Characterization of vegetative and reproductive tillers as a basis to recommend heights and nitrogen doses for the stockpillage of Marandu palisade grass

This article aims to determine the adequate pasture height and the nitrogen dose at the beginning of the deferral period for Urochloa brizantha cv. Marandu (Marandu palisade grass). Two pasture heights (15 and 30 cm) and four nitrogen doses (0, 40, 80, and 120 kg ha -1) were evaluated in a completely randomized factorial (4x2) design with three replications. A linear increase in the number of vegetative tillers was observed with the application of nitrogen in the 15-cm deferred pasture. The lengths of stem and leaf blades of tillers increased linearly with the application of nitrogen. The weight of the vegetative tillers was greater in the 30-cm deferred pasture, compared to the 15-cm one when the pasture was fertilized with 0 and 40 kg ha-1 of N. The nitrogen fertilization linearly increased the weight of both the vegetative tillers in the 15-cm deferred pastures and the reproductive tillers in the 30-cm deferred pastures. The recommendation is that Marandu palisade grass be deferred with 15 cm and fertilized with 80 kg ha-1 of N at the beginning of the deferment period.


Introduction
Nitrogen fertilization in stockpiled pastures allows not only for maintaining the sustainability of the productive system through nutrient reposition, but also for increasing forage production in the winter (AMORIM et al., 2019). 1 Faculdade de Medicina Veterinária da Universidade Federal de Uberlândia, Uberlândia-MG. Reducing the height of the pasture before the start of the stockpiling period is also recommended to provide removal of old forage from the pasture and, in effect, to ensure a greater incidence of light at the base of the plants, which stimulates the development of basal buds in new vegetative tillers (SOUZA et al., 2012). Consequently, a stockpiled pasture is produced with a better structure and nutritional value, potentially resulting in improved animal consumption and performance in the winter.
Variations in the fertilizer dose and the height of the pasture used at the beginning of the stockpiling period promote changes in the number and morphological characteristics of the tillers, which are the growth units of the pasture (SANTOS et al., 2020). Eventually, identifying the tiller population density and the morphological characterization of the tiller classes in stockpiled pastures provides for management strategies that result in a deferred pasture with better characteristics for animal production.
This paper reports on an experiment aimed to determine the appropriate height and nitrogen level dose at the beginning of the stockpiling period for Urochloa brizantha cv. Marandu.

Material and methods
The experiment was carried out in a 150-m² pasture area with To determine the tiller leaf area, 40 leaf blades were harvested at random in each plot. They were cut at the ends so that they were rectangular in shape. Subsequently, the length and width of each were measured to estimate its area. The leaf blades were then dried in a forced ventilation oven at 65 ºC for 72 hours and then weighed. This was used for calculating the specific leaf area (SLA, cm 2 mg -1 ). The leaf area of each tiller was obtained by multiplying the mass of the live leaf blade of the tiller by the SLA.
In ten tillers of each category per plot, measurements were made for the lengths of the pseudostem and leaf blade, as well as for the numbers of live and dead leaves. The length of the pseudostem was measured from the ground level to the younger expanded leaf. The leaf blade length corresponded to the distance from the apex of the expanded leaves to its ligula. Expanding and expanded leaves were considered to be alive; those that had more than 50% of the senescent leaf blade were considered to be dead.
The numbers of reproductive tillers were analyzed descriptively only, due to data variability, with several values equal to zero. For the other variables, statistical analyzes were carried out at the level of 5% probability of occurrence of type I error. For the nitrogen dose, regression analyses were performed with a selection of models that best fit the data within each height pasture assessed. The largest response surface model in the averages of the nitrogen dose factor was as follows: Yi = β0 + β1Ni + β2Ni 2 + ei where: Yi = response variable; Ni = nitrogen dose; β0, β1, β2 = parameters to be estimated; ei = experimental error. The degree of adjustment of the models was assessed by the coefficient of determination and the significance of the regression coefficients was tested by the corrected t-test based on the residuals of the analysis of variance. For height (qualitative factor), a t-test was applied across the four doses of nitrogen.

Results and Discussion
There was a linear and positive effect (P<0,0001) of N doses on the number of vegetative tillers in the 15-cm deferred pasture (Table 2).
However, there was no effect (P <0.05) of nitrogen on the population of vegetative tillers for the 30-cm deferred pasture. In stockpiled pasture with a higher initial height (30 cm), the environment inside the canopy is generally more shaded (SOUZA et al., 2012). This shading is greater when the tallest pasture is fertilized with a high dose of N (120 kg ha -1 ) because, in this situation, the tillers have a higher growth rate (PEREIRA et al., 2015). As a result, tillering is inhibited, which is why there was a lower number of vegetative tillers in the high Marandu grass pasture as compared to the low one when the high dose of N was applied (Table 2).
There was no effect (P=0.1401) of nitrogen fertilization on the number of reproductive tillers in stockpiled pastures with 15 or 30 cm (Table 2). However, lowering the Marandu grass to 15 cm resulted (P=0.0075) in a lower number of reproductive tillers in the deferred pasture, mainly when lower doses of N were applied (Table 2) where brightness is greater.
In general, the length of the stem was greater (P=0,0011) in stockpiled pastures with greater initial height (Tables 3 and 4). Probably, the LAI was at or above the critical LAI in the 30 cm stockpiled pastures, as already discussed. Under this condition, the elongation of the stem is enhanced due to the greater competition for light inside the canopy.
In general, the leaf blade lengths of vegetative (P<0,0001) and reproductive (P=0,0016) tillers increased with nitrogen fertilization in stockpiled pastures with an initial height of 15 and 30 cm (Tables 3 and 4).
Nitrogen has a positive effect on the rate of cell division found in the leaf elongation zone (SKINNER & NELSON, 1995) and accelerates the process of producing new tissues (PEREIRA et al., 2015). For each characteristic, means followed by the same letter in the column do not differ (P>0.10) by the t-test; * Significant by t-test (P <0.05).
Stockpiled pastures with higher initial height may have tiller with longer leaf blades due to the longer stem length (Table 3). In longer tillers, the expanding leaf travels the longest path between its connection point with the meristematic region and the tip of the pseudostem and, consequently, the leaf elongation time is longer, which results in longer leaf blades (SKINNER & NELSON, 1995).
Nitrogen fertilization did not affect the number of live leaves in the 15 cm deferred pasture, both for vegetative tillers (Table 3) and for reproductive tillers (Table 4). There were also no differences (P>0,05) for the number of dead tiller leaves between 15 and 30-cm deferred pastures (Tables 3 and 4). For vegetative tillers, there was no effect (P=0,1513) of N fertilization on the number of dead leaves per tiller in 15-cm stockpiled pasture, but there was a linear increase in this characteristic (P<0,0001) when the pasture was stockpiled to 30 cm (Table 3). For the reproductive tillers, there was no effect (P>0,2108) of nitrogen on the number of dead leaves in the 30cm deferred pasture (P> 0.05), but there was a linear increase of this characteristic (P=0,0033) when the pasture was deferred by 15 cm (Table 4).
The increments in the number of dead leaves, as well as the decrease in the number of live leaves of tillers due to nitrogen fertilization can be justified by the greater shading of tillers in these fertilized canopies.
In the 30 cm stockpiled pastures, the vegetative (P<0,0001) and reproductive (P=0,0109) tillers had an increased linear effect of the leaf area with nitrogen fertilization (Table 5). The positive linear effect observed for the leaf area of vegetative and reproductive tillers is related to the length of the leaf blade, which increased with nitrogen fertilization in the 30-cm deferred pasture tillers (Tables 3 and 4).
A quadratic effect was observed in the 15-cm stockpiled pasture of both types of tillers, i.e, the leaf area of both vegetative and reproductive tillers was positively influenced by nitrogen fertilization to a certain point, after which nitrogen started to have a negative effect on the tillers area.
This may have occurred because higher doses of nitrogen accelerate the rate of leaf senescence (PEREIRA et al., 2011), given the more shaded environment inside fertilized and more developed canopies. For each characteristic, means followed by the same letter in the column do not differ (P>0.10) by the t-test; * Significant by t-test (P <0.05).
The weight of the vegetative tillers was higher (P=0,0100) in the 30cm stockpiled pasture compared to the 15 cm, when the pasture was fertilized with 0 and 40 kg ha -1 of N (Table 6). Probably, competition for light ensues between the tillers at the height of 30 cm, which caused the stem to lengthen and the weight of the tiller to increase. However, when the pasture was fertilized with 80 and 120 kg ha -1 of N (high doses), there were no differences in the weight of the vegetative tillers between the initial heights of the pastures (Table 6). This possibly occurred because the higher doses of nitrogen significantly increased the number of cells in the process of division (SKINNER & NELSON, 1995), which consequently stimulated the high growth of tillers. Nitrogen fertilization influenced positively and linearly (P=0,0064) the weight of vegetative tillers in 15 and 30-cm stockpiled pastures, as well as the weight of reproductive tillers in 30-cm stockpiled pastures (Table 6).
This was because N stimulates the growth of grasses.
When competition for light is small, i.e., when the pastures are lower, there are more small tillers. In higher pastures, the population density is lower for heavier tillers (SBRISSIA & SILVA, 2008). As such, an inverse relationship between the number and weight of the vegetative tillers was expected for the 15 cm stockpiled pasture, which did not occur ( Figure 1A).
This effect is probably because nitrogen simultaneously stimulated tillering and tiller growth. This, in turn, may have been the result of less competition for light within the deferred pasture with lower initial height. In contrast, there was no relationship between the number and the weight of vegetative tillers in the highest pasture at the beginning of the deferral period (30 cm) ( Figure 1B). In this pasture, the number of tillers did not increase with the weight of the tiller probably because of the more shaded environment inside the highest deferred canopy, which inhibits tillering.