Impact of organic no till vegetables sys
Scientia Horticulturae 182 (2015) 145–155
Contents lists available at ScienceDirect
Scientia Horticulturae
journal homepage: www.elsevier.com/locate/scihorti
Impact of organic no-till vegetables systems on soil organic
matter in the Atlantic Forest biome
A. Thomazini a,∗ , E.S. Mendonc¸a a , J.L. Souza b , I.M. Cardoso c , M.L. Garbin a
a
b
c
Department of Plant Production, Federal University of Espírito Santo, 29500-000 Alegre, ES, Brazil
Research of INCAPER—Centro Serrano, BR-262, km 94, 29.375-000 Venda Nova do Imigrante, ES, Brazil
Soil Science Department, Federal University of Vic¸osa, Avenida P.H. Rolfs, s/n, Vicosa 36570-000, MG, Brazil
a r t i c l e
i n f o
Article history:
Received 25 August 2014
Received in revised form
25 November 2014
Accepted 1 December 2014
Keywords:
Green manure
Labile and stable fractions
Soil health
Soil C balance
a b s t r a c t
Soil organic matter is widely recognized as a strategy used to improve soil quality and reduce carbon
emissions to the atmosphere. A field study was carried out to investigate the effects of cover crops in
organic no-till vegetables systems on changes in soil organic matter and CO2 C emissions, in dry and
rainy seasons. We hypothesized that CO2 C emissions are higher in conventional till as compared with
no-till, and that no-till increases soil C sink. The crop rotation comprised a 3-year cropping sequence
involving two crops per year—cabbage (Brassica oleracea L.) in winter and eggplant (Solanum melongena
L.) in summer time. Treatments were no-till on dead mulch of grass (Avena strigosa Schreb. and Zea mays
L.), leguminous (Lupinus albus L. and Crotalaria juncea L.), intercrop (grass and leguminous) and conventional till (no dead mulch) with rotary hoe arranged in a randomized block design on a clayey Oxisol (Typic
Haplustox) at Domingos Martins-ES, Brazil. On 2012 and 2013, disturbed soil samples at three different
layers (0–5, 5–15 and 15–30 cm) and undisturbed samples at 0–10, 10–20 and 20–30 cm, for chemical
and organic matter characterization were taken. CO2 C emissions and soil temperature were measured
in situ on March, May, August and October 2012 and February 2013 (after 3 years of experiment). Conventional till site showed the lowest microporosity values and the highest macroporosity, followed by
lower soil bulk density at 0–10 cm layer. Total organic C ranged from 34.94 to 50.48 g kg−1 in intercrop
and 27.11 to 43.74 g kg−1 in conventional till. Total N ranged from 2.81 to 5.34 g kg−1 in grass and 2.54
to 4.51 g kg−1 in conventional till. Highest C stock was recorded in intercrop. Conventional till showed
lower labile C values while recalcitrant C was higher in the intercrop treatment. The annual average of
CO2 C emissions (mol CO2 m−2 s−1 ) followed the order: grass (15.89) > intercrop (13.77) > leguminous
(13.09) > conventional till (11.20). Highest annual average of soil temperature was recorded in conventional till (23.95 ◦ C). Lowest annual mean of soil water content, microbial biomass C, and highest metabolic
quotient were recorded in conventional till. These results suggest that the use of cover crops and organic
compost in pre-planting promote C increments. The contribution of organic residues increases the water
holding capacity and reduces soil temperature. No-till reduces soil disturbance and promotes a positive
balance of C. Organic no-till vegetable systems is a strategy to increase soil C and should be encouraged
in order to increase soil quality in the Atlantic Forest Biome in Brazil.
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
The Brazilian Atlantic Forest is now reduced to about 11.4 to 16%
of its original cover of approximately 150 million hectares (Ribeiro
et al., 2009). Most deforested areas are composed of agricultural
∗ Corresponding author. Tel.: +55 27 3359 3971; fax: +55 28 3552 8927.
E-mail addresses: andre.thz@gmail.com (A. Thomazini),
eduardo.mendonca@ufes.br (E.S. Mendonc¸a), jacimarsouza@yahoo.com.br
(J.L. Souza), irene@ufv.br (I.M. Cardoso), mlgarbin@gmail.com (M.L. Garbin).
http://dx.doi.org/10.1016/j.scienta.2014.12.002
0304-4238/© 2014 Elsevier B.V. All rights reserved.
systems on degraded soils. Anthropogenic activities lead to land
misuse causing changes in the physical, chemical and biological
attributes of soils (Reicosky et al., 1999; Powlson et al., 2011). This
implies decreases in the storage of organic carbon and nutrients as
well as in the productive capacity of soils, since C is an indicator
used to assess soil quality (Silva and Mendonc¸a, 2007; Ghosh et al.,
2012).
It is widely recognized that soil organic matter is one of the most
important indicators of soil quality and health (Lal, 2004; Ghosh
et al., 2012). Increasing or maintaining soil organic matter is critical to achieve optimum soil functions and crop production (Ghosh
146
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
et al., 2012). When monitoring soil quality in the tropics, sensitive
soil quality indicators need to be identified, mainly due the continuous and intensive vegetable production in these areas (Moeskops
et al., 2012). Soil management can lead to higher decomposition
rates of organic matter decreasing the concentration of this soil
component (Silva and Mendonc¸a, 2007). Agriculture can significantly contribute to elevate atmospheric CO2 concentrations as a
consequence of soil management (Powlson et al., 2011). These C
losses to the atmosphere can be mainly reduced by minimizing
soil disturbance, either with no-till or agroecological management
(Silva and Mendonc¸a, 2007). It is estimated that 89% of the potential
for mitigation of greenhouse gases produced by agriculture relies
on C sequestration (Smith et al., 2008). In addition, increasing the
soil organic C content is an important strategy to deal with climate
changes driven by C emissions to the atmosphere from agricultural
lands.
No-till and organic agriculture increase soil C and N sequestration, and reduce the oxidation of soil organic matter (Bayer et al.,
2009; Campiglia et al., 2014). Continuous input of plant residues
and paucity of soil disturbance promote reductions in CO2 C emissions through decreases in organic matter decomposition rates
(Lal, 2004; Bayer et al., 2009). On other hand, conventional crop
production intensify soil disturbance and, consequently, breakdown the soil aggregates (Bayer et al., 2009). Conventional tillage
is the most common agricultural management for vegetable production in areas formerly occupied by the Atlantic Forest in Brazil.
In addition, vegetable production is historically managed by family
smallholders. Intensive farming or intensive soil preparation in horticulture degrades the soil–plant environment, mostly due to the
reduction in concentration and quality of soil organic matter and
the diversity of soil organisms (Tian et al., 2011). Degradation of soil
organic matter leads to long-term decreases in horticultural productivity. Thus, sustainable tillage is preferable to attain a positive
net balance of C in the highly weathered tropical soils (Mendonc¸a
and Rowell, 1996).
The use of cover crops represent a potentially valuable supply
of organic residues (C source) when they are used in no-tillage
systems and their residues are left on the soil surface (Campiglia
et al., 2014). No-till systems can mitigate CO2 C emissions. This is
because crop rotation and organic residues on soil surface promote
gradual decomposition of organic matter, favoring C incorporation
(Bayer et al., 2009; Conceic¸ão et al., 2013). Physical protection of
organic matter provided by stable aggregates under no-till reduce
organic matter mineralization and lead to C accumulation (Six et al.,
2004). However, there is a lack of information about C storage gains
and CO2 C soil emissions by organic no-till vegetable systems,
especially in the areas formerly occupied by the Atlantic Forest
biome, a well-known biodiversity hotspot (Myers et al., 2000). Here,
we report the results of a long term field experiment conducted
in dry and rainy seasons. We aimed to investigate the effects of
cover crops in organic no-till vegetables systems on changes of
soil organic matter and CO2 C emissions, in dry and rainy seasons.
We hypothesized that CO2 C emissions are higher in conventional
till as compared with no-till, and that no-till increases soil C sink,
leading to improved soil quality.
2. Material and methods
2.1. Site location, characterization and land uses prior to the
experiment
The study was carried out at the 2.5 ha organic agriculture
experimental site of Incaper (Espírito Santo Institute for Research,
Technical Assistance and Rural Extension), municipality of Domingos Martins-ES (20◦ 22′ SE 41◦ 03′ W) altitude of 950 m above the
Fig. 1. Average monthly precipitation and air temperature of the municipality of
Domingos Martins between January 2012 and February 2013. Data from Incaper.
sea. The climate of the region is Aw (tropical climate and dry season in winter), precipitation ranges from 750 to 1500 mm per year,
and all months of the year have average temperatures of 18 ◦ C or
higher. The region is characterized by dry winter and rainy summer
(Köppen, 1923). Mean monthly precipitation and air temperature
are presented in Fig. 1. Soil is classified as Red-Yellow Latosol,
Brazilian Classification System (Embrapa, 2006) or as clayey Oxisol,
Typic Haplustox (Soil Taxonomy, USDA classification). From 1990 to
2009, this area was cultivated with organic vegetables (mainly lettuce, cabbage and eggplant). Organic management was performed
using 15 Mg ha−1 of organic compost (dry mass) amendments. The
composting area followed the indore system (Miller and Jones,
1995) with alternating layers stacked forming cells that received
manual eversion periodically in order to control humidity (50%)
and temperature (60 ◦ C). The method relies on aerobic activity,
although portions of the pile can become anaerobic between turnings. Moreover, it provides better control of flies, more rapid
and uniform decomposition rates and less problems regarding
moisture control (Miller and Jones, 1995). The compost was prepared with a stacked mixture of: grounded green cameron grass
(Pennisetum purpureum Schumach.), coffee husk, crop residues of
maize and beans, and inoculation with chicken manure at the
rate of 50 kg m−3 . Organic compost characteristics were (total
amount): 52% organic matter, 16:1 carbon:nitrogen ratio, 7.3 pH, 2%
nitrogen, 1.2% phosphorus, 1.2% potassium, 4.8% calcium, 0.5% magnesium, 54 mg dm−3 copper, 188 mg dm−3 zinc, 12,424 mg dm−3
iron, 793 mg dm−3 manganese, 25 mg dm−3 boron. More details of
the organic vegetable cropping (1990–2009) can be found in Souza
et al. (2012).
2.2. Experimental design, cover crops and crop rotation
The organic no-till vegetables systems experiment was initiated
in 2009. The experiment comprises four tillage systems, implemented on 4 m × 6 m plots, according to a Randomized Complete
Block Design, with six replicates (totalizing 24 permanent experimental units) covering a total area of 576 m2 . Therefore, the effects
of organic management accumulated over the years. Tillage treatments consisted of:
(i) No-till on dead mulch of grass (grass): black oat (Avena strigosa
Schreb) was used as winter cover crop followed by maize (Zea
mays L.) as summer cover crop.
(ii) No-till on dead mulch of leguminous (leguminous): white lupin
(Lupinus albus, L.) was used as winter cover crop followed by
Sunnhemp (Crotalaria juncea L.) as summer cover crop.
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
(iii) No-till on dead mulch of grass and leguminous (intercrop):
grass and leguminous plants were intercropped using the same
cover crops in grass and leguminous treatments.
(iv) Conventional plow-based tillage (Conventional till): implemented using conventional tillage with rotary hoe one week
before planting, with no cover crop. The tractor used was a rear
rotary mini tiller (Yanmar MRT-650 EX) with the rotary tines
placed right behind the wheels. This is the main vegetable cropping system of the Brazilian horticulture (Souza et al., 2012).
Operation schedule conducted annually in the no-till and conventional till were presented in Table 1. From 2009 to 2013, no-till
was performed with black oat and white lupin as winter cover
crop, followed by cabbage as winter vegetable crop. Maize and
sunnhemp worked as summer cover crop, followed by eggplant as
summer vegetable crop. Black oat and white lupin were sown on
March 2012 as winter cover crops. Cover crop seeds were spread
manually and lightly buried. Cover crops were sown in rows spaced
33 cm from each other for all treatments. The seed rates were 480 g
per plot for black oat and 660 g per plot for white lupin. In the intercropped sampling units, seeds were reduced to half of these values.
On July 2012, cover crops were mowed by mechanical mowing and
cabbage was planted. Cover crop residues were left on the soil surface as organic dead mulch and they were not incorporated into the
soil. One month old cabbage seedlings were transplanted by hand.
The cabbage seedlings were arranged in single rows distant 60 cm
from each other. The distance between the cabbage plants in the
rows was 40 cm.
After winter crop, maize and sunnhemp were sown on October
2012 as summer cover crops. The seed rates were 600 g per plot for
maize and 300 g per plot for sunnhemp. Residues were mowed on
February 2013 followed by eggplant (Solanum melongena) planting.
Eggplant seedlings were grown in tubes of 180 cm3 , using a mixture
of organic compost/soil of 1:2 as substrate. The eggplant seedlings
were arranged in single rows at a distance of 120 cm between them.
The distance between the cabbage plants in the rows was 70 cm.
Cabbage and eggplant received 15 Mg ha−1 of organic compost (dry
mass) at planting in all no-till treatments. Cabbage and eggplant
seedlings were irrigated immediately after transplanting in order
to avoid moisture stress. Inside the rows, the weeds were removed
manually whenever necessary.
2.3. Soil sampling
Soil was sampled in March 2012, at the end of 2011 summer
crop. In each plot, one disturbed soil sample (at three different
layers; 0–5, 5–15 and 15–30 cm, using Dutch augers) and one undisturbed soil sample (0–10, 10–20 and 20–30 cm, by the volumetric
ring method) were taken (Embrapa, 1997). The soil samples were
air dried, grounded and sieved through a 2-mm sieve to remove
larger pieces of root material and the stone fraction. All soil samples were analyzed in the soil laboratory at the Federal University
of Espírito Santo, Agriculture Science Center.
2.4. Soil chemical and physical characterization
Soil chemical and physical characterization is given in Table 2.
The pH was determined on a 1:5 soil:deionised water ratio; the
potential acidity (H + Al) was extracted with Ca(OAc)2 0.5 mol L−1
buffered to pH 7.0, and quantified by titration with NaOH
0.0606 mol L−1 . Exchangeable Ca2+ , Mg2+ and Al3+ were extracted
with 1 mol L−1 KCl and Na and K were extracted with Mehlich−1
(Embrapa, 1997). The element content in the extracts were determined by atomic absorption (Ca2+ , Mg2+ and Al3+ ), flame emission
(K and Na) and photocolorimetry (P). The effective cation exchange
capacity (CECE ) was calculated by sum of cations (Ca2+ , Mg2+ , Na+ ,
147
K+ and Al3+ ) and total cation exchange capacity (CTCT ) estimated by
the sum of bases and potential acidity. The granulometric analysis
was performed by pipette method, 50 rpm, 16 h (Embrapa, 1997).
2.5. Cover crop biomass and C input
Cover crop biomass was collected inside a 1 × 1 m square in each
plot for fresh mass determination. Further, it was dried in oven
with continuous air circulation (60 ◦ C) for dry mass determination.
Total carbon of cover crop biomass was analyzed by loss in ignition at 430 ◦ C for 24 h in muffle furnace (Kiehl, 1985). A proportion
of 950 g C kg−1 biomass for white lupin and sunnhemp, 920 g C kg−1
biomass for black oat and maize and 935 g C kg−1 biomass for intercrop were found after analysis. The factor of 1.724 was used to
convert organic matter of organic compost into organic C based on
the assumption that organic matter contains 580 g C kg−1 biomass
(Carmo and Silva, 2012; Soil Survey Staff, 1996).
2.6. Soil physical attributes
Undisturbed soil samples were saturated in water for 24 h and
then placed in a sand tension table of −6 kPa. Soil microporosity
(Mic ) was calculated after stabilization of water into the volumetric ring (72 h). Bulk density (BD) was performed by the volumetric
ring method and particle density (PD) was determined by the volumetric flask method (Embrapa, 1997). Total porosity (TP) was
calculated using the following equation:
TP = 1 −
BD
PD
(1)
where BD is bulk density (g cm−3 ) and PD is particle density
(g cm−3 ). Macroporosity (Map ) was calculated as the difference
between total porosity and microporosity (Embrapa, 1997).
2.7. Soil organic carbon and nitrogen
Soil subsamples of approximately 20 g were crushed in a mortar
to pass a 250 m mesh, and then analyzed for total soil organic carbon (total organic C), total nitrogen (total N), labile carbon (Clabil )
and recalcitrant carbon (Crecal ). Total soil organic C was performed
by wet oxidation with K2 Cr2 O7 0.167 mol L−1 in the presence of
sulfuric acid with external heating (Yeomans and Bremner, 1988).
Total N was obtained by sulfuric acid digestion followed by Kjeldahl distillation (Bremmer and Mulvaney, 1982; Tedesco et al.,
1995). The fractions of soil organic C were estimated through a
modified Walkely and Black method as described by Chan et al.
(2001) using 2.5, 5 and 10 mL of concentrated H2 SO4 resulting three
acid–aqueous solution ratios of 0.25:1, 0.5:1 and 1:1 (which corresponded, respectively to 3, 6 and 9 mol L−1 H2 SO4 ). The amount
of soil organic C determined using 2.5, 5 and 10 mL of concentrated H2 SO4 when compared with total C, allowed separation of
total C into the following four fractions of decreasing oxidizability:
Fraction I (very labile) organic C oxidizable under 3 mol L−1 H2 SO4 ;
Fraction II (labile) the difference in soil organic C extracted between
6 and 3 mol L−1 H2 SO4 ; Fraction III (less labile) the difference in soil
organic C extracted between 9 and 6 mol L−1 H2 SO4 ; and Fraction IV
(non-labile) residual organic C after reaction with 9 mol L−1 H2 SO4
when compared with total C. The sum of fractions I and II corresponds to the labile C and the sum of fractions III and IV to the
recalcitrant C (Chan et al., 2001). Because of possible changes in
bulk density as a result of cropping system and organic fertilization, the C and N stocks (0–30 cm) were calculated on a mass per
unit volume basis (Ellert and Bettany, 1995), taking the soil mass
of the conventional till as control.
148
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
Table 1
Operational schedule conducted annually in the no-till and conventional till treatments from 2009 to 2013.
Soil sampling 1
Soil CO2-C emission
and soil sampling 2
-----------------------------------------------------2012---------------------------------------------------------2013----Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
------Summer----------------Fall------------------Winter----------------Spring-------------Summer-----
Winter crop - Cabbage
Cover crop sown 3
Cover crop mowed
Cabbage planting
Plowing- Rotary hoe 4
Organic compost
Hand weeding
Summer crop - Eggplant
Cover crop sown 5
Cover crop mowed
Eggplant planting
Plowing- Rotary hoe 4
Organic compost
Determination of total organic C and N, recalcitrant and labile C; 2 Determination of microbial biomass C, soluble C and water content of soil; 3 Black oat and White lupin;
Only for conventional till treatment and there was no cover crop in conventional till; 5 Maize and Sunnhemp; Dates of soil CO2 C emission and soil sampling 2 : 14/03/12;
22/05/12; 10/08/12; 2510/12; 06/02/13.
1
4
2.8. Soil CO2 C emission and soil temperature
Ln(FCO2 C) and the Tsoil is expected where soil temperature is a
limiting factor. Based on the b coefficients it is possible to derive
the Q10 factor, which represents the percentage increase in CO2 C
emission for a 10 ◦ C increase in soil temperature. This is derived as
Q10 = e10 × b (Carvalho et al., 2012).
Measurements of CO2 C emissions were made on March, May,
August, October 2012 and February 2013. CO2 C emissions were
measured using a portable LI-8100 analyzer (LiCor, EUA) coupled
to a dynamic chamber (LI-8100-102), known as survey chamber,
having 10 cm diameter placed on PVC soil collars inserted in the
soil (5 cm depth) before the experiment. Measurements were based
on six replicates in each treatment and lasted for over 1.5 min,
during which time measurements of CO2 C concentrations were
made inside the chamber at 3-s intervals. Annual CO2 C emissions were calculated based on the mean of all measurements.
Soil temperatures (5.0 cm depth) were determined during the gas
flux measurements. The relation between CO2 C (FCO2 C) and soil
temperature (Tsoil ) was described by the following equation:
FCO2 = F0 × exp(b × Tsoil ),
2.9. Soil water content and microbial biomass C
In each plot, disturbed soil samples were collected at 5 cm depth
to determinate soil water content, microbial biomass C, soluble
carbon (Csol ) and metabolic (Qmet ) and microbial quotient (Qmic ).
Soil samples were collected in March, May, August, October 2012
and February 2013. The thermogravimetric method (105–110 ◦ C
for 24 h) was used to determine soil water content (according
to Embrapa, 1997). The C content in the microbial biomass was
determined by the irradiation-extraction method (according to the
methodology developed by Ferreira et al., 1999). The C content
extracted by 0.5 M K2 SO4 (calibrated pH 6.5–6.8) in non-irradiated
samples was used to estimate soluble C. Metabolic quotient was
determined by the ratio between the soil CO2 C emission rate per
(2)
with the natural log (Ln) of the CO2 C emission we
have
Ln(FCO2 C) = Ln(F0 × exp(b × Tsoil )),
the
result
is
Ln(FCO2 C) = Ln(F0) + b × Tsoil . A linear relationship between
Table 2
Chemical and physical characterization of the soils under different management systems in the experimental site.
Treatment
pH
P
H2 O
mg dm−3
K
Na
Ca
Mg
Al
0–5 cm
Grass
Leguminous
Intercrop
Conventional till
6.40
6.44
6.43
6.51
2774.80
2882.95
3243.03
3224.14
324.00
328.67
490.00
360.50
35.33
22.83
92.33
68.00
4.15
4.61
4.76
8.04
1.56
1.42
1.74
2.43
0.00
0.00
0.00
0.00
5–15 cm
Grass
Leguminous
Intercrop
Conventional till
6.37
6.35
6.32
6.52
1676.10
1293.10
1389.63
1445.38
347.67
304.83
285.50
235.80
20.50
14.83
22.33
20.40
4.11
4.11
4.87
6.71
1.13
1.10
1.14
1.40
15–30 cm
Grass
Leguminous
Intercrop
Conventional till
6.35
6.48
6.23
6.45
778.96
661.15
475.30
672.14
230.67
285.50
247.33
143.80
11.67
6.83
3.33
5.20
3.12
3.44
2.87
3.92
0.89
0.75
0.77
1.00
CECT
V
Sand
%
g kg−1
11.86
11.48
8.16
16.25
56.45
61.32
100.00
72.22
0.00
0.00
0.00
0.00
11.09
9.83
6.83
12.51
0.00
0.00
0.00
0.00
9.08
4.95
4.29
9.78
cmolc dm−3
Silt
Clay
580.34
524.07
497.24
461.87
122.04
139.98
144.25
138.25
297.61
335.95
358.51
399.87
55.85
63.14
100.00
69.80
583.38
557.35
485.70
473.19
113.82
117.05
130.17
140.62
302.80
325.60
384.12
386.19
51.11
100.00
100.00
53.56
616.70
580.98
495.64
468.34
89.44
106.35
127.69
129.77
293.87
312.67
376.68
401.88
Grass: no-till on dead mulche of grass; leguminous: no-till on dead mulche of leguminous; intercrop: no-till on dead mulche of grass and leguminous; pH: active acidity; P:
phosphorus; K: potassium; Na: sodium; Ca: calcium; Mg: magnesium; Al: aluminum; CECT : total cation exchange capacity; V: saturation of bases.
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
Table 3
Mean values of fresh mass, dry mass production and C input during winter and
summer cover crop.
Green manure
Fresh mass
Dry mass
3. Results
3.1. Cover crop biomass and C input
C input
Mg ha−1
Winter crop
Black oat
White lupin
Intercropping
Summer crop
Maize
Sunnhemp
Intercropping
149
37.86a
28.54b
37.33a
9.09a
6.61a
8.34a
4.85a
3.65a
4.52a
63.51a
28.64c
46.21b
21.80a
10.69b
16.48ab
11.64a
5.90b
8.94ab
Means followed by the same letter, in the same column, do not differ by Tukey’s test
(p < 0.05). C input = C dry mass of cover crop + C of organic compost.
Mean values of fresh mass, dry mass production and C input
of cover crops are given in Table 3. During the winter crop, fresh
mass production of white lupin was significantly lower than black
oat and intercrop. No significant differences were recorded in winter crop for dry mass production and C input. In summer crop,
fresh mass production of maize was significantly higher than that
of sunnhemp. This result was also observed for dry mass production. The C input was significantly higher in maize plots than the
sunnhemp plots in summer crop.
3.2. Soil physical attributes
microbial biomass C unit. Microbial quotient was calculated by the
ratio between microbial biomass C and total soil organic C (Ferreira
et al., 1999).
2.10. C balance and CO2 equivalent
Carbon balance was calculated by difference between annual
average of CO2 C emissions and C input (organic compost and
green manure). As vegetables crop had similar yields and thus similar values of crop residues, the C input accounted refers to the C
of green manures and organic compost. The equivalence between
C and CO2 was based on the molecular weights of the elements, in
which one mol of CO2 contains 12.011 g C.
2.11. Data analysis
Pearson correlations were performed between soil CO2 C emissions, soil water content and soil temperature between no-till
and conventional till. Data were submitted to analysis of variance
(ANOVA) and means between treatments were compared using the
least significant difference of a Tukey test (p < 0.05) in the SAEG software (Funarbe, 2007). Split-plot analysis of variance for soil CO2 C
emission, soil temperature, soil water content, microbial biomass
C, soluble C, metabolic quotient and microbial quotient were performed. Standard error was calculated from the standard deviation
of the dataset of all replicates.
Microporosity (Mic ), macroporosity (Mac ), total porosity (TP),
bulk density (BD) and particle density (PD) values are given in
Table 4. Higher microporosity values were recorded at 0–10 cm
layer for all plots. Conventional till showed significantly lower
(p < 0.05) microporosity and higher macroporosity as compared to
the no-till treatment. There were no differences between no-till
and conventional till up to 20 cm depth for total porosity. The ratio
between macroporosity and total porosity indicates that no-till has
higher water holding capacity. Bulk density tended to increase with
soil depth.
3.3. Soil organic carbon and nitrogen
Mean values of total organic C, total N, C/N ratio, labile C and
recalcitrant C are given in Fig. 2. In general, as depth increased, total
organic C, total N, Clabil and Crecal tended to decrease. The 0–5 cm
layer had the highest C and N contents. Higher (p < 0.05) total
organic C was recorded in the intercrop treatment (50.48 g kg−1 ) as
compared to conventional till at 0–5 cm layer (43.74 g kg−1 ). There
was no statistical difference for total N among all layers evaluated.
Total N ranged from 2.81 to 5.34 g kg−1 in grass while in conventional till it ranged from 2.54 to 4.51 g kg−1 . The C/N ratio tended to
increase with increasing soil depth. Intercrop showed higher C/N
ratio for all sampled soil layers. Conventional till showed significantly lower means of Clabil as compared with grass up to 15 cm
soil depth. Higher Crecal was recorded for the intercrop when compared with grass at 0–5 and 15–30 cm layer. Crecal tended to be
higher at 5–15 cm layer for the intercrop when compared with
grass. However, no statistical significance was observed. C and
Table 4
Mean values of microporosity (Mic ), macroporosity (Mac ), total porosity (TP), bulk density (BD) and particle density (PD) among different vegetable cropping systems.
Treatment
Mic
Mac
TP
Mac /TP
m3 m−3
0–10 cm
Grass
Leguminous
Intercrop
Conventional till
10–20 cm
Grass
Leguminous
Intercrop
Conventional till
20–30 cm
Grass
Leguminous
Intercrop
Conventional till
BD
PD
g cm−3
0.47a
0.48a
0.48a
0.41b
0.16b
0.16b
0.14b
0.24a
0.63a
0.64a
0.61a
0.65a
0.25b
0.25b
0.22b
0.36a
0.98a
0.98a
0.99a
0.95a
2.70a
2.71a
2.57b
2.72a
0.42a
0.42a
0.42a
0.41a
0.16a
0.16a
0.19a
0.18a
0.57a
0.58a
0.61a
0.59a
0.27a
0.28a
0.31a
0.31a
1.15a
1.12a
1.15a
1.14a
2.72ab
2.65b
2.92a
2.81a
0.41a
0.42a
0.42a
0.42a
0.13b
0.13b
0.16ab
0.18a
0.54b
0.55ab
0.58ab
0.60a
0.24b
0.24b
0.27ab
0.31a
1.19ab
1.21a
1.19ab
1.13b
2.61b
2.73ab
2.83a
2.83a
Grass: no-till on dead mulche of grass. Leguminous: no-till on dead mulche of leguminous. Intercrop: no-till on dead mulche of grass and leguminous. Means followed by
the same letter, in the same column, do not differ by Tukey’s test (p < 0.05).
150
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
Fig. 2. Mean values (n = 6) of total organic C (a), total N (b), C/N ratio (c), labile C (d) and recalcitrant C (e) in the different planting systems. Means followed by the same
letter, did not differ by Tukey’s test (p < 0.05). Horizontal bars represent standard error of the mean. Grass: no-till on dead mulch of grass. Leguminous: no-till on dead mulch
of leguminous. Intercrop: no-till on dead mulch of grass and leguminous.
N stock values in the different vegetables planting systems are
given in Table 5. C stocks were significantly higher in the intercrop (131.2 Mg ha−1 ) when compared with the other treatments.
Conventional till showed C stock of 105 Mg ha−1 . N stock was
12.2 Mg ha−1 in grass and 10 Mg ha−1 in conventional till.
3.4. Soil CO2 C emission and soil temperature
CO2 C emissions and soil temperature values are given in Fig. 3.
Lowest CO2 C emissions were recorded in all plots during May
Table 5
Carbon and nitrogen stocks values in the different planting systems (Mg ha−1 ) in the
sampled soil profile (0–30 cm).
Treatment
Grass
Leguminous
Intercrop
Conventional till
Carbon stock
Nitrogen stock
115.8b
12.2a
110.9b
10.4a
131.2a
10.4a
105b
10a
Grass: no-till on dead mulche of grass. Leguminous: no-till on dead mulche of leguminous. Intercrop: no-till on dead mulche of grass and leguminous. Means followed
by the same letter, in the same row, do not differ by Tukey’s test (p < 0.05).
and August 2012 (Fig. 3a). Mean annual CO2 C emissions were
4.2; 3.64; 3.46 and 2.96 mol CO2 m−2 s−1 in grass, intercrop,
leguminous and conventional till, respectively. These values are
equivalent to an annual efflux of 15.89; 13.77; 13.09 and 11.20 Mg
C CO2 ha−1 year−1 , respectively. Significantly lower CO2 C emissions were recorded in the conventional till treatment during
March 2012, as compared with other treatments. CO2 C emission
values gradually increased from May 2012 to February 2013. During February 2013, the average CO2 C emissions were higher in the
conventional till, with no differences among grass and intercrop.
Soil temperature showed similar seasonal dynamics, presenting
lower averages in the winter (August 2012) and higher mean values in the summer (March 2012 and February 2013) (Fig. 3b).
Annual average soil temperature was 21.18; 21.15; 20.93 and
23.95 ◦ C for grass, leguminous, intercrop and conventional till,
respectively. Significantly higher soil temperature was recorded
in conventional till for all study periods (except for October
2012), when compared with no-till treatments. The Q10 factor
was lower in the intercrop when compared with the conventional
till (Table 6). The lowest b parameter was recorded in intercrop
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
151
Fig. 3. CO2 C emissions (a) and soil temperature (b) in the different planting systems. Same capital letters indicate no significant differences among months and same
lowercase letters represent no significant differences within months for the different treatments by Tukey’s test (p < 0.05). Vertical bars represent standard error of the mean.
Grass: no-till on dead mulch of grass. Leguminous: no-till on dead mulch of leguminous. Intercrop: no-till on dead mulche of grass and leguminous.
treatment, showing less sensitivity to increases in soil temperature.
3.5. Soil water content and microbial biomass C
Soil water content, microbial biomass C, soluble carbon (Csol ),
metabolic (Qmet ) and microbial quotient (Qmic ) are given in
Fig. 4. The annual average soil water content (g g−1 ) followed
the order: intercrop (0.28 g g−1 ) > grass (0.27 g g−1 ) > leguminous
(0.27 g g−1 ) > conventional till (0.20 g g−1 ). Significantly lower soil
water content was recorded in the conventional till, compared
with those of no-till for all study periods (Fig. 4a). There
was a significant association among soil water content, microbial biomass C, soluble C, metabolic and microbial quotient in
the five periods studied. Microbial biomass C decreased in the
colder months (from May to October 2012) and increased in the
warmer period (after October 2012), which coincided with the
higher soil temperatures (Fig. 3b) and soil water content values
(Fig. 4a).
Annual average microbial biomass C was 433.00; 378.67; 380.63
and 246.77 mg kg−1 for grass, leguminous, intercrop and conventional till, respectively. For all study periods, significantly lower
(except February 2013) microbial biomass C was recorded in conventional till, compared with those of the no-till systems (Fig. 4b).
Lower soluble C contents were recorded in August and October
2012 (Fig. 4c). Annual average of soluble C was 133.04; 147.87;
126.75 and 148.42 mg kg−1 for grass, leguminous, intercrop and
conventional till, respectively. There were no differences among
treatments for soluble C in August and October 2012. Lowest
metabolic quotient was recorded during March, May and August,
gradually increasing from May 2012 to February 2013 (Fig. 4d).
Annual average metabolic quotient was 1.58; 1.50; 1.60 and 2.01
for grass, leguminous, intercrop and conventional till. Significantly
higher metabolic quotient was recorded in the conventional till
Table 6
Parameters of the model between CO2
Treatments
Grass
Leguminous
Intercrop
Conventional till
treatment in October 2012, compared with the no-till treatments.
Significantly lower microbial quotient (except February 2013) was
recorded in conventional till. Annual average microbial quotient
was 9.69; 7.84; 7.54 and 5.64% for grass, leguminous, intercrop and
conventional till.
3.6. C balance and CO2 equivalent
C balance between annual input (cover crop and organic compost) and annual losses (CO2 C emissions) are given in Fig. 5.
High C input in no-till is contributing to positive C balance.
The difference between C input and C emitted (CO2 C emissions) was 9.65; 5.50 and 8.74 Mg ha−1 in the grass, leguminous
and intercrop treatments, respectively. C balance was negative
in conventional till (−2.15 Mg ha−1 ), even with annual input of
30 Mg ha−1 organic compost. Carbon balance represents 35.38;
20.16 and 32.04 Mg ha−1 year−1 of CO2 equivalent sequestered
for grass, leguminous and intercrop, respectively. Conventional till
showed negative balance of CO2 equivalent (7.88 Mg ha−1 year−1 ).
4. Discussion
4.1. Cover crop biomass and C input
Cover crop biomass production was significantly affected by
the season, reasonably due to the variation of climatic conditions
(Fig. 1). The average rainfall during the summer cropping cycle
(December–March) was indeed 85% higher than in winter cropping cycle (June–September). The results suggest that higher water
availability and increases in temperature (Fig. 1) contributed to the
high cover crop biomass production during the summer crop by
maize and sunnhemp, as well as C input. The amount of above
ground biomass produced is probably due to more suitable air
temperatures and rainfall which occurred throughout the cover
C emissions and soil temperature, and Q10 factor in the different planting systems during the studied period.
Ln(CO2
C emission) = a + (b × Tsoil )
a
b
R
p
Q10
1.070 ± 0.184
0.494 ± 0.193
0.947 ± 0.200
0.398 ± 0.263
0.016 ± 0.008
0.034 ± 0.009
0.015 ± 0.009
0.027 ± 0.010
0.341
0.582
0.297
0.424
0.065
Contents lists available at ScienceDirect
Scientia Horticulturae
journal homepage: www.elsevier.com/locate/scihorti
Impact of organic no-till vegetables systems on soil organic
matter in the Atlantic Forest biome
A. Thomazini a,∗ , E.S. Mendonc¸a a , J.L. Souza b , I.M. Cardoso c , M.L. Garbin a
a
b
c
Department of Plant Production, Federal University of Espírito Santo, 29500-000 Alegre, ES, Brazil
Research of INCAPER—Centro Serrano, BR-262, km 94, 29.375-000 Venda Nova do Imigrante, ES, Brazil
Soil Science Department, Federal University of Vic¸osa, Avenida P.H. Rolfs, s/n, Vicosa 36570-000, MG, Brazil
a r t i c l e
i n f o
Article history:
Received 25 August 2014
Received in revised form
25 November 2014
Accepted 1 December 2014
Keywords:
Green manure
Labile and stable fractions
Soil health
Soil C balance
a b s t r a c t
Soil organic matter is widely recognized as a strategy used to improve soil quality and reduce carbon
emissions to the atmosphere. A field study was carried out to investigate the effects of cover crops in
organic no-till vegetables systems on changes in soil organic matter and CO2 C emissions, in dry and
rainy seasons. We hypothesized that CO2 C emissions are higher in conventional till as compared with
no-till, and that no-till increases soil C sink. The crop rotation comprised a 3-year cropping sequence
involving two crops per year—cabbage (Brassica oleracea L.) in winter and eggplant (Solanum melongena
L.) in summer time. Treatments were no-till on dead mulch of grass (Avena strigosa Schreb. and Zea mays
L.), leguminous (Lupinus albus L. and Crotalaria juncea L.), intercrop (grass and leguminous) and conventional till (no dead mulch) with rotary hoe arranged in a randomized block design on a clayey Oxisol (Typic
Haplustox) at Domingos Martins-ES, Brazil. On 2012 and 2013, disturbed soil samples at three different
layers (0–5, 5–15 and 15–30 cm) and undisturbed samples at 0–10, 10–20 and 20–30 cm, for chemical
and organic matter characterization were taken. CO2 C emissions and soil temperature were measured
in situ on March, May, August and October 2012 and February 2013 (after 3 years of experiment). Conventional till site showed the lowest microporosity values and the highest macroporosity, followed by
lower soil bulk density at 0–10 cm layer. Total organic C ranged from 34.94 to 50.48 g kg−1 in intercrop
and 27.11 to 43.74 g kg−1 in conventional till. Total N ranged from 2.81 to 5.34 g kg−1 in grass and 2.54
to 4.51 g kg−1 in conventional till. Highest C stock was recorded in intercrop. Conventional till showed
lower labile C values while recalcitrant C was higher in the intercrop treatment. The annual average of
CO2 C emissions (mol CO2 m−2 s−1 ) followed the order: grass (15.89) > intercrop (13.77) > leguminous
(13.09) > conventional till (11.20). Highest annual average of soil temperature was recorded in conventional till (23.95 ◦ C). Lowest annual mean of soil water content, microbial biomass C, and highest metabolic
quotient were recorded in conventional till. These results suggest that the use of cover crops and organic
compost in pre-planting promote C increments. The contribution of organic residues increases the water
holding capacity and reduces soil temperature. No-till reduces soil disturbance and promotes a positive
balance of C. Organic no-till vegetable systems is a strategy to increase soil C and should be encouraged
in order to increase soil quality in the Atlantic Forest Biome in Brazil.
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
The Brazilian Atlantic Forest is now reduced to about 11.4 to 16%
of its original cover of approximately 150 million hectares (Ribeiro
et al., 2009). Most deforested areas are composed of agricultural
∗ Corresponding author. Tel.: +55 27 3359 3971; fax: +55 28 3552 8927.
E-mail addresses: andre.thz@gmail.com (A. Thomazini),
eduardo.mendonca@ufes.br (E.S. Mendonc¸a), jacimarsouza@yahoo.com.br
(J.L. Souza), irene@ufv.br (I.M. Cardoso), mlgarbin@gmail.com (M.L. Garbin).
http://dx.doi.org/10.1016/j.scienta.2014.12.002
0304-4238/© 2014 Elsevier B.V. All rights reserved.
systems on degraded soils. Anthropogenic activities lead to land
misuse causing changes in the physical, chemical and biological
attributes of soils (Reicosky et al., 1999; Powlson et al., 2011). This
implies decreases in the storage of organic carbon and nutrients as
well as in the productive capacity of soils, since C is an indicator
used to assess soil quality (Silva and Mendonc¸a, 2007; Ghosh et al.,
2012).
It is widely recognized that soil organic matter is one of the most
important indicators of soil quality and health (Lal, 2004; Ghosh
et al., 2012). Increasing or maintaining soil organic matter is critical to achieve optimum soil functions and crop production (Ghosh
146
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
et al., 2012). When monitoring soil quality in the tropics, sensitive
soil quality indicators need to be identified, mainly due the continuous and intensive vegetable production in these areas (Moeskops
et al., 2012). Soil management can lead to higher decomposition
rates of organic matter decreasing the concentration of this soil
component (Silva and Mendonc¸a, 2007). Agriculture can significantly contribute to elevate atmospheric CO2 concentrations as a
consequence of soil management (Powlson et al., 2011). These C
losses to the atmosphere can be mainly reduced by minimizing
soil disturbance, either with no-till or agroecological management
(Silva and Mendonc¸a, 2007). It is estimated that 89% of the potential
for mitigation of greenhouse gases produced by agriculture relies
on C sequestration (Smith et al., 2008). In addition, increasing the
soil organic C content is an important strategy to deal with climate
changes driven by C emissions to the atmosphere from agricultural
lands.
No-till and organic agriculture increase soil C and N sequestration, and reduce the oxidation of soil organic matter (Bayer et al.,
2009; Campiglia et al., 2014). Continuous input of plant residues
and paucity of soil disturbance promote reductions in CO2 C emissions through decreases in organic matter decomposition rates
(Lal, 2004; Bayer et al., 2009). On other hand, conventional crop
production intensify soil disturbance and, consequently, breakdown the soil aggregates (Bayer et al., 2009). Conventional tillage
is the most common agricultural management for vegetable production in areas formerly occupied by the Atlantic Forest in Brazil.
In addition, vegetable production is historically managed by family
smallholders. Intensive farming or intensive soil preparation in horticulture degrades the soil–plant environment, mostly due to the
reduction in concentration and quality of soil organic matter and
the diversity of soil organisms (Tian et al., 2011). Degradation of soil
organic matter leads to long-term decreases in horticultural productivity. Thus, sustainable tillage is preferable to attain a positive
net balance of C in the highly weathered tropical soils (Mendonc¸a
and Rowell, 1996).
The use of cover crops represent a potentially valuable supply
of organic residues (C source) when they are used in no-tillage
systems and their residues are left on the soil surface (Campiglia
et al., 2014). No-till systems can mitigate CO2 C emissions. This is
because crop rotation and organic residues on soil surface promote
gradual decomposition of organic matter, favoring C incorporation
(Bayer et al., 2009; Conceic¸ão et al., 2013). Physical protection of
organic matter provided by stable aggregates under no-till reduce
organic matter mineralization and lead to C accumulation (Six et al.,
2004). However, there is a lack of information about C storage gains
and CO2 C soil emissions by organic no-till vegetable systems,
especially in the areas formerly occupied by the Atlantic Forest
biome, a well-known biodiversity hotspot (Myers et al., 2000). Here,
we report the results of a long term field experiment conducted
in dry and rainy seasons. We aimed to investigate the effects of
cover crops in organic no-till vegetables systems on changes of
soil organic matter and CO2 C emissions, in dry and rainy seasons.
We hypothesized that CO2 C emissions are higher in conventional
till as compared with no-till, and that no-till increases soil C sink,
leading to improved soil quality.
2. Material and methods
2.1. Site location, characterization and land uses prior to the
experiment
The study was carried out at the 2.5 ha organic agriculture
experimental site of Incaper (Espírito Santo Institute for Research,
Technical Assistance and Rural Extension), municipality of Domingos Martins-ES (20◦ 22′ SE 41◦ 03′ W) altitude of 950 m above the
Fig. 1. Average monthly precipitation and air temperature of the municipality of
Domingos Martins between January 2012 and February 2013. Data from Incaper.
sea. The climate of the region is Aw (tropical climate and dry season in winter), precipitation ranges from 750 to 1500 mm per year,
and all months of the year have average temperatures of 18 ◦ C or
higher. The region is characterized by dry winter and rainy summer
(Köppen, 1923). Mean monthly precipitation and air temperature
are presented in Fig. 1. Soil is classified as Red-Yellow Latosol,
Brazilian Classification System (Embrapa, 2006) or as clayey Oxisol,
Typic Haplustox (Soil Taxonomy, USDA classification). From 1990 to
2009, this area was cultivated with organic vegetables (mainly lettuce, cabbage and eggplant). Organic management was performed
using 15 Mg ha−1 of organic compost (dry mass) amendments. The
composting area followed the indore system (Miller and Jones,
1995) with alternating layers stacked forming cells that received
manual eversion periodically in order to control humidity (50%)
and temperature (60 ◦ C). The method relies on aerobic activity,
although portions of the pile can become anaerobic between turnings. Moreover, it provides better control of flies, more rapid
and uniform decomposition rates and less problems regarding
moisture control (Miller and Jones, 1995). The compost was prepared with a stacked mixture of: grounded green cameron grass
(Pennisetum purpureum Schumach.), coffee husk, crop residues of
maize and beans, and inoculation with chicken manure at the
rate of 50 kg m−3 . Organic compost characteristics were (total
amount): 52% organic matter, 16:1 carbon:nitrogen ratio, 7.3 pH, 2%
nitrogen, 1.2% phosphorus, 1.2% potassium, 4.8% calcium, 0.5% magnesium, 54 mg dm−3 copper, 188 mg dm−3 zinc, 12,424 mg dm−3
iron, 793 mg dm−3 manganese, 25 mg dm−3 boron. More details of
the organic vegetable cropping (1990–2009) can be found in Souza
et al. (2012).
2.2. Experimental design, cover crops and crop rotation
The organic no-till vegetables systems experiment was initiated
in 2009. The experiment comprises four tillage systems, implemented on 4 m × 6 m plots, according to a Randomized Complete
Block Design, with six replicates (totalizing 24 permanent experimental units) covering a total area of 576 m2 . Therefore, the effects
of organic management accumulated over the years. Tillage treatments consisted of:
(i) No-till on dead mulch of grass (grass): black oat (Avena strigosa
Schreb) was used as winter cover crop followed by maize (Zea
mays L.) as summer cover crop.
(ii) No-till on dead mulch of leguminous (leguminous): white lupin
(Lupinus albus, L.) was used as winter cover crop followed by
Sunnhemp (Crotalaria juncea L.) as summer cover crop.
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
(iii) No-till on dead mulch of grass and leguminous (intercrop):
grass and leguminous plants were intercropped using the same
cover crops in grass and leguminous treatments.
(iv) Conventional plow-based tillage (Conventional till): implemented using conventional tillage with rotary hoe one week
before planting, with no cover crop. The tractor used was a rear
rotary mini tiller (Yanmar MRT-650 EX) with the rotary tines
placed right behind the wheels. This is the main vegetable cropping system of the Brazilian horticulture (Souza et al., 2012).
Operation schedule conducted annually in the no-till and conventional till were presented in Table 1. From 2009 to 2013, no-till
was performed with black oat and white lupin as winter cover
crop, followed by cabbage as winter vegetable crop. Maize and
sunnhemp worked as summer cover crop, followed by eggplant as
summer vegetable crop. Black oat and white lupin were sown on
March 2012 as winter cover crops. Cover crop seeds were spread
manually and lightly buried. Cover crops were sown in rows spaced
33 cm from each other for all treatments. The seed rates were 480 g
per plot for black oat and 660 g per plot for white lupin. In the intercropped sampling units, seeds were reduced to half of these values.
On July 2012, cover crops were mowed by mechanical mowing and
cabbage was planted. Cover crop residues were left on the soil surface as organic dead mulch and they were not incorporated into the
soil. One month old cabbage seedlings were transplanted by hand.
The cabbage seedlings were arranged in single rows distant 60 cm
from each other. The distance between the cabbage plants in the
rows was 40 cm.
After winter crop, maize and sunnhemp were sown on October
2012 as summer cover crops. The seed rates were 600 g per plot for
maize and 300 g per plot for sunnhemp. Residues were mowed on
February 2013 followed by eggplant (Solanum melongena) planting.
Eggplant seedlings were grown in tubes of 180 cm3 , using a mixture
of organic compost/soil of 1:2 as substrate. The eggplant seedlings
were arranged in single rows at a distance of 120 cm between them.
The distance between the cabbage plants in the rows was 70 cm.
Cabbage and eggplant received 15 Mg ha−1 of organic compost (dry
mass) at planting in all no-till treatments. Cabbage and eggplant
seedlings were irrigated immediately after transplanting in order
to avoid moisture stress. Inside the rows, the weeds were removed
manually whenever necessary.
2.3. Soil sampling
Soil was sampled in March 2012, at the end of 2011 summer
crop. In each plot, one disturbed soil sample (at three different
layers; 0–5, 5–15 and 15–30 cm, using Dutch augers) and one undisturbed soil sample (0–10, 10–20 and 20–30 cm, by the volumetric
ring method) were taken (Embrapa, 1997). The soil samples were
air dried, grounded and sieved through a 2-mm sieve to remove
larger pieces of root material and the stone fraction. All soil samples were analyzed in the soil laboratory at the Federal University
of Espírito Santo, Agriculture Science Center.
2.4. Soil chemical and physical characterization
Soil chemical and physical characterization is given in Table 2.
The pH was determined on a 1:5 soil:deionised water ratio; the
potential acidity (H + Al) was extracted with Ca(OAc)2 0.5 mol L−1
buffered to pH 7.0, and quantified by titration with NaOH
0.0606 mol L−1 . Exchangeable Ca2+ , Mg2+ and Al3+ were extracted
with 1 mol L−1 KCl and Na and K were extracted with Mehlich−1
(Embrapa, 1997). The element content in the extracts were determined by atomic absorption (Ca2+ , Mg2+ and Al3+ ), flame emission
(K and Na) and photocolorimetry (P). The effective cation exchange
capacity (CECE ) was calculated by sum of cations (Ca2+ , Mg2+ , Na+ ,
147
K+ and Al3+ ) and total cation exchange capacity (CTCT ) estimated by
the sum of bases and potential acidity. The granulometric analysis
was performed by pipette method, 50 rpm, 16 h (Embrapa, 1997).
2.5. Cover crop biomass and C input
Cover crop biomass was collected inside a 1 × 1 m square in each
plot for fresh mass determination. Further, it was dried in oven
with continuous air circulation (60 ◦ C) for dry mass determination.
Total carbon of cover crop biomass was analyzed by loss in ignition at 430 ◦ C for 24 h in muffle furnace (Kiehl, 1985). A proportion
of 950 g C kg−1 biomass for white lupin and sunnhemp, 920 g C kg−1
biomass for black oat and maize and 935 g C kg−1 biomass for intercrop were found after analysis. The factor of 1.724 was used to
convert organic matter of organic compost into organic C based on
the assumption that organic matter contains 580 g C kg−1 biomass
(Carmo and Silva, 2012; Soil Survey Staff, 1996).
2.6. Soil physical attributes
Undisturbed soil samples were saturated in water for 24 h and
then placed in a sand tension table of −6 kPa. Soil microporosity
(Mic ) was calculated after stabilization of water into the volumetric ring (72 h). Bulk density (BD) was performed by the volumetric
ring method and particle density (PD) was determined by the volumetric flask method (Embrapa, 1997). Total porosity (TP) was
calculated using the following equation:
TP = 1 −
BD
PD
(1)
where BD is bulk density (g cm−3 ) and PD is particle density
(g cm−3 ). Macroporosity (Map ) was calculated as the difference
between total porosity and microporosity (Embrapa, 1997).
2.7. Soil organic carbon and nitrogen
Soil subsamples of approximately 20 g were crushed in a mortar
to pass a 250 m mesh, and then analyzed for total soil organic carbon (total organic C), total nitrogen (total N), labile carbon (Clabil )
and recalcitrant carbon (Crecal ). Total soil organic C was performed
by wet oxidation with K2 Cr2 O7 0.167 mol L−1 in the presence of
sulfuric acid with external heating (Yeomans and Bremner, 1988).
Total N was obtained by sulfuric acid digestion followed by Kjeldahl distillation (Bremmer and Mulvaney, 1982; Tedesco et al.,
1995). The fractions of soil organic C were estimated through a
modified Walkely and Black method as described by Chan et al.
(2001) using 2.5, 5 and 10 mL of concentrated H2 SO4 resulting three
acid–aqueous solution ratios of 0.25:1, 0.5:1 and 1:1 (which corresponded, respectively to 3, 6 and 9 mol L−1 H2 SO4 ). The amount
of soil organic C determined using 2.5, 5 and 10 mL of concentrated H2 SO4 when compared with total C, allowed separation of
total C into the following four fractions of decreasing oxidizability:
Fraction I (very labile) organic C oxidizable under 3 mol L−1 H2 SO4 ;
Fraction II (labile) the difference in soil organic C extracted between
6 and 3 mol L−1 H2 SO4 ; Fraction III (less labile) the difference in soil
organic C extracted between 9 and 6 mol L−1 H2 SO4 ; and Fraction IV
(non-labile) residual organic C after reaction with 9 mol L−1 H2 SO4
when compared with total C. The sum of fractions I and II corresponds to the labile C and the sum of fractions III and IV to the
recalcitrant C (Chan et al., 2001). Because of possible changes in
bulk density as a result of cropping system and organic fertilization, the C and N stocks (0–30 cm) were calculated on a mass per
unit volume basis (Ellert and Bettany, 1995), taking the soil mass
of the conventional till as control.
148
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
Table 1
Operational schedule conducted annually in the no-till and conventional till treatments from 2009 to 2013.
Soil sampling 1
Soil CO2-C emission
and soil sampling 2
-----------------------------------------------------2012---------------------------------------------------------2013----Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
------Summer----------------Fall------------------Winter----------------Spring-------------Summer-----
Winter crop - Cabbage
Cover crop sown 3
Cover crop mowed
Cabbage planting
Plowing- Rotary hoe 4
Organic compost
Hand weeding
Summer crop - Eggplant
Cover crop sown 5
Cover crop mowed
Eggplant planting
Plowing- Rotary hoe 4
Organic compost
Determination of total organic C and N, recalcitrant and labile C; 2 Determination of microbial biomass C, soluble C and water content of soil; 3 Black oat and White lupin;
Only for conventional till treatment and there was no cover crop in conventional till; 5 Maize and Sunnhemp; Dates of soil CO2 C emission and soil sampling 2 : 14/03/12;
22/05/12; 10/08/12; 2510/12; 06/02/13.
1
4
2.8. Soil CO2 C emission and soil temperature
Ln(FCO2 C) and the Tsoil is expected where soil temperature is a
limiting factor. Based on the b coefficients it is possible to derive
the Q10 factor, which represents the percentage increase in CO2 C
emission for a 10 ◦ C increase in soil temperature. This is derived as
Q10 = e10 × b (Carvalho et al., 2012).
Measurements of CO2 C emissions were made on March, May,
August, October 2012 and February 2013. CO2 C emissions were
measured using a portable LI-8100 analyzer (LiCor, EUA) coupled
to a dynamic chamber (LI-8100-102), known as survey chamber,
having 10 cm diameter placed on PVC soil collars inserted in the
soil (5 cm depth) before the experiment. Measurements were based
on six replicates in each treatment and lasted for over 1.5 min,
during which time measurements of CO2 C concentrations were
made inside the chamber at 3-s intervals. Annual CO2 C emissions were calculated based on the mean of all measurements.
Soil temperatures (5.0 cm depth) were determined during the gas
flux measurements. The relation between CO2 C (FCO2 C) and soil
temperature (Tsoil ) was described by the following equation:
FCO2 = F0 × exp(b × Tsoil ),
2.9. Soil water content and microbial biomass C
In each plot, disturbed soil samples were collected at 5 cm depth
to determinate soil water content, microbial biomass C, soluble
carbon (Csol ) and metabolic (Qmet ) and microbial quotient (Qmic ).
Soil samples were collected in March, May, August, October 2012
and February 2013. The thermogravimetric method (105–110 ◦ C
for 24 h) was used to determine soil water content (according
to Embrapa, 1997). The C content in the microbial biomass was
determined by the irradiation-extraction method (according to the
methodology developed by Ferreira et al., 1999). The C content
extracted by 0.5 M K2 SO4 (calibrated pH 6.5–6.8) in non-irradiated
samples was used to estimate soluble C. Metabolic quotient was
determined by the ratio between the soil CO2 C emission rate per
(2)
with the natural log (Ln) of the CO2 C emission we
have
Ln(FCO2 C) = Ln(F0 × exp(b × Tsoil )),
the
result
is
Ln(FCO2 C) = Ln(F0) + b × Tsoil . A linear relationship between
Table 2
Chemical and physical characterization of the soils under different management systems in the experimental site.
Treatment
pH
P
H2 O
mg dm−3
K
Na
Ca
Mg
Al
0–5 cm
Grass
Leguminous
Intercrop
Conventional till
6.40
6.44
6.43
6.51
2774.80
2882.95
3243.03
3224.14
324.00
328.67
490.00
360.50
35.33
22.83
92.33
68.00
4.15
4.61
4.76
8.04
1.56
1.42
1.74
2.43
0.00
0.00
0.00
0.00
5–15 cm
Grass
Leguminous
Intercrop
Conventional till
6.37
6.35
6.32
6.52
1676.10
1293.10
1389.63
1445.38
347.67
304.83
285.50
235.80
20.50
14.83
22.33
20.40
4.11
4.11
4.87
6.71
1.13
1.10
1.14
1.40
15–30 cm
Grass
Leguminous
Intercrop
Conventional till
6.35
6.48
6.23
6.45
778.96
661.15
475.30
672.14
230.67
285.50
247.33
143.80
11.67
6.83
3.33
5.20
3.12
3.44
2.87
3.92
0.89
0.75
0.77
1.00
CECT
V
Sand
%
g kg−1
11.86
11.48
8.16
16.25
56.45
61.32
100.00
72.22
0.00
0.00
0.00
0.00
11.09
9.83
6.83
12.51
0.00
0.00
0.00
0.00
9.08
4.95
4.29
9.78
cmolc dm−3
Silt
Clay
580.34
524.07
497.24
461.87
122.04
139.98
144.25
138.25
297.61
335.95
358.51
399.87
55.85
63.14
100.00
69.80
583.38
557.35
485.70
473.19
113.82
117.05
130.17
140.62
302.80
325.60
384.12
386.19
51.11
100.00
100.00
53.56
616.70
580.98
495.64
468.34
89.44
106.35
127.69
129.77
293.87
312.67
376.68
401.88
Grass: no-till on dead mulche of grass; leguminous: no-till on dead mulche of leguminous; intercrop: no-till on dead mulche of grass and leguminous; pH: active acidity; P:
phosphorus; K: potassium; Na: sodium; Ca: calcium; Mg: magnesium; Al: aluminum; CECT : total cation exchange capacity; V: saturation of bases.
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
Table 3
Mean values of fresh mass, dry mass production and C input during winter and
summer cover crop.
Green manure
Fresh mass
Dry mass
3. Results
3.1. Cover crop biomass and C input
C input
Mg ha−1
Winter crop
Black oat
White lupin
Intercropping
Summer crop
Maize
Sunnhemp
Intercropping
149
37.86a
28.54b
37.33a
9.09a
6.61a
8.34a
4.85a
3.65a
4.52a
63.51a
28.64c
46.21b
21.80a
10.69b
16.48ab
11.64a
5.90b
8.94ab
Means followed by the same letter, in the same column, do not differ by Tukey’s test
(p < 0.05). C input = C dry mass of cover crop + C of organic compost.
Mean values of fresh mass, dry mass production and C input
of cover crops are given in Table 3. During the winter crop, fresh
mass production of white lupin was significantly lower than black
oat and intercrop. No significant differences were recorded in winter crop for dry mass production and C input. In summer crop,
fresh mass production of maize was significantly higher than that
of sunnhemp. This result was also observed for dry mass production. The C input was significantly higher in maize plots than the
sunnhemp plots in summer crop.
3.2. Soil physical attributes
microbial biomass C unit. Microbial quotient was calculated by the
ratio between microbial biomass C and total soil organic C (Ferreira
et al., 1999).
2.10. C balance and CO2 equivalent
Carbon balance was calculated by difference between annual
average of CO2 C emissions and C input (organic compost and
green manure). As vegetables crop had similar yields and thus similar values of crop residues, the C input accounted refers to the C
of green manures and organic compost. The equivalence between
C and CO2 was based on the molecular weights of the elements, in
which one mol of CO2 contains 12.011 g C.
2.11. Data analysis
Pearson correlations were performed between soil CO2 C emissions, soil water content and soil temperature between no-till
and conventional till. Data were submitted to analysis of variance
(ANOVA) and means between treatments were compared using the
least significant difference of a Tukey test (p < 0.05) in the SAEG software (Funarbe, 2007). Split-plot analysis of variance for soil CO2 C
emission, soil temperature, soil water content, microbial biomass
C, soluble C, metabolic quotient and microbial quotient were performed. Standard error was calculated from the standard deviation
of the dataset of all replicates.
Microporosity (Mic ), macroporosity (Mac ), total porosity (TP),
bulk density (BD) and particle density (PD) values are given in
Table 4. Higher microporosity values were recorded at 0–10 cm
layer for all plots. Conventional till showed significantly lower
(p < 0.05) microporosity and higher macroporosity as compared to
the no-till treatment. There were no differences between no-till
and conventional till up to 20 cm depth for total porosity. The ratio
between macroporosity and total porosity indicates that no-till has
higher water holding capacity. Bulk density tended to increase with
soil depth.
3.3. Soil organic carbon and nitrogen
Mean values of total organic C, total N, C/N ratio, labile C and
recalcitrant C are given in Fig. 2. In general, as depth increased, total
organic C, total N, Clabil and Crecal tended to decrease. The 0–5 cm
layer had the highest C and N contents. Higher (p < 0.05) total
organic C was recorded in the intercrop treatment (50.48 g kg−1 ) as
compared to conventional till at 0–5 cm layer (43.74 g kg−1 ). There
was no statistical difference for total N among all layers evaluated.
Total N ranged from 2.81 to 5.34 g kg−1 in grass while in conventional till it ranged from 2.54 to 4.51 g kg−1 . The C/N ratio tended to
increase with increasing soil depth. Intercrop showed higher C/N
ratio for all sampled soil layers. Conventional till showed significantly lower means of Clabil as compared with grass up to 15 cm
soil depth. Higher Crecal was recorded for the intercrop when compared with grass at 0–5 and 15–30 cm layer. Crecal tended to be
higher at 5–15 cm layer for the intercrop when compared with
grass. However, no statistical significance was observed. C and
Table 4
Mean values of microporosity (Mic ), macroporosity (Mac ), total porosity (TP), bulk density (BD) and particle density (PD) among different vegetable cropping systems.
Treatment
Mic
Mac
TP
Mac /TP
m3 m−3
0–10 cm
Grass
Leguminous
Intercrop
Conventional till
10–20 cm
Grass
Leguminous
Intercrop
Conventional till
20–30 cm
Grass
Leguminous
Intercrop
Conventional till
BD
PD
g cm−3
0.47a
0.48a
0.48a
0.41b
0.16b
0.16b
0.14b
0.24a
0.63a
0.64a
0.61a
0.65a
0.25b
0.25b
0.22b
0.36a
0.98a
0.98a
0.99a
0.95a
2.70a
2.71a
2.57b
2.72a
0.42a
0.42a
0.42a
0.41a
0.16a
0.16a
0.19a
0.18a
0.57a
0.58a
0.61a
0.59a
0.27a
0.28a
0.31a
0.31a
1.15a
1.12a
1.15a
1.14a
2.72ab
2.65b
2.92a
2.81a
0.41a
0.42a
0.42a
0.42a
0.13b
0.13b
0.16ab
0.18a
0.54b
0.55ab
0.58ab
0.60a
0.24b
0.24b
0.27ab
0.31a
1.19ab
1.21a
1.19ab
1.13b
2.61b
2.73ab
2.83a
2.83a
Grass: no-till on dead mulche of grass. Leguminous: no-till on dead mulche of leguminous. Intercrop: no-till on dead mulche of grass and leguminous. Means followed by
the same letter, in the same column, do not differ by Tukey’s test (p < 0.05).
150
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
Fig. 2. Mean values (n = 6) of total organic C (a), total N (b), C/N ratio (c), labile C (d) and recalcitrant C (e) in the different planting systems. Means followed by the same
letter, did not differ by Tukey’s test (p < 0.05). Horizontal bars represent standard error of the mean. Grass: no-till on dead mulch of grass. Leguminous: no-till on dead mulch
of leguminous. Intercrop: no-till on dead mulch of grass and leguminous.
N stock values in the different vegetables planting systems are
given in Table 5. C stocks were significantly higher in the intercrop (131.2 Mg ha−1 ) when compared with the other treatments.
Conventional till showed C stock of 105 Mg ha−1 . N stock was
12.2 Mg ha−1 in grass and 10 Mg ha−1 in conventional till.
3.4. Soil CO2 C emission and soil temperature
CO2 C emissions and soil temperature values are given in Fig. 3.
Lowest CO2 C emissions were recorded in all plots during May
Table 5
Carbon and nitrogen stocks values in the different planting systems (Mg ha−1 ) in the
sampled soil profile (0–30 cm).
Treatment
Grass
Leguminous
Intercrop
Conventional till
Carbon stock
Nitrogen stock
115.8b
12.2a
110.9b
10.4a
131.2a
10.4a
105b
10a
Grass: no-till on dead mulche of grass. Leguminous: no-till on dead mulche of leguminous. Intercrop: no-till on dead mulche of grass and leguminous. Means followed
by the same letter, in the same row, do not differ by Tukey’s test (p < 0.05).
and August 2012 (Fig. 3a). Mean annual CO2 C emissions were
4.2; 3.64; 3.46 and 2.96 mol CO2 m−2 s−1 in grass, intercrop,
leguminous and conventional till, respectively. These values are
equivalent to an annual efflux of 15.89; 13.77; 13.09 and 11.20 Mg
C CO2 ha−1 year−1 , respectively. Significantly lower CO2 C emissions were recorded in the conventional till treatment during
March 2012, as compared with other treatments. CO2 C emission
values gradually increased from May 2012 to February 2013. During February 2013, the average CO2 C emissions were higher in the
conventional till, with no differences among grass and intercrop.
Soil temperature showed similar seasonal dynamics, presenting
lower averages in the winter (August 2012) and higher mean values in the summer (March 2012 and February 2013) (Fig. 3b).
Annual average soil temperature was 21.18; 21.15; 20.93 and
23.95 ◦ C for grass, leguminous, intercrop and conventional till,
respectively. Significantly higher soil temperature was recorded
in conventional till for all study periods (except for October
2012), when compared with no-till treatments. The Q10 factor
was lower in the intercrop when compared with the conventional
till (Table 6). The lowest b parameter was recorded in intercrop
A. Thomazini et al. / Scientia Horticulturae 182 (2015) 145–155
151
Fig. 3. CO2 C emissions (a) and soil temperature (b) in the different planting systems. Same capital letters indicate no significant differences among months and same
lowercase letters represent no significant differences within months for the different treatments by Tukey’s test (p < 0.05). Vertical bars represent standard error of the mean.
Grass: no-till on dead mulch of grass. Leguminous: no-till on dead mulch of leguminous. Intercrop: no-till on dead mulche of grass and leguminous.
treatment, showing less sensitivity to increases in soil temperature.
3.5. Soil water content and microbial biomass C
Soil water content, microbial biomass C, soluble carbon (Csol ),
metabolic (Qmet ) and microbial quotient (Qmic ) are given in
Fig. 4. The annual average soil water content (g g−1 ) followed
the order: intercrop (0.28 g g−1 ) > grass (0.27 g g−1 ) > leguminous
(0.27 g g−1 ) > conventional till (0.20 g g−1 ). Significantly lower soil
water content was recorded in the conventional till, compared
with those of no-till for all study periods (Fig. 4a). There
was a significant association among soil water content, microbial biomass C, soluble C, metabolic and microbial quotient in
the five periods studied. Microbial biomass C decreased in the
colder months (from May to October 2012) and increased in the
warmer period (after October 2012), which coincided with the
higher soil temperatures (Fig. 3b) and soil water content values
(Fig. 4a).
Annual average microbial biomass C was 433.00; 378.67; 380.63
and 246.77 mg kg−1 for grass, leguminous, intercrop and conventional till, respectively. For all study periods, significantly lower
(except February 2013) microbial biomass C was recorded in conventional till, compared with those of the no-till systems (Fig. 4b).
Lower soluble C contents were recorded in August and October
2012 (Fig. 4c). Annual average of soluble C was 133.04; 147.87;
126.75 and 148.42 mg kg−1 for grass, leguminous, intercrop and
conventional till, respectively. There were no differences among
treatments for soluble C in August and October 2012. Lowest
metabolic quotient was recorded during March, May and August,
gradually increasing from May 2012 to February 2013 (Fig. 4d).
Annual average metabolic quotient was 1.58; 1.50; 1.60 and 2.01
for grass, leguminous, intercrop and conventional till. Significantly
higher metabolic quotient was recorded in the conventional till
Table 6
Parameters of the model between CO2
Treatments
Grass
Leguminous
Intercrop
Conventional till
treatment in October 2012, compared with the no-till treatments.
Significantly lower microbial quotient (except February 2013) was
recorded in conventional till. Annual average microbial quotient
was 9.69; 7.84; 7.54 and 5.64% for grass, leguminous, intercrop and
conventional till.
3.6. C balance and CO2 equivalent
C balance between annual input (cover crop and organic compost) and annual losses (CO2 C emissions) are given in Fig. 5.
High C input in no-till is contributing to positive C balance.
The difference between C input and C emitted (CO2 C emissions) was 9.65; 5.50 and 8.74 Mg ha−1 in the grass, leguminous
and intercrop treatments, respectively. C balance was negative
in conventional till (−2.15 Mg ha−1 ), even with annual input of
30 Mg ha−1 organic compost. Carbon balance represents 35.38;
20.16 and 32.04 Mg ha−1 year−1 of CO2 equivalent sequestered
for grass, leguminous and intercrop, respectively. Conventional till
showed negative balance of CO2 equivalent (7.88 Mg ha−1 year−1 ).
4. Discussion
4.1. Cover crop biomass and C input
Cover crop biomass production was significantly affected by
the season, reasonably due to the variation of climatic conditions
(Fig. 1). The average rainfall during the summer cropping cycle
(December–March) was indeed 85% higher than in winter cropping cycle (June–September). The results suggest that higher water
availability and increases in temperature (Fig. 1) contributed to the
high cover crop biomass production during the summer crop by
maize and sunnhemp, as well as C input. The amount of above
ground biomass produced is probably due to more suitable air
temperatures and rainfall which occurred throughout the cover
C emissions and soil temperature, and Q10 factor in the different planting systems during the studied period.
Ln(CO2
C emission) = a + (b × Tsoil )
a
b
R
p
Q10
1.070 ± 0.184
0.494 ± 0.193
0.947 ± 0.200
0.398 ± 0.263
0.016 ± 0.008
0.034 ± 0.009
0.015 ± 0.009
0.027 ± 0.010
0.341
0.582
0.297
0.424
0.065