Influence of harvest time in accumulation of biomass between cassava cultivars

Lack of cultivar adapted to the planting area, as the ideal harvest period undertakes the productive potential of cassava. The study was conducted to evaluate the influence of different harvest seasons on total production of fresh and dry weight of four cassava cultivars. The used cultivars were: Branca de Santa Catarina (BSC), IAC 13, IAC 14-18 and Fiber that they were planted on 09/20/11 at the headquarters of the Regional Center Valley Paranapanema Paulista Agency for Agribusiness Technology (APTA), in Assis, São Paulo. The experimental design was completely randomized in 4x6 (four cultivars and six harvest seasons) factorial schemes with four replicates, being 4, 8, 12, 16, 20 and 24 months after planting (MAP) the harvest season, totaling 24 treatments and 48 installments. The total production of fresh biomass (FB) and dry matter (DM) in tons per hectare of 12 plants per harvest season was evaluated. Regarding the production of FB as DM among cultivars was not observed significant differences between them. However, the cultivars showed accumulation of FB linearly. The best time to harvest in order to higher accumulation of DM occurred at 12 and 24 months, corresponding to the end of the first and second round of cassava cultivation, respectively.


INTRODUCTION
The cassava plant (Manihot esculeta Crantz) belongs to the family Euphorbiaceae, originally from the American tropics where it is cultivated for nearly four thousand years.It is currently one of the crops with the highest potential for energy production even at low agronomic conditions and limited economic partner, since cassava has drought resistance, tolerance to low soil fertility, which facilitates its cultivation and promote a high yield potential (El-Sharkawy et al., 2008).
In recent years the cassava has been the energetic basis for millions of low-income people in many tropical and subtropical countries around the World (FAO, 2014).In Brazil conditions, the current crop has grown just over 2.3 million hectares, with an estimated production of 23.4 million tons of cassava roots and yield per area of 17.7 t ha  (1979), is much higher, reaching 90 t ha -1 of fresh roots.There are reports in Kenya, evaluating cassava varieties, that observed yields higher than 100 t ha -1 , with some genotypes showed yield potential of roots above 150 t ha -1 (IITA, 2005).The lack of cultivar specific and adapted to the planting area, commits to generating information about maximum yield, as the harvest season.The ignorance of the ideal harvest time may result in losses to the producer either through loss due to early harvest, when to cultivate not reached its maximum level of dry matter accumulation in roots, or when late harvest, increases the rate of decay roots and the fiber content, decreasing in quality (Alves, 2002;Cenoz et al., 2005;Lopes et al., 2010;Alves, 2006).Although the period of higher dry matter accumulation in roots coincides with the maturity of branches that can be performed with one or two cycles, 12 or 24 months, respectively some cultivars differ from these periods either by genetics or by climatic influences (Valle and Lorenzi, 2014).
In general, farmers have demanded shorter cycle's materials.However, even the most demanding focuses on materials with higher levels of starch and dry matter accumulation, in order to add value to the final product.However, it is necessary to develop more specific regional and cassava studies in order to achieve greater efficiency and productivity.Given the above, this research was carried out with the objective of evaluating the influence of different harvest seasons on total production of fresh and dry weight of four cassava cultivars.

MATERIALS AND METHODS
The study was conducted at the headquarters of the Regional Center Valley Paranapanema Paulista Agency for Agribusiness Technology (APTA), located in Assis, SP (latitude 22º 40'S, 50º 26'W longitude and altitude average 563 m) between 09/20/2011 (planting) to 9.24.13 (last harvest).The climate, according to Köppen Cwa is subtropical dry, hot summer and wet winter.Climatic data recorded during the experimental period are shown in Figure 1.
The soil preparation consisted of plowing with disc plow to depth of 20 cm, and did not occur distort and leveling because it is a sandy soil, there is no need for this procedure.The herbicide trifluralin was applied to the soil in a commercial dose of the 20 g i.a.ha -1 .In the experiment was worked with the natural fertility of soil, it did not receive fertilizer to avoid interfere for the treatments.The furrow was done on soil with 10 cm deep and 100 cm distance between grooves.
Four cassava cultivars commonly grown by farmers from São Paulo were used: Branca de Santa Catarina, large industrial use, low habit and open branching; IAC 13, midrange and high closed branch; IAC 14-18 midsize, closed branches and grow fiber for industrial use without branching habit.All cultivars were planted at spacing of (100 cm × 100 cm) which represented a total of 10,000 plants ha -1 .
The planting material of cassava was produced in Roots and Tubers Section from IAC. Good branches with physiological and health quality, 10 cm long and five viable buds were placed horizontally in the planting furrows, on the propagation way to cassava.
The experimental design was completely randomized in 4×6 factorial schemes with four replicates, with four cultivars and six harvest seasons 4, 8, 12, 16, 20 and 24 months after planting (MAP), totaling 24 treatments and 48 plots.The plots consisted of four rows of eight plants, using as useful the two central lines totaling 12 harvested plants per plot.The cuttings were planted on 09/20/11, and four months later, it was proceeded the first harvest and repeated it five times spaced four months until the last harvest on 09/24/2013, thus covering all phenological stages of cassava.The plants were collected, weighed on analytical balance and recorded the fresh biomass.After weighing, the plants were dried in oven with forced air ventilation at 65ºC for 72 h; they were ground in Willey mill to determine the dry weight.After this, these data were corrected to tons per hectare by multiplying the weight per plant by total plants per hectare.
Data were subjected to variance analysis by F test at 5% significance.It was held normality of Shapiro-Wilk and homogeneity of variance using the Bartllet test.Just as the response variable dry biomass did not meet the assumption of homogeneity, then the same was submitted to the Box and Cox transformation (Box and Cox, 1964), which provided for the examination of variance conditions.
Due to these quantitative variables, the results for harvest times on fresh and dry weight were assessed by regression analysis.The variable response in fresh biomass was adjusted with linear regression (ŷ = a + bx) model, while the data dry mass adjusted to cubic polynomial regression, the model represented by (ŷ = a + bx + cx 2 + dx 3 ).The mean qualitative cultivars factor was compared by Tukey test at the 5% level of probability.Statistical analyzes were performed using software (R Development Core Team, 2014).

RESULTS AND DISCUSSION
For the variables fresh biomass (FB) and dry matter (DM) were significant, except for the time factor of harvest (p≤0.05).The varieties factor and the effect of double interaction between factors was not significant (p≥0.05)(Table 1).Table 2 shows the data about Tukey test involving FB and DM among the four cassava cultivars.
It may be noted that there was no significant difference (p≥0.05) on the cultivars.The averages shown in Table 2, it is observed that the values are within the limits recommended by other studies (Vidigal Filho et al., 2000;Alves and Modesto Junior, 2009;Dias et al., 2009).
The fact is that significant differences did not occur, demonstrates that both can be planted by the producers of the Valley Paranapanema in Assis, SP region.The DM content is usually the feature that determines the greater or lesser amount paid by industries from producers at the time of marketing, since it is directly related to the performance of various industrial products derived from cassava (Schons et al., 2007).In this context are desirable cultivars that account for the greatest productions of tuberous roots are also those that have the highest levels of FB maximize the yield of the final product per unit of cultivated area, which was verified in this work.
It was observed from Figure 2 that the FB of cassava cultivars showed a linear behavior as a function of harvest time.With the persistence of plants in the field for each four month will be no increase in the production of FB of 2.66 t ha -1 . Since the adopted model explains 94% of the observed production of FB in relation to different harvest times.Similarly, Azevedo et al. (2006) studied the production of the aerial part of cassava cultivars found that the production of FB increased linearly up to 12 months after planting, the period evaluated by the authors.
According to the statements of Mendonça et al. (2003) in crop of cassava, it is performed earlier with a vegetative cycle, 6-14 months after planting.However, when dealing with cultivars for industrial purposes, as the materials using in this study, the crop can be performed with one or two cycles, at 12 or 24 months after planting, respectively, which coincide with higher yields when compared to samples taken with a cycle.Carvalho et al. (2007) evaluated eleven cassava cultivars in three harvest seasons, 14, 16 and 18 months, also found that the average yield of fresh roots increased linearly as a function of harvest times, and the average values of 21, 22 and 24 t ha -1 , respectively.Although some authors claim that cassava does not have a specific time for your harvest and can be harvested in accordance with the needs of the producer, however, its processing must occur within 72 h after harvest due to enzymatic transformations and microorganisms action thus compromising the quality of the starch present in the roots (Lorenzi, 2003;Mattos, 2002).
Unlike FB and DM do not maintain behavior incrementally, but specific periods of greater accumulation, and generally the best times to harvest cassava aiming DM should consider the physiological state of the plant, because it had introduced the total or partial defoliation period before beginning the new branch, which corresponds to 12 and 24 months after planting (Alves, 2006;Otsubo et al., 2008.).The average DM yield obtained set cubic polynomial regression (Figure 3).Several studies corroborate the results obtained in this study (Sagrilo et al., 2007;Otsubo et al., 2008;Vidigal Filho et al., 2000).
It is observed that from four to 12 months after planting there is an increase in DM accumulation of cassava varieties of 1.56 t ha -1 to 2.06 t ha -1 , respectively, an accumulation of approximately 75% to about eight months (Figure 3).This is due to the fact that from four to eight months the plant canopies, the amount of paper sheets can intercept much of the light that falls on them, resulting in this period the largest partition DM to branches and roots.
According to Alves (2006) to complete the first cycle, which corresponds on average 12 months, the cassava plant has the highest rates of accumulation of DM in the roots of booking, this is an ideal time to harvest the roots.However, in this work we proceeded to conduct the plants to a second round of culture, extending up to the 24 months.The results obtained in this study are in agreement with reports of Távora et al. (1995); Peressin et al. (1998) when they state that after 12 months, when starting the second cycle, there is a drop in the total DM, this is due to resume in the rate of growth of leaves and branches and the issuance of new roots.According to the authors the maximum period of this redistribution occurs in about four to six months, which confirms the results obtained here, for a fall of 4% of DM from 12 to 16 months (Figure 3) was observed.Similarly observed in the first cycle, the behavior of DM accumulation is repeated in the second, this time accumulating increasingly the DM of 16 to 24 months after planting.
Thus, our findings confirm the well-defined cycles of cassava year after year, and the translocation of

Conclusions
There is no difference in the accumulation of fresh and dry weight among cultivars: Branca de Santa Catarina, IAC 13, IAC 14-18 and fiber.Cultivars accumulate fresh weight linearly.The highest accumulation of dry biomass occurs at the end of each production cycle, 12 and 24 months after planting.

Figure 2 .
Figure 2. Average yield of fresh biomass (FB) of cassava plants (t ha -1 ) depending on the time of harvest, regardless of the cultivars.

Figure 3 .
Figure 3. Average production of dry matter (DM) of cassava plants depending on the time of harvest, regardless of the varieties.

Table 1 .
Analysis of variance for factorial 4×6 on production of fresh biomass (FB) and dry matter (DM) in t ha -1 .

Table 2 .
Tukey test for total production of fresh biomass (FB) and dry (DM) in tons per hectare between cassava cultivars BSC, IAC 13, IAC 14-18 and fiber.
Means followed by the same letters within the column do not differ by Tukey test (p≥0.05); 1 Branca de Santa Catarina.