Macronutrients in watermelon plants fertilized with potassium and cattle manure

Macronutrientes em plantas de melancia adubadas com potássio e esterco bovino R E S U M O O potássio é o nutriente mineral exigido em maior quantidade pela melancieira principalmente na fase de produção. Este trabalho objetivou avaliar o efeito de doses de potássio e esterco bovino na composição foliar de melancieira. Os tratamentos foram distribuídos em blocos ao acaso com três repetições e 21 plantas por parcela em arranjo fatorial 3 × 5 + 1, referentes às três doses de esterco bovino 0; 1.100 e 2.520 g cova-1), para elevar o teor de matéria orgânica de 0,67% que o solo possuía, para 3 e 6%, cinco doses de K2O (0; 7,5; 15; 22,5 e 30 kg ha -1) utilizando-se KCl como fonte e um tratamento adicional (sem adubação orgânica e mineral). No início da frutificação foram colhidas folhas para avaliação do estado nutricional das plantas em macronutrientes (N, P, K, Ca, Mg e S). O esterco bovino estimula a absorção de macronutrientes pela melancieira; no solo com 3% de matéria orgânica os melhores resultados corresponderam à dose de K2O de 15 kg ha -1 e para 6% entre 11 e 13 kg ha-1.


Introduction
The mineral nutrition of watermelon [Citrullus lanatus (Thunb.)Matsum.& Nakai] is one of the factors that mostly contribute the most to the increase in yield and production quality of this crop, which stands out for exporting large amounts of nutrients, and potassium is the nutrient most extracted from 45 to 65 days after transplanting (Gonsalves et al., 2011;Almeida et al., 2012).
The adequate supply of K to watermelon increments the contents of the element in leaf dry matter and stimulates the absorption of other essential nutrients (El-Bassiony et al., 2012).For this, Cavalcanti (2008) recommends doses between 30 and 60 kg ha -1 of K 2 O, in the state of Pernambuco.In other northeastern states, such as Paraíba, the manuals of recommendation are scarce or outdated and become an obstacle for the expansion of local crops, warning for the necessity of studies with fertilization in local crops.
Besides mineral fertilization, the supply of organic matter to the soil is of great importance for the cultivation of watermelon and other vegetables, especially in sandy soils of arid regions that have low contents, because the climate conditions favor the acceleration of oxidation of soil organic matter (Nicolae et al., 2014).
Most producers of vegetables in the state of Paraíba, according to Silva et al. (2011), use cattle manure as the only source of nutrient for the crops, which means that, besides mineral fertilization, it is necessary to also invest in studies with doses of organic matter, because the manures are in many cases the only sources available to family farmers.
This study aimed to evaluate the leaf composition of macronutrients in watermelon plants cultivated in soil with doses of potassium and cattle manure.

Material and Methods
The experiment was carried out at the settlement 'Campo Comprido' , in Patos, PB, Brazil, from March to June 2012.The climate of the region, according to Köppen's classification, is BSh, semiarid, with mean annual temperatures above 25 ºC and mean annual rainfall below 1000 mm, with irregular rains during the rainy period.In 2012, the annual rainfall was 221 mm in the experimental area.The soil in the experimental area was classified as Fluvic Neosol.
Before the experiment, soil samples were collected in the layer of 0-20 cm for chemical and physical characterization (Table 1) according to the methodologies suggested by EMBRAPA (2011).
The experimental design was randomized blocks with three replicates in a 3 × 5 + 1 factorial arrangement, referring to the addition of three doses of cattle manure (0.0; 1,100 and 2,520 g hole -1 ) to increase the organic matter content of the soil, from 0.67% to 3.0 and 6.0%, five doses of K 2 O in the form of potassium chloride (0.0; 7.5; 15.0; 22.5 and 30.0 kg ha -1 ) as suggested by Cavalcanti (2008), and an additional treatment without fertilization, as commonly performed in the region.The experimental unit consisted of 21 plants per plot, distributed in three rows of seven plants.
The cattle manure doses were applied as basal fertilization during the opening and preparation of the holes, 30 days before sowing.The holes were open with dimensions of 0.20 x 0.20 x 0.30 m, volume of 12 dm 3 and spacing of 2.0 × 2.0 m.The variety 'Crimson Sweet' was selected for cultivation because of its acceptance in the regional market.
The cattle manure doses were obtained using the following expression: Table 1.Soil chemical and physical characterization in the layer of 0-20 cm
The doses of K 2 O in the form of potassium chloride were fractionated as 0, 25, 50, 75 and 100% of the maximum dose (30 kg ha -1 ) recommended by Cavalcanti (2008) for watermelon in soil with K contents above 0.30 cmol c ; thus, 0.0, 7.5, 15, 22.5 and 30 kg ha -1 of K 2 O were provided at 40 days after seedlings emergence.
All treatments, except the additional, received nitrogen fertilization (120 kg ha -1 ) in the form of urea, divided into two applications, 30 and 90 kg ha -1 of N, at 20 and 40 days after seedlings emergence (Cavalcanti, 2008).No phosphate fertilization was applied, due to the high values observed in the soil and in the cattle manure.
At sowing, three seeds were planted per hole and, after emergence, when plants showed three pairs of permanent leaves, thinning was performed, leaving only the most vigorous plant per hole.
Irrigation from sowing to emergence was daily performed through sprinklers, maintaining the soil with moisture (1) content adequate for the germination process.From emergence to fruiting, irrigation was performed using the conventional sprinkler method, supplying every 48 h a water depth equal to the reference evaporation obtained by the product of Class A pan evaporation and the factor 0.75, as suggested by Andrade et al. (2013).
At the beginning of fruiting, 25 leaves (5 th healthy leaf counted from the apex of the branch) were collected per plot, disregarding the apical tuft, as recommended by Trani & Raij (1997).After the samples were washed in tap water and submersed in deionized water, the material was placed in paper bags, dried in a forced-air oven at temperature of 65 ºC for 72 h and ground in a Wiley-type mill (TE-650 ® ) using a 20-mesh sieve.Leaf dry matter was analyzed for: total N, determined through the micro-Kjeldahl method (sulfuric digestion); P, through colorimetry of metavanadate; K, through flame emission photometry; Ca and Mg, through atomic absorption spectrophotometry; and S, through gravimetry of barium sulfate (Malavolta et al., 1997).
The data were subjected to analysis of variance by F test; the means referring to cattle manure doses were compared by Tukey test at 0.05 probability level and those referring to K doses by regression.The analysis of contrasts between fertilized treatments and the additional treatment was performed through the Dunnett's test at 0.05 probability level using the statistical program SAS.

Results and Discussion
The interaction between doses of K and cattle manure did not have significant effects on leaf N content; however, the mean contents observed in Table 2 are within the interval of 25-50 g kg -1 , considered by Trani & Raij (1997) as adequate for watermelon.Even considering that manure is a source of N to plants, the absence of significance between the doses of this input for the N contents of the plants can be explained by the supply of 120 kg ha -1 of N in the form of urea in total area, which resulted in equalization of the contents in the plants.
In the treatment with supply of 1,100 g of cattle manure, the increase in K 2 O doses promoted linear increment of 0.0263 g kg -1 in leaf P content, from 4.8 to 5.6 g kg -1 of leaf dry matter, which represents an increment of 16.5% between plants in soil without and with the highest dose of K 2 O (Figure 1).
This increment is a response of the stimulating action of K in the release of P from the soil, which was also reported by El-Bassiony et al. (2012), who observed that K doses in the soil stimulate P absorption in watermelons and pumpkins.
In the treatments without and with addition of 2,520 g of cattle manure, leaf P contents were increased to 5.55 and 5.82 g kg -1 of P for the maximum estimated doses of 19.4 and 18.6 kg ha -1 of K 2 O (Figure 1).According to Trani & Raij (1997), leaf P contents varying between 3.0 and 7.0 g kg -1 are adequate for the watermelon crop.It is observed, based on the amplitude, that the plants cultivated in the soil without the addition of cattle manure (dose of 0.0) and K 2 O were adequate supplied in P.
The increment in leaf P accumulation was more pronounced in the treatments with the highest dose of cattle manure, 2,520 g hole -1 , probably due to the higher supply of P with the highest dose of manure and the action of the organic acids produced during the decomposition of organic matter, which blocks P absorption sites in the soil, decreasing the fixation by the soil (Guppy et al., 2005).
This information is also consistent with Abul-Soud et al. ( 2010), in the cucumber crop (Cucumis anguria L.), Bindiya et al. (2012) and Santos et al. (2012), who reported increments of P in the leaf tissue of pumpkin (Cucurbita pepo L.) due to doses of organic inputs applied to the soil.
The adequate supply to plants can be a response of the high content of the nutrient in the soil (51 mg dm -3 ) as also observed by Geleta et al. (2011) in watermelon plants cultivated in soils with high P contents.
Leaf K contents did not differ as a function of the doses of cattle manure and K 2 O added to the soil, but the mean contents are within the range of 20 to 60 g kg -1 (Table 3) and, according to Trani & Raij (1997), plants were balanced with respect to the macronutrient.

K2O dose (kg ha -1 )
Cattle manure (g hole -1 ) 0 1,100   The absence of significant difference of K contents in leaf dry matter, associated with the adequate nutrition of the plants, means that the contents existing in the soil met the requirement of the crop.Despite the absence of significant differences, the previous supply of this element cannot be neglected, since the soil in the area is sandy and, under irrigation regime, the suppression of K could cause deficiency of this nutrient due to losses through leaching (Nicolae et al., 2014).
Even considering the plants in the treatments without K 2 O and cattle manure had contents sufficient for their demand, the watermelon crop, for being a plant that demands high amounts of K (Almeida et al., 2012), require the maintenance of sufficient levels of the element in the soil during its cycle, which justifies the recommendation of K by Cavalcanti (2008) even in soil with high contents of the nutrient.
In plants of the treatments without cattle manure, leaf Ca contents linearly increased from 15.53 to 22.1 g kg -1 , an increment of 42.31% between plants without and with the highest dose of K 2 O (Figure 2).This result differs from those in the literature, in which the addition of K to the soil inhibits the availability of Ca to plants (Silva & Trevizam, 2015).
In the treatments with 1,100 g hole -1 , represented by the equation ŷ = 16.44 + 0.8281x -0.0277**x 2 (R² = 0.52) and 2,520 g hole -1 of cattle manure (Figure 2), the increment in K doses increased the values of Ca in plant leaves to contents of 22.5 and 22.8 g kg -1 , respectively, at the maximum estimated K 2 O doses of 14.9 and 13.1 kg ha -1 .
The behavior of the data is similar to those reported by Nascimento et al. (2015) for watermelon and Santos et al. (2012) for pumpkin, a plant of the same family of watermelon, who reported increments in leaf Ca contents due to the increase in doses of organic materials in the soil.
The decrease in Ca in the treatments with K 2 O doses above the maximum estimated doses can be due to the competitive inhibition between K and Ca in the exchange sites of the soil (Silva & Trevizam, 2015).Still in agreement with the previously mentioned authors, when there are high levels of K in the soil, the plant grows more, causing reduction in the contents of Ca due to the dilution effect.
The leaf Ca contents of 22.1, 22.5 and 22.8 g kg -1 , respectively, in plants without and with the addition of cattle manure, to increase the content of soil organic matter from 0.67% to 3.0 and 6.0%, are below the lower critical limit that meets the requirement of watermelon, which oscillates between 25 and 50 g kg -1 (Trani & Raij, 1997).
In the treatments without manure (0 g hole -1 ) and with the dose of 1,100 g hole -1 , leaf Mg contents did not fit to any mathematical model as a function of the K 2 O doses, with mean contents of 5.26 and 5.97 g kg -1 , respectively (Figure 3).
In the treatment with 2,520 g hole -1 of cattle manure, the increment in K 2 O doses caused reduction in Mg accumulation in the leaf dry matter of the plants.In spite of that, the mean content of Mg accumulated in the leaves (6.62 g kg -1 ) was numerically higher than those observed in the treatments with 0.0 and 1,100 g hole -1 of cattle manure.As observed for the Ca content, the absorption of Mg was also inhibited by the increment in K doses.
Similar behavior was reported by Santos et al. (2012), studying doses of organic inputs applied to the soil on the leaf Mg contents in pumpkin.Based on the results in all situations, the contents of Mg are within the range of 5-12 g kg -1 established as adequate for the watermelon crop (Trani & Raij, 1997) Due to the increase in the doses of cattle manure, the contents of S fitted to a quadratic model of polynomial regression (Figure 4).The supply of the doses of 0.0 g hole -1 (ŷ = 2.41708 + 0.02003x -0.00053 NS x 2 R² = 0.58), 1,100 g hole -1 (ŷ = 2.49128 + 0.02932x -0.00087*x 2 R² = 0.53) and 2,520 g hole -1 of cattle manure, associated with doses of 18.9, 16.8 and 12.8 kg ha -1 of K 2 O, was responsible, respectively, for the highest contents of S 2.60, 2.73 and 2.79 g kg -1 (Figure 4).
The superiority of S contents in the treatments with higher amount of cattle manure, in relation to the lower doses, is due to the fact that the manure is source of S to the plants and, therefore, higher doses contribute with greater increments in the availability of the element in the soil to the plants (Higashikawa et al., 2010).
Watermelon plants, including those in treatments without the addition of cattle manure, showed S contents in the range of 2.0 to 3.0 g kg -1 , an indication that they were nutritionally balanced with respect to this element (Trani & Raij, 1997).
In relation to the test of contrasts between the additional treatment (without fertilizer) and the treatments with fertilization, it is observed that, except for Ca, the leaf contents of the other macronutrients were higher in the treatments that received manure and K (Table 4).
The manure, besides providing elements in small amounts to the soil, especially N, P and S, increases the efficiency of mineral fertilization and, therefore, the availability of nutrients to plants (Higashikawa et al., 2010).
For the contents of N and K in the leaf tissue, the superiority is mainly due to the supply of urea and potassium chloride.The inferiority of Ca contents in the treatments with K 2 O and manure strengthens the hypothesis of competition between these elements, compromising the absorption of Ca (Silva & Trevizam, 2015).

Figure 2 .
Figure 2. Leaf contents of calcium in watermelon cultivated in soil with K 2 O doses, without -0 (-) and with 2,520 g hole -1 of cattle manure (---) Figure 4. Leaf contents of sulfur in watermelon cultivated in soil with doses of K 2 O and 2,520 g hole -1 of cattle manure (----)