Potential hydrogen ion of Quartzarenic Neosol with joint application of lime and gypsum

The objective of the study is to investigate the interaction of lime and gypsum during incubation of soil samples in response to pH and also check the possibility of using more than one treatment for a single model by model identity method. The analysis of the potential hydrogen ion (pH) was performed in the Laboratory of Agricultural Chemistry, Federal Institute Goiano, Campus Rio Verde GO, Brazil. The soil used is classified as Quartzarenic Neosol loam sandy and its corresponding soil taxonomy is Quartzipsamment. The experiment was completely randomized with four replications in a factorial 5 × 3 design, five doses of lime (CaCO3) DL (0, 1, 2, 3 and 4 t ha -1 ) and three doses of gypsum (CaSO4) DG ( 0, 0.7 and 5 t ha -1 ). The sources of CaCO3 and CaSO4, dolomitic lime and gypsum were used, respectively. Incubation was done in plastic containers with 20 ml of soil, and pH analysis was done every two days for 10 days. In general, it was frequently observed that reducing the pH of the reaction in the soil was attributed to higher DG applied together with CaCO3. Due to the difference in all models tested by identity models, it is explained that the range of CaSO4 doses and incubation periods promoted major change in the model either by interception, angulation or both. The reduction of pH in the soil is attributed to higher rates of gypsum used together with lime.


INTRODUCTION
One of the factors that limit crop productivity is the acidity of the soil, once liming is used for its correction (Fageria, 2001).Several authors have identified the importance of using lime together with gypsum to treat soil acidity profile (Caires et al., 2008;Maschietto, 2009;Dalla Nora et al., 2013;Zandoná et al., 2015).
Alone, liming raises the potential hydrogenion (pH) of the soil, neutralizes Al 3+ toxicity, adds Ca 2+ and Mg 2+ , provides favorable conditions for root growth and water and nutrient uptake by plants (Zandoná et al., 2015).
Gypsum is used in tillage system to minimize acidity problems.Low Al 3+ toxicity is precipitated when it reacts with gypsum (Zambrosi et al., 2007).This causes it to be less toxic (AlSO 4+ ) and increases the Ca 2+ concentration and S underground (Neis et al., 2010).This can make the lime to be active on the surface and subsoil soil (Caires et al., 2003).The main criticism is it results in the imbalance between the soil bases, considerably raising the exchangeable calcium content (Araujo et al., 2015).
When there is an interest to study quantitative variables, a variable response (pH) using one or more explanatory variables such as combined application of gypsum and lime, the regression analysis is a data analysis technique widely used.Due to different explanatory variables, regression analysis may be applied separately to obtain different results according to the number of situations.
Typically, the response variable Y and the set of factors, called the regression variables X i , i = 1, 2, ..., n are measured in separate groups, to compare the resulting models if they differ (Seber, 1977).
Knowing the importance of identity models, it is important to apply this method on the response of soils to potential hydrogenionic based on the explanatory variables (gypsum doses and periods of incubation under lime function).This is because acidity limits the productivity of crops in sandy soils, beyond the limited studies on the reaction of gypsum and lime.
The hypothesis of this work is that gypsum and lime doses applied together has effect on soil pH at different periods of incubation, and further on, identity model was used for gypsum doses alone and for both lime and gypsum doses.

MATERIALS AND METHODS
The analysis of the potential hydrogenionic (pH) was performed in the Laboratory of Agricultural Chemistry from the Federal Institute Goiano, Campus Rio Verde -GO, Brazil.The soil used is classified as a Quartzarenic Neosol loam sandy and its corresponding one in soil taxonomy is Quartzipsamment (Embrapa, 2013).
The experiment was completely randomized with four replications in a factorial 5 × 3 design, five doses of lime (CaCO 3 ) -DL (0, 1, 2, 3 and 4 t ha -1 ) and three doses of gypsum (CaSO 4 ) -DG ( 0, 0.7 and 5 t ha -1 ).The sources of CaCO 3 and CaSO 4 were used as dolomitic lime and gypsum, respectively.Incubation was done in plastic containers with 20 ml of soil and pH analysis was done every two days for 10 days, totaling five periods of incubation periods.
During the incubation period of the soil, moisture was maintained at 60%, by measuring the mass of the sample units.At 2, 4, 6, 8 and 10 days after the beginning of the incubation period, the samples were placed in a forced-air oven at 45°C for 24 h.Subsequently, the soil subsamples were withdrawn for determination of pH potentiometrically in soil suspension solution of ratio 1:2.5 in distilled water, KCl 1 mol L -1 and CaCl 2 0.01 mol L -1 .The suspension was read, with contact time (Embrapa, 2011).
The variance of interactions was analyzed (p<0.05) and identity of models (p<0.10) by the F test, set models linear as a function of the DL and represented the behavior of pH in surface graphs.After characterizing with F test (p<0.10), the combination of the models between the DG for each incubation periods (IP) and between the IP for each DG, it was considered that the hypothesis of similarity of the parameters a's and b's for the two sets of observations is rejected when Fc > Ft.

RESULTS AND DISCUSSION
The application of CaSO 4 and CaCO 3 together led to significant changes (p <0.05) in potential hydrogenionic (pH).At a dose of CaCO 3 (DL) 4 t ha -1 , the absence of CaSO 4 led to higher pH (H 2 O) in relation to DG 5 t ha -1 and the incubation periods (IP) of 4, 6 and 10 days.The differences were 0.68, 1.89 and 1.87, respectively (Figure 1B, C and D).
According to Araujo et al. (2015), elementary sulfur and calcium sulfate positively affected the chemical properties of the soil, improving its fertility.Moreover, the elementary sulfur is more efficient than calcium sulphate in decreasing alkalinity.However, the use of this product requires the application of additional calcium in the soil.
From DL 2 t ha -1 regardless of DG, pH proved to be appropriate for the soil in 2 and 4 days IP (Figure 1A and  B), but at 6, 8 and 10 days of incubation, the pH was maintained from appropriate DL 2 t ha -1 without application of CaSO 4 (Figure 1C, D and E).Frade Junior (2013) verified that the average values of pH in water, which increased with the increasing doses of CaCO 3 , raised the pH to approximately neutral ground for the three soils (Spodosol Humilúvico, Yellow Latosol and Ultisol Alítico Plinthic).
The pH given in water showed a linear polynomial fit in each DG depending on the DL in all IPs (Figure 1).The greatest increases in pH of DL function were checked at DG 5 t ha -1 IP in 2 and 4 days; there was an increase of 0.55 and 0.61, respectively in response to an increase of 1 t ha -1 CaCO 3 (Figure 1A and B).Costa (2015) observed that the surface application of limestone, with or without gypsum, were effective in reducing soil acidity to a depth of 0.20 m The greatest increases in pH at IP of 6, 8 and 10 days were observed in the absence of CaSO 4 ; there was an increase of 0.71, 0.48, and 0.68, respectively in response to an increase of 1 t ha -1 of CaCO 3 (Figure 1C, D and E).According to Natale et al. (2007), with increased doses of correctives, there was a linear increase of pH effect on dystrophic Red Latosol.
In 2 and 4 days IP (Figure 2A and B), the pH showed higher values in higher DL and DG based on the methodology for determining KCl; unlike IP 6, 8 and 10 days (Figure 2C, D and E) that showed higher pH values in greater DL and absence of CaSO 4 .Chi et al. (2012) found that gypsum was effective in improving the physical and chemical properties of the soil, increasing the aggregate stability.5   6,0   6,0   6,0  6,0   6,0   5,5  5,5   5,5   5,5   5,5   5,0   5,0   5,0   5,0   4,5   4,5   4,5   4,5   4,0   4  In DL 4 t ha -1 , DG 0.7 t ha -1 gave higher pH values (KCl); at DG 5 t ha -1 and IP in 4, 6 and 10 days, the differences were 1.01, 1.47, and 1.05, respectively (Figure 2B, C and  E).Although at DL 4 t ha -1 , the absence of CaSO 4 gave pH (KCl) of 1.17, more than that of DG 5 t ha -1 with a difference of 1.17 at 10 days IP (Figure 2E).According to Ebeling et al. (2008), the lowest pH values in KCl solution may result from KCl solution, which when in contact with the ground, induces the exchange of cations, releasing H + and Al 3+ ions, due to the higher concentration of K + ions.

6,
The pH determined by KCl presented polynomial fit of the second order DG 0.7 t ha -1 under DL at IP of 2 days and the other set of linear regression (Figure 2).The highest pH of function DL was checked at DG 5 t ha -1 and 2 days IP, with an increase of 0.67 when 1 t ha -1 of CaCO 3 was raised (Figure 2A).The surface application of gypsum reduced the exchangeable acidity (Al 3+ ) and increased the levels of Ca 2+ and S-SO 4 2-in surface and subsurface, but reduced the surface levels of Mg 2+ (Costa, 2015).
The pH determined by KCl presented polynomial fit of the second order DG 0.7 t ha -1 on DL function of the IP 2 days and the other set of linear regression (Figure 2).The greatest increases in pH of DL function were checked at DG 5 t ha -1 and 2 days IP, with an increase of 0.67 when 1 t ha -1 of CaCO 3 (Figure 2A) was raised.Ferreira et al. (2013) verified that soil Ca content showed a linear relationship with soil C under gypsum+lime (p<0.0001) and gypsum applications (p<0.0001).
In IP 4, 6, 8 and 10 days, the largest estimated values of the pH being 6.39, 6.82, 6.10, and 6.12, respectively were found at DL 4 t ha -1 and without CaSO 4 (Figure 2).Several authors found high yields of annual crops due to the increase in pH in the soil, as Fageria et al. (2005) found that increased productivity of beans in 6.4 pH water content in distrophic Red Latosol (Oxisol) which initially had a pH of 5.7.Fageria and Baligar (2003) found that higher yields with respect to the appropriate pH are through adequate supply of Ca and Mg and balance between basic cations.
In DL 2 t ha -1 , DG 0.7 t ha -1 gave higher pH values (CaCl 2 ) with respect to DG 5 t ha -1 and IP of 2, 4 and 8 days, the differences were 0.92, 1.43 and 2.06, respectively (Figure 3A, B and D).In DL 3 t ha -1 and IP of 2 days, the absence of CaSO 4 resulted in higher pH values being 1.54 and 0.98 higher than the DG 0.7 of 5 t ha -1 , respectively (Figure 3A); in the same DL, DG 0.7 t ha -1 gave higher pH values than the DG of 5 t ha -1 in IP of 6 and 10 days; an increase of 1.06 and 1.01, respectively (Figure 3C and E).Rosa et al. (2012), evaluating lime and gypsum in distroferric Red Latosol, found that the application of increasing doses of lime caused an increase in pH and the use of gypsum influenced positively the culture only at 871.3 kg ha -1 and lime of above 2000 kg ha -1 .
The pH given in CaCl 2 showed second order polynomial fit of the DG 0.7 to 5 t ha -1 on DL function of the IP 8 days and the other set of linear regression (Figure 3).The greatest increases in pH at IP of 2, 4 and 6 were recorded in the absence of CaSO 4 0.63, 0.58, and 0.71, respectively, an increase of 1 t ha -1 of CaCO 3 (Figure 3A, B and C).The greatest increases in pH of DL function were checked at DG 0.7 t ha -1 and IP of 8 to 10 days; there was an increase of 0.56 and 0.67, respectively; increased by 1 t ha -1 CaCO 3 (Figure 3D and  E).
Studying the effect of adding lime and gypsum on the base in soil columns, Dal Bó et al. (1986) found that treatment with gypsum isolated accelerated magnesium and calcium displacement in depth, so it is likely that this change in calcium and the changes introduced by incubation periods justify the rejection of the identity of tested models.
In IP of 2, 4 and 6 days, the largest estimated values of the pH being 6.26, 5.74, and 6.43, respectively, were found at DL 4 t ha -1 and without CaSO 4 ; but 6.27 and 6.30 values after 8 and 10 days of incubation of the samples were superior to the others (Figure 3).
The results obtained by the interaction of gypsum and lime doses applied jointly in the soil samples are related to the results of Carvalho and Nascente (2014)'s study on the influence of lime and gypsum in the production of biomass and cycling millet nutrients.It was found that there were increases of dry biomass production with the application of lime and gypsum provided there were no changes in the production of dry biomass of millet.
In general, it was frequently observed that reducing the pH of the reaction fixed in the soil was attributed to higher DG applied together with CaCO 3 .
According to the analysis of variance F (p> 0.1), it was found that the adjusted models of 0, 0.7, and 5.0 t ha -1 were similar between the IP and dose (Table 1) of both models set under DL (Table 2).Based on Regazzi and Silva (2004), the approximated F should be preferred, since it provides a rate of type I error, regardless of the sample size.
Due to the differences in all models tested by identity models, it is explained that the range of CaSO 4 doses and incubation periods promoted major change in the model either by interception, angulation or both.Ribeiro et al. (2005), comparing all adjusted equations, found that the various ions present statistical difference between the adjusted models, except for sodium, and significant among some equations adjusted per time, with the largest differences observed for chloride and calcium with 46.15 and 40.66%, respectively.

Conclusion
The reduction of pH correction in the soil is given to higher rates of gypsum used together with lime.Some parameters of the hypothesis models in some combination of gypsum doses within a period of Figure 1.pH behavior water determined according to the relation of the doses of CaCO 3 and CaSO 4 in the incubation period two day (A), four days (B), six days (C), eight days (D) and ten days (E).

Figure 2 .
Figure 2. pH behavior KCl determined according to the relation of the doses of CaCO 3 and CaSO 4 in the incubation period two day (A), four days (B), six days (C), eight days (D) and ten days (E).

Figure 3 .
Figure 3. pH behavior CaCl 2 determined according to the relation of the doses of CaCO 3 and CaSO 4 in the incubation period two day (A), four days (B), six days (C), eight days (D) and ten days (E).

Table 1 .
F calculated in the model identity to the combinations of the incubation periods as a function of lime and gypsum in each dose for the different method of determining pH.

Table 2 .
F calculated models of identity for the combinations of doses of CaSO 4 due to the CaCO 3 doses of each incubation period and method for measuring the pH.