Available sulphur by different extractants in soils of the state of Ceará, Brazil¹

Enxofre disponível por diferentes extratores em solos do estado do Ceará, Brasil

ABSTRACT

Despite the relevance of sulphur (S) to plant development, studies in Brazil on its availability in the soil are restricted to a few crops and regions. The aim of this study was to compare extractants for assessing the availability of S, and establish critical levels for soils in the state of Ceará. In the laboratory, available S was extracted from 23 soils (0 - 0.20 m), using two extractants: a solution of Ca(H₂PO₄)₂, 500 mg L^-1 P in H₂O (MCP) and NH₄OAc in HOAc (AMA). In an experiment set up in a greenhouse, forage sorghum ‘EA116’ (Sorghum vulgari Pers.) was used as an indicator plant in two successive crops. A completely randomised experimental design was employed, in a 23 x 2 factorial scheme with three replications, where the factors corresponded to the soils and to the doses of S (0 and 80 mg per experimental unit). The MCP was more efficient in extracting available S than was the AMA, although the S content obtained by both did show a significant correlation (r = 0.96). The available S using MCP showed a better correlation, with an increase in production and in absorbed S, suggesting MCP to be the most suitable extractant. The critical levels for S were 3.90 and 3.40 mg dm^-3 for MCP and AMA, respectively.

Keywords:
Extraction methods; Critical level; Leaf S content

RESUMO

Apesar da relevância do enxofre (S) para o desenvolvimento das plantas, estudos sobre sua disponibilidade no solo são restritos à algumas culturas e regiões do Brasil. Assim, objetivou-se com este trabalho comparar extratores para avaliar a disponibilidade de S e estabelecer níveis críticos em solos do estado do Ceará. Em laboratório foi extraído o S disponível de 23 solos (0 - 0,20 m), empregando-se dois extratores: soluções de Ca(H₂PO₄)₂, 500 mg L^-1 de P, em H₂O (MCP) e NH₄OAc em HOAc (AMA). Em experimento instalado em casa de vegetação foi usado, como planta indicadora, o sorgo forrageiro (Sorghum vulgari Pers.), cultivar EA116, em dois cultivos sucessivos. O delineamento experimental foi inteiramente casualizado, em esquema fatorial (23 x 2), com três repetições. Os termos do fatorial corresponderam aos solos e doses de S (0 e 80 mg por unidade experimental). O MCP apresentou maior eficiência de extração de S disponível que o AMA, embora o teor de S obtido por ambos tenha apresentado correlação significativa (r = 0,96). O S disponível por MCP apresentou melhor correlação com incremento de produção e S absorvido, sugerindo ser o mais indicado. Os níveis críticos de S foram de 3,90 e 3,40 mg dm^-3 para MCP e AMA, respectivamente.

Keywords:
Extraction methods; Critical level; Leaf S content

INTRODUCTION

Sulphur (S) is considered the fourth most important nutrient for plants, after N, P and K (JAMAL; MOON; ABDIN, 2010), as it is a component of several proteins, enzymes and cofactors (SOLOMON et al., 2011). The neglect of this nutrient in fertiliser recommendations, and the drastic reduction in input from the atmosphere (DIVITO et al., 2015; VIEIRA-FILHO; LEHMANN; FORNARO, 2015), have been the cause of S deficiency in plants, limiting productivity and reducing product quality (JOHNSON et al., 2018; SCHERER, 2009).

The SO₄^2- ion is the form absorbed by plants. Adsorption and desorption processes in the soil are mainly controlled by the SO₄^2- concentration in the soil solution, the pH, characteristics of the colloidal surfaces, and the concentration of other anions in solution (SCHERER, 2009; ZHAO et al., 2017). Due to these processes, SO₄^2- is found in the soil at various degrees of availability for plants.

There are various extractants and techniques for assessing available S, but the difficulties in establishing critical levels for soil S by the current methods of analysis are due not only to the loss of efficiency of the method itself (NOVAIS et al., 2015), but also to other factors, such as forms of S not accessed by the extraction method, desorption rates compatible with the needs of the plant, absorption at deeper layers, mineralisation rate of the residue and organic matter, and the entry of SO₄^2- with rainwater or irrigation, masking the plant's response to S fertilisation (RAMPIM et al., 2011; TIECHER et al., 2012).

The availability of S in the soil and the requirements of the plant regulate the processes of SO₄^2- absorption and assimilation (TAKAHASHI et al., 2011), underlining the need to estimate the available soil S so as to correct deficiencies and favour the metabolic processes of the crops.

Standing out among the available extractants for S are 0.5 mol L^-1 ammonium acetate diluted in 0.25 mol L^-1 acetic acid (NH₄OAc), and 0.01 mol L^-1 calcium phosphate (Ca(H₂PO₄)₂), considered similar in extraction ability in the 0 - 0.20 m layer in maize and wheat crops (BLUM et al., 2014). According to Barrow and Debnath (2015), the adsorption force of phosphate is considered superior to that of SO₄^2-, which justifies the use of phosphate solutions to displace the adsorbed SO₄^2-. Fernandes, Freire and Oliveira (2007) found that ammonium acetate extractant in acetic acid extracted more S from clayey soils with higher levels of organic matter, while monocalcium phosphate diluted in acetic acid was efficient in extracting sulphur regardless of the physical or chemical characteristics of the soil.

Despite the importance of S for plants, there is limited information on the critical levels of this element for soils in different regions, especially under semi-arid conditions. It is therefore necessary to carry out studies on the availability of sulphur for plants and its relationship with the characteristics of the soil. Based on the hypothesis that the SO₄^2- ion can be extracted with greater or lesser efficiency by different solutions, the aim of this study was to compare the effectiveness of two extractants in assessing the availability of sulphur, and to establish critical levels for representative soils of the semi-arid region of the state of Ceará.

MATERIAL AND METHODS

An experiment was conducted in a greenhouse using soil samples with a wide range of physical and chemical properties belonging to the following classes: Haplic Cambisol, Quartzarenic Neosol, Litholic Neosol, Haplic Planosol, Red-Yellow Ultisol and Red-Yellow Oxisol.

Soil samples were collected from the 0 - 0.20 m surface layer in the districts of Limoeiro do Norte (4 samples), São João do Jaguaribe (1 sample), Itapebuçu (1 sample), Russas (1 sample)), Palhano (2 samples), Palmaceae (1 sample), Aracoiaba (1 sample), Mulungu (2 samples), Baturité (1 sample), Redenção (1 sample), Itaitinga (1 sample), Ipú (1 sample). Tianguá (1 sample), Viçosa do Ceará (1 sample), Ubajara (1 sample), Pentecoste (1 sample) and Pacajus (2 samples). After collection, the samples were air-dried and passed through a 2-mm sieve (ADFS) for physical and chemical analysis (Table 1).

Table 1
Chemical and physical attributes of the soil samples under study

1/Soil	2/pH	3/EC	4/Ca+Mg	5/Na	5/K	6/H+Al	7/Ex	8/V	9/N	10/OM	11/Snd	Slt	Ds	P
		dS m-1	______________ (cmolc dm-3)__________________					__________________ (%)__________________					g cm-3	mg dm-3
1	6.0	0.32	6.7	0.14	0.69	2.3	9.8	76	0.08	1.4	33	42	1.14	1.0
2	6.3	0.35	5.2	0.14	0.39	1.8	7.5	76	0.12	1.5	53	27	1.26	1.0
3	6.5	0.34	11.2	0.22	0.16	1.8	13.4	86	0.05	0.7	46	13	1.40	73.0
4	5.3	0.30	---	0.13	0.16	1.7	2.0	14	0.04	0.8	63	2	1.66	3.0
5	5.7	0.69	7.0	0.43	0.26	3.9	11.6	66	0.17	3.1	43	12	1.13	24.0
6	4.7	0.35	---	0.11	0.05	1.5	1.7	9	0.02	0.4	87	2	1.74	1.0
7	5.3	0.32	3.2	0.21	0.37	1.8	5.8	68	0.04	0.7	72	11	1.49	1.0
8	5.0	0.31	8.5	0.43	0.15	4.0	13.1	69	0.04	0.8	21	28	1.21	2.0
9	5.7	0.37	3.7	0.15	0.30	3.0	7.2	58	0.09	1.7	71	11	1.31	25.0
10	5.0	0.33	3.3	0.18	0.22	3.8	7.5	50	0.07	1.5	68	9	1.41	8.0
11	5.2	0.24	---	0.12	0.15	2.0	2.3	11	0.03	0.7	87	3	1.53	5.0
12	5.4	0.92	4.3	0.30	0.57	3.5	8.7	60	0.14	2.5	58	17	1.23	14.0
13	4.5	0.47	1.0	0.16	0.26	5.8	7.4	21	0.12	2.8	59	17	1.23	7.0
14	5.5	0.74	5.9	0.16	0.40	4.0	9.9	60	0.19	3.5	63	15	1.17	6.0
15	5.5	0.16	---	0.12	0.14	1.3	1.6	16	0.02	0.3	81	6	1.63	2.0
16	4.7	0.19	---	0.10	0.05	1.7	1.9	8	0.01	0.4	93	1	1.47	2.0
17	5.0	0.18	---	0.12	0.04	2.2	2.4	6	0.01	0.5	91	1	1.61	2.0
18	5.6	0.23	---	0.12	0.10	2.3	2.5	8	0.04	0.8	82	2	1.58	5.0
19	3.8	0.20	---	0.04	0.04	4.5	4.6	1	0.06	1.2	76	12	1.32	5.0
20	5.0	0.16	---	0.04	0.11	2.3	2.5	6	0.06	0.8	78	10	1.43	9.0
21	3.8	0.28	---	0.03	0.04	3.8	3.9	2	0.04	5.0	79	7	1.33	10.0
22	5.8	0.20	---	0.10	0.43	2.4	7.0	67	0.12	0.3	79	5	1.37	17.0
23	5.0	0.16	---	0.10	0.06	5.9	7.5	22	0.14	2.4	75	11	1.30	20.0
Amplitude	3.8- 6.5	0.16 - 0.90	-- - 9.0	0.03 - 0.43	0.04 - 0.69	1.3 - 5.9	1.6 - 13.4	1 - 86	0.01 - 0.19	0.3 - 5.0	21 - 93	1 - 42	1.13 - 1.74	1.0 - 73
Mean	5.2	0.34	5.45	0.16	0.22	2.9	6.2	37	0.07	1.5	68	11	1.39

• ^1/
  1 and 2 - CXbe - Eutrophic Tb Haplic Cambisol; 3 and 8 - RRe - Eutrophic Regolithic Neosol; 4, 6, 13, 17 and 18 - PVAd - Dystrophic Red-Yellow Ultisol; 5 - RLd - Eutrophic Litholic Neosol; 7 and 12- SXe - Haplic Planosol; 10, 11 and 14 - PVAe - Eutrophic Red-Yellow Argisol; 15, 16 and 21 - RQo - Dystrophic Quartzarenic Neosol; 19, 20 and 23 - LVAd - Dystrophic Red-Yellow Latosol; 9 and 22 - RL - Litholic Neosol.
• ^2/
  pH in H₂O (1:2.5);
• ^3/
  EC - Electrical conductivity in dS m^-1- determined in the saturation extract and measured with a soilbridge salt bridge;
• ^4/
  KCl 1 mol L
• ^-1
  extractant in cmol_c dm^-3;
• ^5/
  Mehlich-1 extractant, with Na and K determined by flame photometry, and P determined by colorimetry;
• ^6/
  0.5 mol L^-1 Ca(OAc)₂ extractant, pH 7.0;
• ^7/
  Ex - Cation exchange capacity;
• ^8/
  V - Base saturation (%);
• ^9/
  Total nitrogen, Kjedahl method;
• ^10/
  OM - Organic matter, Walkley-Black method;
• ^11/
  Sand and Clay - pipette method;
• ^12/
  Bulk density - Test-tube method.

The statistical design was completely randomised, in a 23 x 2 factorial scheme, with three replications, giving a total of 138 experimental units. The factors corresponded to 23 soils and two doses of S (0 and 80 mg dm^-3). The experimental unit comprised a black polyethylene plastic bag with a capacity of 2 dm^-3, containing 1.6 dm^-3 of soil and five sorghum plants. The weight of the soil in the experimental units ranged from 1.8 to 2.8 kg, due to the density of the different soil classes used.

In the greenhouse, the soils received a mixture of CaCO³ and MgCO³ at a molar ratio of 3:1 whenever necessary and after homogenisation and wetting were incubated for 30 days. The dose of soil corrective was estimated so as to reach 70% base saturation or 2 cmol_c dm^-3 of exchangeable Ca + Mg. During incubation, the moisture in the samples was kept close to 80% of field capacity by weighing each week.

At the end of the incubation period, S was applied according to each treatment, using CaSO₄.2H₂O in the form of p.a. reagent. The soils were again homogenised, wetted and incubated for another 10 days, at which time planting was carried out. The treatments with no S received calcium in the form of CaCl₂ in the same proportion as the soils that received the sulphur.

Forage sorghum ‘EA116’ (Sorghum vulgari Pers.) was used as an indicator plant. Sorghum is grown in very dry and/or very hot areas and environmental conditions, where the productivity of other cereals is uneconomical (GALVÃO et al., 2015). Another important aspect of forage sorghum is its regrowth capacity, which can reach a production of up to 60% compared to the first cut (VON PINHO et al., 2007). The crop is therefore adapted to the semi-arid conditions of Ceará. Furthermore, due to regrowth, there was no need to sow the second crop in the experiment. Ten seeds were sown in each experimental unit. Six days after germination, the plants were thinned, leaving five plants per unit. The experiment was irrigated daily with distilled water.

After thinning, each experimental unit was fertilised based on the nutrient requirements of the crop. The N and K were divided into three equal doses at 6, 15 and 25 days. The reagents and doses, in mg per experimental unit, were NH₄NO₃ (380) and KCl (120-400, based on the K content of the soil). All the P was applied in the form of NaH₂PO₄, (250-750 mg per experimental unit, based on the soil content). Ca and S were applied using CaSO₄.2H₂O and CaCl₂.2H₂O (180 and 100 mg per experimental unit, respectively). Micronutrients were applied (mg per experimental unit) using the following reagents and doses: H₃BO₃ (1.40), MnCl₂.4H₂O (5.50), ZnCl₂ (6.00), CuCl (2.00) and H₂MoO₄.H₂O (0.29).

Forty days after sowing, the first cut of the shoots was carried out 5 cm from the ground. A second crop, from the sorghum regrowth, was maintained for 40 days. The second crop also received N, P and K fertiliser and micronutrients using the same doses and reagents. The material collected from each cut was placed in paper bags and taken to the laboratory to be dried in an oven.

The available S content in the soils under study was determined in the laboratory with the use of two extractants: 1) NH₄OAc 0.5 mol L^-1 in HOAc 0.25 mol L^-1 (ammonium acetate in acetic acid - AMA) (BARDSLEY; LANCASTER, 1960). A soil to solution ratio of 1:2.5 (10 g of soil and 25 mL of extractant solution) was used and stirred for 30 minutes, after which 0.25 g of activated charcoal was added and stirred for a further 3 minutes. The suspension was then filtered, and a 10 ml aliquot was transferred to measuring tubes which received 1 ml of 6N HCl with 20 ppm S and 0.5 g of barium chloride. A turbidimetric determination of the S in the samples was carried out within 2 to 8 minutes after the barium chloride had dissolved, using a molecular absorption spectrophotometer at a wavelength of 420 nm. 2) Ca(H₂PO₄)₂ in water, with 500 mg P dm^-3 (monocalcium phosphate in water - MCP) (FOX; OLSON; RHOADS, 1964). For this extractant, a soil to solution ratio of 1:5 was used (20 g of soil and 100 mL of extractant), stirred for 30 min and decanted for 12 to 15 h. From this suspension, 40 ml were removed and concentrated by evaporation at 120°C until dry. The residue was digested on a hot plate until white smoke was produced, when the temperature was reduced, and digestion continued for another 15 min. After cooling, 10 ml of distilled water and 1 ml of gum acacia solution in acetic acid were added. The S was determined by turbidimetry, using a spectrophotometer at a wavelength of 420 nm.

From the material collected in the greenhouse, the dry matter weight (DMW) was evaluated in the aerial part of the sorghum (g per experimental unit) obtained after drying to constant weight at 60°C. From the DMW, the relative production (RP) and increase in production (IP) were determined using Equations 1 and 2.

(Eq. 1)$RP=\frac{DMW production with no S}{DMW production with S} x 100$

(Eq. 2)$\mathrm{I}\mathrm{P}=100-\mathrm{R}\mathrm{P}$

With the dry matter ground and passed through a 0.84-mm sieve, the following were determined in perchloric-nitric extract: S by turbidimetry, P by colorimetry and K by flame photometry. N was determined by semi-micro-Kjeldahl distillation in sulphuric extract. The absorption of the nutrients under analysis was determined with the formula: mg of extracted nutrient = % nutrient x dry matter (g per experimental unit) x 10.

The critical levels for S in both the soil and plant were determined by means of the Cate and Nelson graphical method (1965), using the data for available soil S (X) and relative production (Y), and in the treatment with no S, data for the S content of the plant (X) and relative production (Y).

The data from both sorghum crops were submitted to analysis of variance. The relationships between the different attributes evaluated in the soil and in the plant were analysed by simple linear correlation.

RESULTS AND DISCUSSION

Table 2 shows the S content in the soils under natural conditions (without the addition of S), using two extractants. The contents obtained with MCP were greater than those obtained with AMA. The superiority of MCP can be explained by the phosphate ion having greater power to displace the adsorbed sulphate than does the acetate ion, determining higher values of extractable sulphate (AYLMORE; KARIM; QUIRK, 1967; BARROW; DEBNATH, 2015; ZHAO et al. al., 2017). For both extractants, the lower and upper limits of each range of variation in the S content corresponded to the same soils, 16-RQo and 19-LVAd.

Table 2
Available S content by two extractants, monocalcium phosphate (MCP) and ammonium acetate (AMA), in soil samples from the state of Ceará without the addition of S

Extractant	Soil1/
	1CXbe	2CXbe	3RRe	4PVAd	5RLd	6PVAd	7SXe	8RRe
	_________________________________________________ mg dm-3_____________________________________________
MCP	5.2	2.9	4.3	2.0	4.5	2.3	5.2	5.6
AMA	4.1	2.4	2.7	1.6	4.1	1.8	3.4	4.9
	9RL	10PVAe	11PVAe	12SXe	13PVAd	14PVAe	15RQo	16RQo
	_____________________________________________ mg dm-3_________________________________________________
MCP	3.2	2.9	1.7	5.9	7.1	6.0	2.3	1.4
AMA	2.5	2.7	1.2	4.1	5.9	5.9	1.9	1.1
	17PVAd	18PVAd	19LVAd	20LVAd	21RQo	22RL	23LVAd
	_____________________________________________ mg dm-3_________________________________________________
MCP	2.0	2.1	8.6	4.0	4.8	4.9	5.8
AMA	1.7	2.2	6.9	2.6	4.2	3.1	4.8
Amplitude: 7.2 (MCP) and 5.7 (AMA)
Mean: 4.1 (MCP) and 3.3 (AMA)

• ^1/
  1,2 - CXbe - Eutrophic Haplic Cambisol; 3, 8 - RRe - Eutrophic Regolithic Neosol; 4,6,13,17,18 - PVAd - Dystrophic Red-Yellow Ultisol; 5 - RLd - Eutrophic Litholic Neosol; 7,12- SXe - Haplic Planosol; 10,11,14 - PVAe - Eutrophic Red-Yellow Ultisol; 15,16,21 - RQo - Dystrophic Quartzarenic Neosol; 19,20,23 - LVAd - Dystrophic Red-Yellow Latosol; 9,22 - RL - Litholic Neosol.

Table 3 shows the linear correlation coefficients resulting from the relationships between the available S by the two extractants plus various properties of the soil. The behaviour of both extractants was similar within each property. The correlation coefficients for most soil properties were not significant, possibly due to the varied mineralogical composition of the soils and the narrow range of variation in the values of their properties.

The levels of clay and non-exchangeable H were the only properties that showed a positive correlation (p ≤ 0.01), revealing clay as a labile and non-labile reservoir of S, possibly forming covalent bonds with the non-exchangeable H. No correlation was found between organic matter and available S, possibly due to the predominance of inorganic forms. The pH also affects the way the plant responds to S fertilisation, so that a higher pH favours leaching of SO₄^2- to the subsurface layers (SUTAR et al., 2017). This did not occur in the soils under study due to the low pH (≤ 5.8) in 20 of the 23 soils.

Table 3
Correlation coefficient and linear regression equations for available sulphur (Y) by two extractants, monocalcium phosphate (MCP) and ammonium acetate (AMA), with various soil attributes

Property	Equation	R
% clay	$\hat{\mathrm{Y}}=2.17+0.09\mathrm{X}\left(\text{MCP}\right)$ $\hat{\mathrm{Y}}=1.36+0.11\mathrm{X}\left(\text{AMA}\right)$	0.51** 0.54**
% organic matter	$\hat{\mathrm{Y}}=3.26+0.07\mathrm{X}\left(\text{MCP}\right)$ $\hat{\mathrm{Y}}=2.30+0.16\mathrm{X}\left(\text{AMA}\right)$	0.24ns 0.38ns
pH	$\hat{\mathrm{Y}}=8.95-0.99\text{X}\left(\text{MCP}\right)$ $\hat{\mathrm{Y}}=7.89-0.85\text{X}\left(\text{AMA}\right)$	-0.27ns -0.31ns
Sum of bases	$\hat{\mathrm{Y}}=3.55+0.07\mathrm{X}\left(\text{MCP}\right)$ $\hat{\mathrm{Y}}=3.01+0.17\mathrm{X}\left(\text{AMA}\right)$	0.25ns 0.34ns
Non-exchangeable H	$\hat{\mathrm{Y}}=0.12+0.57\mathrm{X}\left(\text{MCP}\right)$ $\hat{\mathrm{Y}}=0.17+1.39\mathrm{X}\left(\text{AMA}\right)$	0.85** 0.89**
Al3+	$\hat{\mathrm{Y}}=0.39+0.04\mathrm{X}\left(\text{MCP}\right)$ $\hat{\mathrm{Y}}=2.57+1.45\mathrm{X}\left(\text{AMA}\right)$	0.30ns 0.30ns
V%	$\hat{\mathrm{Y}}=31.86+1.02\mathrm{X}\left(\text{MCP}\right)$ $\hat{\mathrm{Y}}=3.28+0.004\mathrm{X}\left(\text{AMA}\right)$	0.10ns 0.07ns

• ^ns
  - not significant;
• ^**
  - significant at level of 1%.

Dry matter production (DMW), relative production (RP) and the increase in production (IP) in the two successive cuts are shown in Table 4. Based on these data, the analysis of variance was carried out for the two cuts separately, which by F-test (p≤0.05) showed significant responses to the S doses, to the soils and to the dose x soil interaction. For this study, only a breakdown of the S doses within each soil was considered.

Table 4
Dry matter production (DMW), relative production (RP) and increase in production (IP) by plants of forage sorghum in treatments with S (So) and with no S (S₁)

Soil	DMW						RP		IP
	1st cut				2nd cut		1st cut	2nd cut	1st cut	2nd cut
	S0		S1		S0	S1
	__________ g per experimental unit______________						________________________ (%)_______________________
1 CXbe		10.73ª		11.27a	8.50b	17.17a	95	50	5	51
2 CXbe		9.70ª		10.80a	9.73b	18.57a	90	52	10	48
3 RRe		10.43ª		11.87a	6.40b	14.50a	88	44	12	56
4 PVAd		3.38b		9.22a	5.63b	11.57a	37	49	63	51
5 RLd		11.76ª		9.25b	16.56a	18.04a	127	92	--	8
6 PVAd		6.64b		8.11a	5.20b	16.24a	82	32	18	68
7 SXe		10.48b		11.19a	7.75b	15.65a	94	50	6	51
8 RRe		9.28ª		10.16a	13.60b	19.52a	91	70	9	30
9 RL		10.61ª		11.55a	12.10b	21.11a	92	57	8	43
10 PVAe		12.04ª		12.14a	12.98b	20.92a	99	62	1	38
11 PVAe		8.53b		12.00a	6.36b	22.05a	71	29	29	71
12 SXe		12.19ª		10.09b	16.81b	23.98a	121	70	--	30
13 PVAd		13.64ª		13.67a	16.70b	22.72a	100	74	--	27
14 PVAe		13.56ª		13.91a	12.38b	23.04a	97	54	3	46
15 RQo		7.09b		9.94a	4.41b	17.53a	71	25	29	75
16 RQo		7.57b		10.07a	4.18b	16.11a	75	26	25	74
17 PVAd		9.74ª		10.28a	5.03b	17.81a	95	28	5	72
18 PVAd		7.56b		11.19a	7.11b	20.92a	68	34	32	66
19 LVAd		7.50ª		7.02a	15.96a	18.68a	107	85	--	15
20 LVAd		10.61ª		8.73b	15.14b	21.92a	122	69	--	31
21 RQo		9.27ª		10.03a	11.91b	22.26a	92	54	8	47
22 RL		13.27ª		12.62a	14.54b	21.13a	105	69	--	32
23 LVAd		9.97ª		9.78a	12.08b	27.62a	102	44	--	56
Mean		9.80		10.64	10.48	19.50	-	-	-	-

• Mean values followed by the same letter in a row for each variable and cut did not differ by F-test (p≤0.05).
• ^1/
  1,2 - CXbe - Eutrophic Cambisol; 3, 8 - RRe - Eutrophic Regolithic Neosol; 4,6,13,17,18 - PVAd - Dystrophic Red-Yellow Ultisol; 5 - RLd - Eutrophic Litholic Neosol; 7,12- SXe - Haplic Planosol; 10,11,14 - PVAe - Eutrophic Red-Yellow Ultisol; 15,16,21 - RQo - Dystrophic Quartzarenic Neosol; 19,20,23 - LVAd - Dystrophic Red-Yellow Latosol; 9 - RL - Litholic Neosol;

In the first cut, for seven of the soils under study, the plants responded positively to the application of S, with 4PVAd standing out due to an increase in production of 63% (Table 4). In general, the soils provided enough S to meet the demands of the plant, which, as a Poacea, has a low S requirement. Singh et al. (2015) also found that the agronomic efficiency of rice grains per kg of applied S was reduced as the availability of the S increased. In addition to the low S demand of sorghum, the low response during the first cut can be explained by the increase in P availability (fertilisation) which, due to competition for the phosphate ion by the exchange sites, displaced SO₄^2- to the soil solution (BARROW; DEBNATH, 2015; POZZA et al., 2007).

With the second cut, except for two soils (5-RLd and 19-LVAd), plants not fertilised with S had a lower DMW (Table 4). This shows the vulnerability of the soil after being used with no S fertilisation, which leads to increased S deficiency in the crops, and limits productivity and product quality (JOHNSON et al., 2018; SCHERER, 2009; VERMEIREN et al., 2018). In the fertilised soils, the response was due to the greater availability of S in the labile phase, and the plant regrowth already having a developed root system, thereby reducing the input of photoassimilates and nutrients for root development.

Sulphur availability maximises the activity of the nitrogenase enzyme in N-deficient soils (DEVI et al., 2012), contributes to a proper N to S ratio in the plant and, consequently, increases protein synthesis, production and product quality (IBAÑEZ et al., 2020; JAMAL; MOON; ABDIN, 2010; STEINFURTH et al., 2012).

The increase in production (IP) in the first cut varied from almost zero to 63%, but in most soils the IP was less than 20% (Table 4), underlining the low response to the added S. There was a greater response in the second cut, with the IP ranging from 8% to 75%, and 15 soils showing more than a 40% increase in production.

In the first cut, the critical level for soil S was 2.40 mg dm^-3 for both MCP and AMA (Figures 1A and 1C). Below these values, the probability of nutrient deficiency is great, with a consequent response to fertilisation (CATE JUNIOR; NELSON, 1965). For this cut, it was found that 16 soils had an available S content greater than the critical level established for both extractants, explaining the low response of the plants in the first cut.

In the second cut (Figures 1B and 1D), keeping the same critical level of 2.40 mg dm^-3, the number of soils that showed an available S content greater than the critical level was reduced to 13 and 14, using MCP and AMA respectively. However, using both extractants, the RP went up to 50%, with only one soil having an RP greater than 90%.

Figure 1
Relationship between the relative production of dry matter weight in sorghum plants and the available S, for MCP in the first cut (A) and second cut (C), and for AMA in the first cut (B) and second cut (D)

The values obtained for critical level (CL) are below the range of variation in available S, of between 5 and 14 mg dm^-3 for species with a low and high S demand, respectively (CARMONA et al., 2009; NASCIMENTO; MORELLI, 1980; PIAS et al., 2019). However, Fontes et al. (1982) found a CL of 1.2 mg dm^-3, close to that found in this study, using sorghum and MCP in water.

The low CL found in this study is attributed to the low S requirement of the crop, since more demanding species require levels greater than 7.5 mg dm^-3, especially in soils with low levels of clay and organic matter (PIAS et al., 2019). Some studies of soils with an S content greater than 4 mg dm^-3 found no plant response to S, except for Brassica napus L. var. napus (canola) (RHEINHEIMER et al., 2007; TIECHER et al., 2012), which is included in the group with a high S requirement.

The critical level (CL) was obtained for the two successive cuts by relating the relative production with the S content of the plants with no fertilisation (Figures 2A and 2B). In the first cut, the CL for S in the leaf tissue was 0.6 g S kg^-1 for a relative production greater than 90%. In the second cut, except for one soil, all the plants had an RP of less than 90%, with levels below the CL in most soils. Due to data dispersion, it was not possible to establish the CL for S in plants fertilised with the nutrient.

Figure 2
Relative production as a function of plant S content and the critical level for S in sorghum plants, in the first cut (A) and the second cut (B)

The S concentration in plant tissue varies between 1 and 5 g kg^-1, with the concentration decreasing in the following order: Cruciferae, Leguminosae and Gramineae (LUCHETA; LAMBAIS, 2012). The low levels of S found in this study can therefore be explained by sorghum being a species with a low S requirement. Studies reviewed by Pias et al. (2019) showed that maize and wheat, from the same family as sorghum, had a critical level for leaf S in the range of 0.8 to 2.5 mg dm^-3, with a relative grain production greater than 65%. S deficiency affects the growth, development, resistance to disease, and performance of the plants, and has a great impact on the nutritional quality of the crops (KORPIVA; MALAGOLI; TAKAHASHI, 2019).

The successive cultivation of sorghum without the addition of S caused a reduction of S in the leaf tissue to below the CL in the plants of 17 soils (Figures 2A and 2B). The S content of plant tissue is an uncertain variable due to its low correlation with production (DIVITO et al., 2015; FRANDOLOSO et al., 2010; MODA et al., 2013). However, analysis of the leaf tissue can be an effective tool for monitoring the need for S fertilisation in soils with low SO₄² availability (PIAS et al., 2019), as in the soils under study.

The values for relative production as a function of the N/S and P/S ratios were plotted in Figures 3A and 3B, respectively. For N/S, the critical level (CL) was 22.5, values above that level indicate a S deficiency, limiting dry matter production in the sorghum. It was found that in the first cut, the plants in 10 soils had an N/S ratio lower than the CL and an RP greater than 90%; in six soils the plants had an N/S ratio greater than the CL and an RP equal to or greater than 90%. In the second cut, except for one soil, the RP of the plants was less than 90%, caused by the S expended by the first crop.

Figure 3
Critical level for the N/S (A) and P/S (B) ratios in two cuts of sorghum from the treatment with no sulphur (S₀)

Although the N/S ratio did not remain stable throughout the growth cycle of the plant, it can be used as a tool for diagnosing S during cultivation. As a result, for diagnosing S deficiency in wheat, Reussi Calvo et al. (2012) suggested determining the S concentration in the tissue and the N/S ratio. Ercoli et al. (2011) point out that this ratio should be used for soils with good S availability, as plants tend to absorb far more S than they require (luxury consumption).

Relative production as a function of the P/S ratio was plotted in Figure 3B, where the critical level (CL) obtained was 10. This relationship proved to be unreliable due to the high RP in plants with a CL greater than 10.

Comparing available S by the two extractants, a statistically significant, positive linear correlation was found at a level of 1% (r² = 0.92). In principle, this close correlation suggests that both extractants can be used to assess the available S.

When the available S was correlated with the absorbed S (S-Abs) and the increase dry matter production of the aerial part of the sorghum (IP) (Table 5), it was found that in both cuts the extractants showed a positive linear correlation with the S-Abs and a negative correlation with the IP. However, the MCP extractant proved to be superior, showing a greater correlation with both variables.

Table 5
Correlation coefficients between the soil S extracted with monocalcium phosphate (MCP) and ammonium acetate (AMA), and the S absorbed (S-abs) by the plants and the increase in production (IP), in both cuts

Variable	Correlation coefficient
	1st cut	2nd cut
S-abs x S - MCP	0.89**	0.74**
S-abs x S - AMA	0.63**	0.69**
IP x S - MCP	-0.64**	-0.71**
IP x S - AMA	-0.60**	-0.54**

• ^*
  Significant at a level of 5%;^**Significant at a level of 1%.

Thus, despite the correlation showing the possibility of using both extractants for assessing the available S, as also seen by Blum et al. (2014) in maize and wheat, MCP should be preferred, as it shows a higher correlation with the increase in production and with the S absorbed by the plants. The availability of S in the soil and the requirements of the plant regulate the processes of SO₄^2- absorption and assimilation (TAKAHASHI et al., 2011). By properly estimating the available soil S, deficiencies can be corrected, favouring the metabolic processes of the crops and increasing productivity.

CONCLUSIONS

1. The critical levels of S in forage sorghum for the soils under study were 3.90 and 3.40 mg dm^-3 for the MCP and AMA extractants, respectively;

2. The available S by monocalcium phosphate (MCP) showed a better correlation with the increase in production and the S absorbed by the plants for the soils under study;

3. MCP is the recommended extractant for assessing available S in the soils of Ceará.

REFERENCES

• AYLMORE, L. A. G.; KARIM, M.; QUIRK, J. P. Adsorption and desorption of sulphate ions by soil constituents. Soil Science, v. 1, p. 10-15, 1967.
• BARDSLEY, C. E.; LANCASTER, J. D. Determinations of reserve sulfur and soluble sulfates in soils. Soil Science Society America Proceedings, v. 24, p. 265-268, 1960.
• BARROW, N. J.; DEBNATH, A. Effect of phosphate and pH on sulphate sorption and desorption. European Journal of Soil Science, v. 66, p. 286-297, 2015.
• BLUM, S. C. et al. Assessing available soil sulphur from phosphogypsum applications in a no-till cropping system. Experimental Agriculture, v. 50, n. 4, p. 516-532, 2014.
• CARMONA, F. C. et al. Disponibilidade no solo, estado nutricional e recomendação de enxofre para o arroz irrigado. Revista Brasileira de Ciência do Solo, v. 33, n. 2, p. 345-355, 2009.
• CATE JUNIOR, R. B.; NELSON, L. A. A rapid method for correlation of soil test analyses with plant response data. Raleigh, NC: North Carolina State University Agricultural Experiment Station, 1965. (International Soil Testing. Technical Bulletin, 1).
• DEVI, K. N. et al. Influence of sulphur and boron fertilization on yield, quality, nutrient uptake and economics of soybean (Glycine max) under upland conditions. Journal of Agricultural Science, v. 4, p. 1-10, 2012.
• DIVITO, G. A. et al. Diagnosis of S deficiency in soybean crops: performance of S and N:S determinations in leaf, shoot and seed. Field Crop Research, v. 180, p. 167-75, 2015.
• ERCOLI, L. et al Durum wheat grain yield and quality as affected by S rate under Mediterranean conditions. European Journal of Agronomy, v. 35, p. 63-70, 2011.
• FERNANDES, M. B.; FREIRE, F. J.; OLIVEIRA, A. C. Níveis críticos de enxofre em solos de Pernambuco. Revista Caatinga, v. 20, n. 3, p. 93-103, 2007.
• FONTES, M. P. F. et al. Nível Crítico de enxofre em latossolos e recuperação do sulfato adicionado por diferentes extratores químicos, em casa de vegetação. Revista Brasileira de Ciência do Solo, v. 6, p. 226-30, 1982.
• FOX, R. L.; OLSON, R. A.; RHOADS, H. F. Factors influencing the availability of sulfur fertilizers to alfafa and corn. Soil Science Society Proceedings, v. 28, p. 406-408, 1964.
• FRANDOLOSO, J. F. et al Eficiência de adubos fosfatados associados a enxofre elementar na cultura do milho. Revista Ceres, v. 57, n. 5, p. 686-694, 2010.
• GALVÃO, J. R. et al. Adubação potássica em híbridos de sorgo forrageiro cultivados em sistemas de manejo do solo na Amazônia oriental. Revista Caatinga, v. 28, n. 4, p. 70-79, 2015.
• IBAÑEZ, T. B. et al. Sulfur modulates yield and storage proteins in soybean grains. Scientia Agricola, v. 78, n. 1, e20190020, 2020.
• JAMAL, A.; MOON, Y. S.; ABDIN, M. Z. Enzyme activity assessment of peanut (Arachis hypogea) under slow-release sulphur fertilization. Australian Journal of Crop Science, v. 4, n. 3, p. 169-174, 2010.
• JOHNSON, J. et al. The response of soil solution chemistry in European forests to decreasing acid deposition. Global Change Biology, v. 24, p. 3603-3619, 2018.
• KOPRIVA, S.; MALAGOLI, M.; TAKAHASHI, H. Sulfur nutrition: impacts on plant development, metabolism, and stress responses. Journal of Experimental Botany, v. 70, n. 16, p. 4069-4073, 2019.
• LUCHETA, A. R.; LAMBAIS, M. R. Sulfur in agriculture. Revista Brasileira de Ciência do Solo, v. 36, p. 1369-1379, 2012.
• MODA, L. R. et al Gessagem na cultura da soja no sistema de plantio direto com e sem adubação potássica. Revista Agroambiente On-line, v. 2, n. 7, p. 129-135, 2013.
• NASCIMENTO, J. A. L.; MORELLI, M. Enxofre em solos do Rio Grande do Sul. I. Formas nos solos. Revista Brasileira de Ciência do Solo, v. 4, n. 3, p. 131-135, 1980.
• NOVAIS, S. V. et al. Loss of extraction capacity of mehlich-1 and monocalcium phosphate as a variable of remaining p and its relationship to critical levels of soil phosphorus and sulfur. Revista Brasileira de Ciência do Solo, v. 39, p. 1079-1087, 2015.
• PIAS, O. H. C. et al. Crop yield responses to sulfur fertilization in Brazilian no-till soils: a systematic review. Revista Brasileira de Ciência do Solo, v. 43, p. 1-21, 2019.
• POZZA, A. A. A. et al. Retenção e dessorção competitivas de ânions inorgânicos em gibbsita natural de solo. Pesquisa Agropecuária Brasileira, v. 42, n. 11, p. 1627-1633, nov. 2007.
• RAMPIM, L. et al. Atributos químicos de solo e resposta do trigo e da soja ao gesso em sistema semeadura direta. Revista Brasileira de Ciência do Solo, v. 35, n. 5, p. 1687-1698, 2011.
• REUSSI CALVO, N. I. et al. Stability of foliar nitrogen: sulfur ratio in spring red wheat and sulfur dilution curve. Journal of Plant Nutrition, v. 35, p. 990-1003, 2012.
• RHEINHEIMER, D. S. et al. Resposta à aplicação e recuperação de enxofre em cultivos de casa de vegetação em solos com diferentes teores de argila e matéria orgânica. Ciência Rural, v. 37, n. 2, p. 363-371, 2007.
• SCHERER, H. W. Sulfur in soils. Journal of Plant Nutrition and Soil Science, v. 172, p. 326-335, 2009.
• SINGH, V. K. et al Status of available sulfur in soils of north-western indo-gangetic plain and western himalayan region and responses of rice and wheat to applied sulfur in farmer’s Fields. Agriculte Research, v. 4, n. 1, p. 76-92, 2015.
• SOLOMON, D. et al. Speciation and long - and short-term molecular-level dynamics of soil organic sulfur studied by x-ray absorption near-edge structure spectroscopy. Journal of Environmental Quality, v. 40, p. 704-718, 2011.
• STEINFURTH, D. et al. Time-dependent distribution of sulphur, sulphate and glutathione in wheat tissues and grain as affected by three sulphur fertilization levels and late S fertilization. Journal of Plant Physiology, v. 169, p. 72-77, 2012.
• SUTAR, R. et al. Sulphur nutrition in maize - a critical review. International Journal of Pure & Applied Bioscience, v. 5, n. 6, p. 1582-96, 2017.
• TAKAHASHI, H. et al. Sulfur assimilation in photosynthetic organisms: molecular functions and regulations of transporters and assimilatory enzymes. Annual Review of Plant Biology, v. 62, p.157-84, 2011.
• TIECHER, T. et al. Resposta de culturas e disponibilidade de enxofre em solos com diferentes teores de argila e matéria orgânica submetidos à adubação sulfatada. Bragantia, v. 71, n. 4, p. 518-527, 2012.
• VERMEIREN, C. et al. Model-based rationalization of sulphur mineralization in soils using 35 S isotope dilution. Soil Biology and Biochemistry, v. 120, p.1-11, 2018.
• VIEIRA-FILHO, M. S.; LEHMANN, C.; FORNARO, A. Influence of local sources and topography on air quality and rainwater composition in Cubatão and São Paulo, Brazil. Atmospheric Environment, v. 101, p. 200-208, 2015.
• VON PINHO, R. G. et al Produtividade e qualidade da silagem de milho e sorgo em função da época de semeadura. Bragantia, v. 66, n. 2, p. 235-245, 2007.
• ZHAO, B. et al. Sulfate sorption on rape (Brassica campestris L.) straw biochar, loess soil and a biochar-soil mixture. Journal of Environmental Management, v. 201, p. 309-314, 2017.

Edited by

Editor-in-Article: Profa. Mirian Cristina Gomes Costa - [e-mail]

Publication Dates

• Publication in this collection
  12 Nov 2021
• Date of issue
  2021

History

• Received
  23 July 2020
• Accepted
  15 Mar 2021