilk ments that received maintenance fertilizer was 4.0 and
tests, respectively. An ANOVA was also made to evaluate 13.4 t of grain ha−1 , respectively. There is also growing crop
the yield development of soybean and corn crops through yield potential over the years. In addition, there is a large
the growing seasons, considering repeated measurements initial dispersion of different treatment effects, which is
(Vivaldi, 1999). Regression analyzes were also performed reduced, but not eliminated, in later seasons of the exper-
between the relative yield and residual P stock present in iment. For example, in the first harvest treatments that
the soil, adjusted to the Mitscherlich model via the “nls” received correction fertilization produced on average 180%
function of the statistical package R version 3.5.2 (R Core more grains than those that did not receive this initial P
Team, 2018). input, while in the last harvest this difference was reduced
In order to verify the statistical equality of these models to 9%. Similarly, treatments that received maintenance fer-
for each cultivated crop, within each P source used and vice tilizer with TSP produced 29% more than those with RRP
versa, the model identity analysis was performed (Regazzi maintenance in the first harvest, an advantage that was
& Silva, 2010). Four Mitscherlich nonlinear models were reduced to 6.5% in the last crop.
generated resulting from the relationship between residual The significance of the main effects and interactions
P in the soil (P) and relative yield (RY), given by: of treatments on grain yield was summarized in Table 4.
There is a consistent positive effect of soil correction,
𝑅𝑌 𝑖 = 𝑏1 − 𝑏2 .exp(𝑏3.𝑃𝑖) + 𝑒𝑖 which lasted throughout the entire duration of the exper-
iment, with the exception of the 15th crop (soybean). The
where: maintenance source also had effects on almost all crops,
Yi : corresponds to the ith value of the response variable, and many interacted with the correction factor. On the
i = 1, 2, . . . , N observations; other hand, the P fertilizer application method was not sig-
Pi : corresponds to the ith value of the explanatory vari- nificant in every season, but it did have a significant effect
able, i = 1, 2, . . . , N observations; on the accumulated total yield.
βk : corresponds to the parameter of the non-linear Table 5 shows the total relative accumulated yields at the
model, k = 1, 2, 3. end of the experiment. There was a significant effect of cor-
εi : corresponds to random errors. rection fertilization compared with the control, either with
TA B L E 4 Analysis of variance of the different variation sources in the yield of the 16 crops. C: phosphorus correction source; M: phosphorus maintenance source; A: application method
1999/ 2000/ 2001/ 2002/ 2003/ 2004/ 2005/ 2006/ 2007/ 2008/ 2009/ 2010/ 2011/ 2012/ 2013/ 2014/
ANOVA 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
ZANCANARO DE OLIVEIRA et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Factors (Soybean) (Soybean) (Soybean) (Corn) (Soybean) (Corn) (Soybean) (Corn) (Soybean) (Corn) (Soybean) (Corn) (Soybean) (Corn) (Soybean) (Corn)
† † † † † † *** † * † ** † * * **
C ns
† † † † ** † † † † † ** ** *
M ns ns ns
† * † ** ** *** * **
A ns ns ns ns ns ns ns ns
† † ** * ** ** *** * *
C×M ns ns ns ns ns ns ns
* * * **
C×A ns ns ns ns ns ns ns ns ns ns ns ns
** **
M×A ns ns ns ns ns ns ns ns ns ns ns ns ns ns
C × M × ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns
A
Phase 1 Phase 2 Phase 3 Phase 4 Total Soybean Corn total
total
† † † † † † †
C
† † † † † †
M ns
** † ** * *
A ns ns
† * ** * † † ***
C×M
C×A ns ns ns ns ns ns ns
M×A ns ns ns ns ns ns ns
C × M × ns ns ns ns ns ns ns
A
Note. ns: not significant.
†
P < .0001.
*
.05 > P > .01.
**
.01 > P > .001.
***
.001 > P > .0001.
4309
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4310 ZANCANARO DE OLIVEIRA et al.
F I G U R E 1 Annual soybean and corn yield in the 16 experiment growing seasons, started in 1999, in a clayey Oxisol with very low initial
STP, in different soil correction and phosphate maintenance fertilization strategies. Values represent the means of the two application methods
and three replicates (n = 6). Bars represent the HSD at 5% probability according to Tukey’s test. Yields for the control treatment with no P
fertilization (Treatment 1) varied between 182.3 kg ha−1 and 392.2 kg ha−1 for soybean and between 52.5 kg ha−1 and 1,205.5 kg ha−1 for corn
T A B L E 5 Total yields after the 16 crops cultivated in the tions of TSP was especially higher when compared to the
experiment, relative to the mean of treatments under correction and use of RRP when there was no corrective fertilization. In
maintenance fertilization with triple superphosphate (TSP) this case, 13% more grains were produced, whereas in the
(100% = 113,606.3 kg ha−1 ). Values represent the mean of the case of corrective phosphate fertilization, this difference
different application methods and replicates (n = 6), except for
was reduced to an average of 5.1% (Table 5). Use of the TSP
controls without application of maintenance P (n = 3, number of
and RRP mixture in equal P proportions generally showed
replicates)
an intermediate behavior between the results observed
P correction
when these sources were applied separately. Therefore,
P maintenance Control TSP RRP
this treatment was omitted from the subsequent figures,
Relative yield, %
a
but always considered in the ANOVA.
Control 5.6Bd 27.0Ac 28.4Ac
Figure 2 shows time trends for the yield of (a) soybean
TSP 97.0Ba 100.0Aa 101.9Aa and (b) corn in different treatments subjected or not to
RRP 84.0Bc 96.3Ab 95.4Ac maintenance P applications. Where P was applied as cor-
TSP + RRP 92.4Bb 99.2Aa 98.1Ab rective fertilization in the beginning of the trial and no
Note. RRP: reactive rock phosphate. maintenance was made, crop yields dropped severely. This
a
Capital letters compare different soil correction strategies (treatments in the drop was more intense in the case of TSP correction in the
same row) and lower case letters compare those in the same column (mainte-
soybean crops. Control treatment with no P application
nance sources) according to Tukey’s test (P < .05).
retained very low crop yields throughout the experiment.
Treatments that received maintenance fertilization but not
TSP or RRP, and there was no difference between the two the initial dose of P showed the highest yield improvement,
sources for all maintenance conditions. The main differ- especially in the soybean crop. By the end of the experi-
ence observed between soil P correction strategies was in ment, though, all treatments subjected to the annual P dose
the control treatments, that is, without annual P mainte- reached similar yields.
nance fertilization. In this case, a single corrective fertiliza-
tion at the beginning of the experiment with 105 kg ha−1
of P allowed for production of about 27. 7% of the total 3.2 Correction and maintenance
produced in the reference treatment (correction and main- phosphorus effects in different cultivation
tenance with TSP, with total production of 113,606.3 kg phases
grains ha−1 ), while no application of any P source limited
the productivity to 5.6% of the total observed in the refer- Effects of the treatments on crop yields varied over the
ence treatment. In relation to phosphate maintenance, the growing seasons. Thus, to facilitate the comprehension
application of TSP resulted in higher crop yields compared of these effects, the average relative yield of crops was
to the sparingly soluble source (RRP), regardless of the cor- grouped in 4-yr phases. The reference treatment was that
rection condition (P < .05). Total yield with annual applica- most similar to what is practiced by farmers in the region:
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZANCANARO DE OLIVEIRA et al. 4311
(a)
(b)
F I G U R E 2 (a) Soybean and (b) corn crop yields evolution through the growing seasons. Letters in parenthesis compare different time
trends in yield development according to Tukey’s test (P < .05). Values represent the mean of the different application methods and replicates
(n = 6), except for controls without application of maintenance P (n = 3, number of replicates)
corrective fertilization and maintenance with TSP, band when maintenance was performed with the sparingly solu-
applying the maintenance fertilizer in the sowing furrow. ble source, corrective fertilization with TSP produced 12.2%
Phase 1 is then composed of three soybean crops and only more grains than with RRP.
one corn crop, and phases 2–4 are composed of two grow- The second phase in the soybean crop was character-
ing seasons for each crop. Figure 3 explores the relative ized by main correction effects, with treatments without
yields per phase for each crop. corrective fertilization producing an average of 11.1% less
In the soybean crop (Figure 3a), the effect of soil cor- than those with application of TSP or RRP, with no sig-
rection with P was markedly noticed in the first analyzed nificant difference between the two sources. As a source
phase, since this was the crop grown in the first 3 of the 4 yr. of P for maintenance fertilization, TSP promoted higher
In this phase, lack of a correction application with 105 kg yields than RRP (P < .05), with an average advantage
of P ha−1 severely restricted plant growth, especially when of 536 kg ha−1 per crop or 18.3%. In the third phase,
the maintenance source used was RRP. In this case, yields the absence of corrective fertilization was only significant
were only 41% of that observed in the reference treatment, when maintenance was performed with RRP. As for the
while in treatments with TSP application this value rose to source of P in maintenance, yields were similar between
74%. In the other two correction conditions, the advantage TSP and RRP when correction was made with the soluble
of the soluble source used in maintenance, although less source. However, when soil was not previously corrected
pronounced, was also significant. Comparing the two P or when it was corrected with RRP, maintenance fertiliza-
sources in soil correction, there was no difference between tion with TSP provided greater yields. In the fourth phase,
them when using TSP for maintenance. On the other hand, for soybean, there was no significant difference between
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4312 ZANCANARO DE OLIVEIRA et al.
(a)
F I G U R E 5 Residual fertilizer P in soil (kg P ha−1 ) before each of the 16 cropping seasons, started in 1999. Values represent application
method means for treatments with maintenance fertilization (n = 6) and replicate means for treatments without maintenance fertilization
(controls, n = 3)
fertilization, due to the exportation of P in harvested grains each P source. In order to obtain 90% of the reference treat-
and no annual replacement via maintenance fertilization. ment yield, it was necessary to have 107.7 kg of residual
In treatments in which only maintenance fertilizer was P ha−1 in soil for the cultivation of soybean and 126.6 kg
applied, a gradual increase in soil P stocks can be observed, of P ha−1 for the cultivation of corn, when the source
characterizing a gradual correction of P contents (Fig- used was TSP. The statistical comparison of these models
ure 5). Residual P values in these treatments approached showed that there is no difference between the response
those observed where soil was initially corrected and of soybean and corn to the residual P in soil (P = .1611).
maintained with annual fertilization (treatments under Thus, these values are close to that recommended by Sousa
total corrective fertilization). As a result, there was also an and Lobato (2004) for corrective P fertilization accord-
approximation in yields obtained in these two treatment ing to the clay content and initial Mehlich-1 P found in
groups (Figures 2 and 3). the studied soil (130 kg of P ha−1 ), valid for annual grain
Due to the often-higher yields in treatments with main- crops.
tenance fertilizer in the form of TSP (Figure 1), greater P When RRP was used there was a need for higher residual
offtake via the grains gradually resulted in reduced P accu- P doses in soil in relation to what was necessary with TSP
mulation in these treatments when compared to RRP. At to obtain 90% of the relative yield. This fact was significant
the end of the 16 crops, total P stocks in soil of treatments for both soybean (P < .001) and corn crops (P < .001). These
maintained with RRP, regardless of the P correction and values were 206.9 and 193.2 kg P ha−1 for soybean and corn,
maintenance strategies, was on average 28.2% higher. respectively. Comparing both crop responses when using
the sparingly soluble source (RRP), there was no difference
between the models (P = .3413).
3.5 Critical residual and labile By adopting a single model involving the two crops, a
phosphorus levels critical value of 113.6 kg of P ha−1 was found for residual
TSP. For this residual value in soil, the critical Mehlich-1P
The P input–output balance in soil of the treatments fertil- level was 4.1 mg kg−1 (Figure 7a). This value is lower than
ized exclusively with TSP or RRP, calculated from the dif- that proposed by Sousa and Lobato (2004) to obtain 90% of
ference between fertilizer inputs and P export in the grains the yield potential. For RRP, the residual critical value of P
during the 16 yr of experimentation, allowed for establish- in the soil was 205.2 kg ha−1 , which provides 20.4 mg kg−1
ment of critical residual P levels in soil based on its rela- of Mehlich-1 P (Figure 7a).
tionship with crop yield and STP (Figures 6 and 7). For the critical values of 113.6 kg P ha−1 in soil for resid-
Figure 6 shows regression models between residual P in ual TSP and 205.2 kg P ha−1 for RRP, expected Bray-1 P lev-
soil and the relative yield of soybean and corn crops for els are 5.7 and 3.9 mg kg−1 , respectively (Figure 7b).
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4314 ZANCANARO DE OLIVEIRA et al.
F I G U R E 6 Relative yield as a function of the residual fertilizer P stock ha−1 present in the soil. Values are calculated as a percentage of the
reference treatment yield (Treatment 10, correction fertilization with triple superphosphate (TSP) and maintenance with band-applied TSP).
Data for treatments fertilized exclusively with TSP (Treatments 2, 3, 8, 9 and 10) or reactive rock phosphate (RRP) (4, 5, 15, 18 and 19). P: residual
P in soil (kg ha−1 ); RY: relative yield. All model coefficients were significant at 0.1% by the t test, n = 45 for soybean and n = 35 for corn. *Models
differ statistically according to the F test at 5% probability. ns: Models do not differ according to the F test at 5% probability
F I G U R E 9 Total recovery of P contained in harvested products (grains), as a percentage of P applied as fertilizers and soil conditioners,
after the end of the experiment. Uppercase letters compare controls and maintenance sources in each soil correction strategy, and lowercase
letters compare application methods in a given correction and maintenance source. Means with the same letter do not differ according to
Tukey’s test (P < .05), n = 3. Total P exported in the control treatment (Treatment 1) with no correction or maintenance phosphate fertilizer
application, with P applied only as agricultural gypsum: 17.8 kg ha−1
Up until the third phase, even in treatments under total with expectations, with no synergetic effect that justifies
corrective fertilization where the residual P (Figure 5) was its recommendation.
sufficient to promote more than 90% relative yield (Fig- The regression models between relative yield and resid-
ure 6), lower yields were observed with the use of RRP for ual P in soil were the same between soybean and corn
maintenance compared to TSP (Figure 3). In the fourth and crops within each phosphate source (Figure 6). This is
last studied phase, when the residual P content in soil pre- in accordance with the official recommendation for the
viously treated with correction P fertilization approached region (Sousa & Lobato, 2004), which does not differen-
values sufficient for 100% relative yield, there was no dif- tiate annual grain crops in terms of the recommendation
ference between the P maintenance sources. This indicates for corrective or maintenance P fertilization.
that for maintenance, due to the slow solubilization effect,
the application of RRP is more suitable for soils where
STP is not limiting. This benefit is observed in the longer 4.3 Phosphorus application method and
term (Oliveira et al., 2019), in a manner similar to its use grain yield in long-term no-till system
as a P correction source. In these situations, crop nutrition
depends little on recently applied P fertilizer, where soil is Increasing P stocks throughout the cropping seasons in
the main provider of P to the plants (Mclaren et al., 2015). treatments that received maintenance P caused STP values
Due to lower yields frequently observed with the use of to rise (Figures 5 and 7). In phases 3 and 4, with adequate
RRP for maintenance, P recovery efficiency with the use P stocks and STP values in soil, broadcast application
of this source at the end of the experiment was lower than increased yields by 2.1 and 2.9% on average over band-
with the soluble source in all correction conditions. On the applied P fertilizer, respectively (Figure 4). This fact was
other hand, the control treatment grown only with correc- also reflected in the greater P recovery under broadcast
tive RRP fertilization at the beginning of the experiment application (Figure 9).
exported 83.1% of the added P, whereas with TSP correc- There are probably three reasons for this. The P concen-
tion, this value was 75.8%. However, this difference was tration in surface layers decreases the soil buffer capac-
not significant. ity (Barrow et al., 2018), that is, P-binding intensity to
Maintenance with the TSP and RRP mixture in equal P soil colloids in surface layers is reduced since adsorption
parts showed intermediate crop response between the sep- sites are quickly saturated. Because crop rows rarely over-
arate application of these sources (Table 5), which is in line lapped perfectly for two or more growing seasons, the P
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZANCANARO DE OLIVEIRA et al. 4317
concentration effect is reduced over the years when the fer- sons such as high organic matter content and less phos-
tilizer is band applied. Although the fertilized soil volume phate interaction with soil solid phase (Tiecher, Calegari,
extends to deeper layers in this case, this does not seem to Caner, & Rheinheimer, 2017), critical levels were reduced
counterbalance the effect of high P availability found in the (Sousa, Rein, Goedert, Lobato, & Nunes, 2010). This was
superficial layers with broadcast application. probably due to the increase in organic P fractions that play
Although maximum P adsorption capacity is not an important role in crop nutrition, but whose availabil-
affected by soil organic matter in highly weathered soils ity is not considered by traditional STP methods, such as
in Brazil (Fink, Inda, Bayer, Torrent, & Barrón, 2014), it Mehlich-1 (Sousa et al., 2010).
can reduce the P-binding energy intensity with soil colloids Bray-1 and Olsen (0.5 M NaHCO3 ) soil tests are often
(Fink, Inda, Tiecher, & Barrón, 2016b). This means that considered alternatives for the assessment of labile P in
uneven P distribution throughout the soil profile under soils fertilized with phosphate rocks. In our results, how-
broadcast P application, and its similarity with the distri- ever, it is noticeable an underestimation of labile P in soil
bution of C contents under no-till system (NT), may be fertilized with RRP when using the Bray-1 test, since STP
a positive factor. For example, Coelho et al. (2019) cited values are considerably lower than that observed for TSP
organic C accumulation in the 0- to 7.5-cm soil layer as a for a given residual P (Figure 7b). Even when considering
possible reason for the greater P use efficiency observed in that greater amounts of residual RRP P are necessary to
broadcast P application. obtain the same relative yields as TSP, as discussed above,
The third possible justification for the higher observed P Bray-1 P values are still low. For example, for the produc-
use efficiency under broadcast application may be related tion of 90% relative yield, it is estimated the need for critical
to the low organic C/P ratio found in NT superficial layers, values of 113.6 kg residual TSP P ha−1 and 205.2 kg residual
which promotes high phosphatase enzyme activity (Sousa, RRP P ha−1 in soil (Figure 6). The expected Bray-1 P levels
Nunes, & Rein, 2019). A reduction of this ratio can be for these respective amounts of residual P are 5.7 mg kg−1
enhanced in broadcast application due to the even higher for TSP and lower than that for RRP (3.9 mg kg−1 ), despite
values of labile organic and inorganic P observed in the first the extra 91.6 kg P ha−1 (Figure 7b). Figure 7a also shows
5-cm layer (Nunes, 2014). Thus, potential P supply from an opposite problem, an apparent overestimation of labile
organic forms, not accounted for in the Mehlich-1 method, P when Mehlich-1 soil test is used to assess lability in RRP
can be very significant when phosphatase enzyme activity fertilized soils, which is due to solubilization of unreacted
is increased. RRP particles by the acid extract.
It should be noted, however, that there are doubts The aforementioned problems were discussed by
regarding the effect of prolonged droughts on P absorption Oliveira et al. (2019) in the same soil type. The authors
in areas where P is mainly concentrated in shallow surface noticed similar underestimation between Bray-1 and
layers, as occurs in NT under broadcast P application due Olsen tests and an overestimation of Mehlich-1 P when
to the reduction of water availability in the narrow layer dealing with RRP fertilized soils. A series of similar results
rich in P. Experimentally, however, there were no major are also discussed by Zapata and Roy (2004). Despite
yield losses when a water deficit was induced compared these concerns, Oliveira et al. (2019) affirm all these three
to P application at a depth of 5 cm (Hansel et al., 2017a). methods can all adequately assess STP, provided that
This might be related to the ability of some plants to raise results are interpreted considering the respective P source.
soil moisture closer to the surface, although this is limited More recently some studies have suggested that high-
(Shen, Zhang, & Li, 2011). yielding modern varieties would require a reassessment of
soil P critical levels (Hopkins & Hansen, 2019). However,
under well-established NT conditions not only STP criti-
4.4 Critical STP levels in NT may be cal levels appear to be adequate (Antonangelo, Firmano,
lower than expected Ferracciú Alleoni, Oliveira, & Zhang, 2019), but also there
are many areas receiving high P inputs compared to out-
In order to obtain 90% yield potential in relation to the puts in harvested products, what allows for opportunities
reference treatment, approximately 113.6 kg ha−1 of resid- for farmers to exploit residual P (Withers et al., 2018).
ual P from TSP was required in soil of which the expected
Mehlich-1 P content for this P stock was about 4.1 mg kg−1
(Figure 7); this is lower than that recommended for the 5 CONCLUSIONS
Cerrado region by Sousa and Lobato (2004). It is proba-
bly due to the fact that the calibration curves adopted by Corrective P fertilization promoted a rapid response in
these authors were based on the conventional tillage sys- crop yields, with an equivalence between the P sources
tem (CT). In NT as in the present experiment, for rea- in most evaluated cultivation phases. Exceptions were the
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4318 ZANCANARO DE OLIVEIRA et al.
higher yields observed for TSP correction when mainte- Bihn, T., Pichot, J., & Beunard, P. (1978). Caracterisation et compara-
nance was performed with RRP in the first phase of the ison des phosphates naturels tricalciques d’Afrique de l’Ouest en
soybean crop, and the higher yields obtained with RRP vue de leur utilisation directe en agriculture. L’Agronomie Tropi-
cale, 33, 136–145.
correction under TSP maintenance in the second phase of
Bolan, N. S., & Hedley, M. J. (1990). Dissolution of phosphate rocks
the corn crop. However, in the first crop harvest high TSP
in soils. 2. Effect of pH on the dissolution and plant availability
solubility was more efficient in supplying P to crops. In of phosphate rock in soil with pH dependent charge. Fertilizer
the uncorrected condition, high yields were only obtained Research, 24(3), 125–134. https://doi.org/10.1007/BF01073580
when residual P in the soil was sufficient for its correction, Bray, R.H., & Kurtz, L. T. (1945). Determination of total, organic,
especially when using RRP for maintenance. The applica- and available forms of phosphorus in soils. Soil Science, 59(1),
tion method did not influence average yields in the first 39–46.
Coelho, M. J. A., Diaz, D. R., Hettiarachchi, G. M., Hansel, F. D.,
two phases, however with accumulating P stocks in soil,
& Pavinato, P. S. (2019). Soil phosphorus fractions and legacy in
broadcast application improved yields in phases 3 and 4.
a corn-soybean rotation on Mollisols in Kansas, USA. Geoderma
Estimated residual stocks of 113.6 kg of P ha−1 using TSP Regional, 18, e00228. https://doi.org/10.1016/j.geodrs.2019.e00228
and 205.2 kg of P ha−1 in RRP equivalent were necessary Dodd, R. J., & Sharpley, A. N. (2015). Recognizing the role of soil
to obtain 90% of the reference yield potential. These val- organic phosphorus in soil fertility and water quality. Resources,
ues permit that Mehlich-1 P is increased to critical levels, Conservation and Recycling, 105, 282–293. https://doi.org/10.1016/
where the value of 4.1 mg kg−1 for TSP is lower than that j.resconrec.2015.10.001
Embrapa. (1999). Handbook for chemical analysis of soils, plants
recommended for the region and may be due to the contri-
and fertilizers. (In Portuguese.) Empresa Brasileira de Pesquisa
bution of organic P in NT, the fraction not accounted for in Agropecuária. Brasília, Brazil: Embrapa Informacão Tecnológica.
the Mehlich-1 method. Fink, J. R., Inda, A. V., Bavaresco, J., Barrón, V., Torrent, J., & Bayer, C.
(2016a). Adsorption and desorption of phosphorus in subtropical
AC K N OW L E D G M E N T S soils as affected by management system and mineralogy. Soil and
This research was supported by Embrapa Cerrados and Tillage Research, 155, 62–68. https://doi.org/10.1016/j.still.2015.07.
was financed in part by the Coordenação de Aperfeiçoa- 017
Fink, J. R., Inda, A. V., Bayer, C., Torrent, J., & Barrón, V. (2014).
mento de Pessoal de Nível Superior–Brasil (CAPES)–
Mineralogy and phosphorus adsorption in soils of south and
Finance Code 001. central-west Brazil under conventional and no-tillage systems.
Acta Scientiarum. Agronomy, 36(3), 379. https://doi.org/10.4025/
CONFLICT OF INTEREST actasciagron.v36i3.17937
The authors declare that they have no conflict of interest. Fink, J. R., Inda, A. V., Tiecher, T., & Barrón, V. (2016b). Iron
oxides and organic matter on soil phosphorus availability. Ciên-
ORCID cia e Agrotecnologia, 40(4), 369–379. https://doi.org/10.1590/1413-
70542016404023016
Luiz E. Zancanaro de Oliveira https://orcid.org/0000-
Hansel, F. D., Amado, T. J. C., Ruiz Diaz, D. A., Rosso, L. H. M.,
0002-4374-9631
Nicoloso, F. T., & Schorr, M. (2017a). Phosphorus fertilizer place-
ment and tillage affect soybean root growth and drought tol-
REFERENCES erance. Agronomy Journal, 109(6), 2936–2944. https://doi.org/10.
Alvares, C. A., Stape, J. L., Sentelhas, P. C., De Moraes Gonçalves, J. 2134/agronj2017.04.0202
L., & Sparovek, G. (2013). Köppen’s climate classification map for Hansel, F. D., Ruiz Diaz, D. A., Amado, T. J. C., & Rosso, L. H. M.
Brazil. Meteorologische Zeitschrift, 22(6), 711–728. https://doi.org/ (2017b). Deep banding increases phosphorus removal by soybean
10.1127/0941-2948/2013/0507 grown under no-tillage production systems. Agronomy Journal,
Antonangelo, J. A., Firmano, R. F., Ferracciú Alleoni, L. R., Oliveira, 109(3), 1091–1098. https://doi.org/10.2134/agronj2016.09.0533
A., & Zhang, H. (2019). Soybean yield response to phosphorus fer- Hedley, M. J., Stewart, J. W. B., & Chauhan, B. S. (1982). Changes
tilization in an oxisol under long-term no-till management. Soil in inorganic and organic soil phosphorus fractions induced by
Science Society of America Journal, 83(1), 173–180. https://doi.org/ cultivation practices and by laboratory incubations. Soil Science
10.2136/sssaj2018.07.0251 Society of America Journal, 46(5), 970–976. https://doi.org/10.2136/
Barrow, N. J. (1980). Evaluation and utilization of residual phospho- sssaj1982.03615995004600050017x
rus in soils. In F. Kasewneh & G. Sample (Eds.), The role of phos- Hopkins, B. G., & Hansen, N. C. (2019). Phosphorus management in
phorus in agriculture (pp. 333–359). Madison, WI: ASA. high-yield systems. Journal of Environmental Quality, 1280, 1265–
Barrow, N. J. (1985). Comparing the effectiveness of fertilizers. Fertil- 1280. https://doi.org/10.2134/jeq2019.03.0130
izer Research, 8(1), 85–90. https://doi.org/10.1007/BF01048909 Johnston, A. E., & Poulton, P. R. (2019). Phosphorus in agriculture:
Barrow, N. J. (2015). Soil phosphate chemistry and the P-sparing A review of results from 175 years of research at Rothamsted, UK.
effect of previous phosphate applications. Plant and Soil, 397(1–2), Journal of Environmental Quality, 48(5), 1133. https://doi.org/10.
401–409. https://doi.org/10.1007/s11104-015-2514-5 2134/jeq2019.02.0078
Barrow, N. J., Barman, P., & Debnath, A. (2018). Three residual bene- Liu, J., Hu, Y., Yang, J., Abdi, D., & Cade-Menun, B. J. (2014). Investi-
fits of applying phosphate fertilizer. Soil Science Society of America gation of soil legacy phosphorus transformation in long-term agri-
Journal, 82(5), 1168–1176. https://doi.org/10.2136/sssaj2018.03.0115 cultural fields using sequential fractionation, P K-edge XANES
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZANCANARO DE OLIVEIRA et al. 4319
and solution P NMR spectroscopy. Environmental Science & Tech- em dados de experimento com delineamento em blocos casualiza-
nology, 49(1), 168–176. https://doi.org/10.1021/es504420n dos. Revista Ceres, 57(3), 315–320. https://doi.org/10.1590/s0034-
Lou, H., Zhao, C., Yang, S., Shi, L., Wang, Y., Ren, X., & Bai, J. (2018). 737x2010000300005
Quantitative evaluation of legacy phosphorus and its spatial dis- Rosendo dos Santos, V., Soltangheisi, A., Franco, H. C. J., Kolln, O.,
tribution. Journal of Environmental Management, 211, 296–305. Vitti, A. C., Dias, C. S., & Pavinato, P. (2018). Phosphate sources
https://doi.org/10.1016/j.jenvman.2018.01.062 and their placement affecting soil phosphorus pools in sugarcane.
Lu, D., Song, H., Jiang, S., Chen, X., Wang, H., & Zhou, J. (2019). Inte- Agronomy, 8(12), 1–15. https://doi.org/10.3390/agronomy8120283
grated phosphorus placement and form for improving wheat grain Rowe, H., Withers, P. J. A., Baas, P., Chan, N. I., Doody, D., Holiman,
yield. Agronomy Journal, 111, 1998–2004. https://doi.org/10.2134/ J., . . . Weintraub, M. N. (2016). Integrating legacy soil phosphorus
agronj2018.09.0559 into sustainable nutrient management strategies for future food,
Lun, F., Liu, J., Ciais, P., Nesme, T., Chang, J., Wang, R., . . . Ober- bioenergy and water security. Nutrient Cycling in Agroecosystems,
steiner, M. (2018). Global and regional phosphorus budgets in agri- 104(3), 393–412. https://doi.org/10.1007/s10705-015-9726-1
cultural systems and their implications for phosphorus-use effi- Roy, E. D., Richards, P. D., Martinelli, L. A., Della Coletta, L., Lins, S.
ciency. Earth System Science Data, 10(1), 1–18. https://doi.org/10. R. M., Vazquez, F. F., . . . Porder, S. (2016). The phosphorus cost of
5194/essd-10-1-2018 agricultural intensification in the tropics. Nature Plants, 2(5), 2–7.
McLaren, T. I., Mclaughlin, M. J., Mcbeath, T. M., Simpson, R. J., https://doi.org/10.1038/NPLANTS.2016.43
Smernik, R. J., Guppy, C. N., & Richardson, A. E. (2015). The Roy, E. D., Willig, E., Richards, P. D., Martinelli, L. A., Vazquez, F. F.,
fate of fertiliser P in soil under pasture and uptake by subterra- Pegorini, L., . . . Porder, S. (2017). Soil phosphorus sorption capac-
neum clover –A field study using 33 P-labelled single superphos- ity after three decades of intensive fertilization in Mato Grosso,
phate. Plant and Soil, 401, 23–38. https://doi.org/10.1007/s11104- Brazil. Agriculture Ecosystems and Environment, 249(July), 206–
015-2610-6 214. https://doi.org/10.1016/j.agee.2017.08.004
Menezes-Blackburn, D., Giles, C., Darch, T., George, T. S., Black- Shen, Y., Zhang, Y., & Li, S. (2011). Nutrient effects on diurnal varia-
well, M., Stutter, M., . . . Haygarth, P. M. (2018). Opportunities tion and magnitude of hydraulic lift in winter wheat. Agricultural
for mobilizing recalcitrant phosphorus from agricultural soils: A Water Management, 98(10), 1589–1594. https://doi.org/10.1016/j.
review. Plant and Soil, 427(1), 5–16. https://doi.org/10.1007/s11104- agwat.2011.05.012
017-3362-2 Silveira, T. C., Pegoraro, R. F., Kondo, M. K., Portugal, A. F., &
Murphy, J., & Riley, J. P. (1962). A modified single solution method for Resende, Á. V. (2018). Sorghum yield after liming and combi-
the determination of phosphate in natural waters. Analytica Chim- nations of phosphorus sources. Revista Brasileira de Engenharia
ica Acta, 27, 31–36. https://doi.org/10.1016/S0003-2670(00)88444-5 Agrícola e Ambiental, 22(4), 243–248. https://doi.org/10.1590/1807-
Nicolodi, M., Anghinoni, I., & Salet, R. L. (2002). Alternativa a coleta 1929/agriambi.v22n4p243-248
de uma secção transversal, com pá de corte, na largura da entre- Soltangheisi, A., Rodrigues, M., Coelho, M. J. A., Gasperini, A. M.,
linha, na amostragem do solo em lavouras com adubação em linha Sartor, L. R., & Pavinato, P. S. (2018). Changes in soil phosphorus
no sistema plantio direto. Revista Plantio Direto, 69(May/June), lability promoted by phosphate sources and cover crops. Soil and
22–28. Tillage Research, 179, 20–28. https://doi.org/10.1016/j.still.2018.01.
Nunes, R. S. (2014). Phosphorus use efficiency in soil management 006
systems and phosphate fertilizer for a long period. (In Portuguese, Sousa, D. M. G., & Lobato, E. (2004). Cerrado: Correção do solo e
with English abstract.) Brasília, Brazil: Faculdade de Agronomia e adubação (2nd ed.). Planaltina, Brazil: Embrapa Informação Tec-
Medicina Veterinária, Universidade de Brasília. nológica.
Nunes, R. D. S., de Sousa, D. M. G., Goedert, W. J., & Vivaldi, L. Sousa, D. M. G., Nunes, R. S., & Rein, T. A. (2019). Manejo do fósforo
J. (2011). Distribuição de fósforo no solo em razão do sistema em sistemas de cultivo na região do Cerrado. In E. C. Severiano, M.
de cultivo e manejo da adubação fosfatada. Revista Brasileira F. Morais, & A. M. de Paula (Eds.), Tópicos em ciência do solo (Vol.
de Ciência do Solo, 35(3), 877–888. https://doi.org/10.1590/S0100- 10, pp. 239–282). Viçosa, Brazil: Sociedade Brasileira de Ciência do
06832011000300022 Solo.
Oliveira, L. E. Z., de, R., Nunes, S., de Sousa, D. M. G., Busato, J. G., Sousa, D. M. G., Rein, T. A., Goedert, W. J., Lobato, E., & Nunes,
& de Figueiredo, C. C. (2019). Response of maize to different soil R. S. (2010). Fósforo. In L. I. Prochnow, V. Casarin, & S. R. Stipp
residual phosphorus conditions. Agronomy Journal, 111(6), 3291– (Eds.), Boas práticas para uso eficiente de fertilizantes (Vol. 2,
3300. https://doi.org/10.2134/agronj2018.11.0710. pp. 67–132). Piracicaba, Brazil: Instituto Nacional da Propriedade
Olsen, S. R., Cole, C. V., Watanab, F. S., & Dean, L. A. (1954). Esti- Industrial.
mation of available phosphorus in soils by extraction with sodium Tiecher, T., Calegari, A., Caner, L., & Rheinheimer, D. dos S. (2017).
bicarbonate. USDA Circ. 939. Washington, DC: USDA. Soil fertility and nutrient budget after 23-years of different soil
Preston, C. L., Ruiz Diaz, D. A., & Mengel, D. B. (2019). Corn response tillage systems and winter cover crops in a subtropical Oxisol. Geo-
to long-term phosphorus fertilizer application rate and placement derma, 308, 78–85. https://doi.org/10.1016/j.geoderma.2017.08.028
with strip-tillage. Agronomy Journal, 111(2), 841–850. https://doi. USDA-NRCS. (2003). Soil taxonomy: Keys to soil taxonomy (9th ed.).
org/10.2134/agronj2017.07.0422 Washington, DC: USDA-NRCS.
R Core Team. (2018). R: A language and environment for statisti- van der Bom, F. J. T., McLaren, T. I., Doolette, A. L., Magid, J.,
cal computing. Vienna: R Foundation for Statistical Computing. Frossard, E., Oberson, A., & Jensen, L. S. (2019). Influence of long-
Retrieved from https://www.R-project.org term phosphorus fertilisation history on the availability and chem-
Regazzi, A. J., & Silva, C. H. O. (2010). Testes para verificar a igualdade ical nature of soil phosphorus. Geoderma, 355(August), 113909.
de parâmetros e a identidade de modelos de regressão não-linear https://doi.org/10.1016/j.geoderma.2019.113909
14350645, 2020, 5, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.1002/agj2.20324 by CAPES, Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4320 ZANCANARO DE OLIVEIRA et al.
Vivaldi, L. J. (1999). Analysis of experiments with repeated measure- phorus adsorbing Oxisol of Central Brazil. Soil Science Society
ments on time or space. (In Portuguese.) Empresa Brasileira de of America Journal, 45(3), 540. https://doi.org/10.2136/sssaj1981.
Pesquisa Agropecuária. Planaltina, Brazil: Embrapa Cerrados. 03615995004500030021x
Withers, P. J. A., Rodrigues, M., Soltangheisi, A., De Carvalho, T. S., Zapata, F., & Roy, N. R. (2004). Use of phosphate rocks for sustainable
Guilherme, L. R. G., Benites, V.De.M., . . . Pavinato, P. S. (2018). agriculture. Rome, Italy: Food and Agriculture Organization of the
Transitions to sustainable management of phosphorus in Brazil- United Nations, Land and Water Development Division, Interna-
ian agriculture. Scientific Reports, 8(1), 1–13. https://doi.org/10. tional Atomic Energy Agency.
1038/s41598-018-20887-z
World Bank Group. (2020). GEM commodities, World Bank Group.
Retrieved from https://datacatalog.worldbank.org/dataset/gem-
How to cite this article: Zancanaro de Oliveira
commodities
Yost, R. S., Kamprath, E. J., Lobato, E., & Naderman, G. (1979). LE, de Sousa DMG, de Figueiredo CC, Nunes RdS,
Phosphorus response of corn on an Oxisol as influenced by rates Malaquias JV. Long-term phosphate fertilization
and placement. Soil Science Society of America Journal, 43(2), 338. strategies evaluation in a Brazilian Oxisol.
https://doi.org/10.2136/sssaj1979.03615995004300020020x Agronomy Journal. 2020;112:4303–4320.
Yost, R. S., Kamprath, E. J., Naderman, G. C., & Lobato, E. (1981). https://doi.org/10.1002/agj2.20324
Residual effects of phosphorus applications on a high phos-