1. Introduction
Nitrogen is the primary limiting macronutrient for tropical forage production, as it is the most extracted element by plants from the soil [
1]. Its role in protein synthesis and structural organ development makes it essential for sustaining plant growth. Consequently, the absence or insufficiency of nitrogen in soils is a leading cause of pasture degradation and reduced productivity in tropical grasses [
2]. Replenishing soil nitrogen through fertilization is, therefore, a critical practice in tropical forage-based systems.
Megathyrsus maximus cultivars are known for their high productivity, with annual forage yield exceeding 25 t DM ha
−1 [
3]. However, there are important limitations to consider when using these grasses. They exhibit low productivity under water deficit conditions [
4] and require strict control of grazing height. When managed above the recommended height of 35 cm [
5], there is an excessive production of stems and dead material which are morphological components that reduce forage intake [
6,
7], lowering the nutritional value of the forage and impairing animal performance [
8,
9].
Pasture degradation is led by the lack of nutrient replenishment and overgrazing, which lowers forage productivity and facilitates weed growth. In this context, nitrogen fertilization improves pasture resilience by promoting leaf area development, tiller density, and total biomass yield [
10,
11], leading to increased leaf elongation and appearance rates, reduced phyllochron [
11,
12], and changes in the bromatological composition of forages [
10].
Moreover, fertilized forages exhibit a higher crude protein content and changes in the allocation of protein in different structures within the grass [
13,
14]. Nonetheless, the excessive return of nitrogen to soil is associated with environmental risks [
15], such as nitrate leaching and toxicity, highlight the need for efficient nitrogen management. Nitrogen leaching losses range from 8 to 30 kg ha
−1 yr
−1 under moderate application rates and can exceed 100 kg ha
−1 yr
−1 with high nitrogen fertilization rates [
16].
For this reason, to avoid losses, nitrogen doses must be studied, as nitrogen is the most commonly recommended nutrient for the maintenance and recovery fertilization of pastures. Although maximum nitrogen application rates have been established for
Brachiaria brizantha cultivars [
17], such recommendations are lacking for short
Megathyrsus maximus cultivars. Several studies have evaluated nitrogen fertilization rates for
Megathyrsus maximus cultivars; however, there is still no standardization regarding the splitting of nitrogen applications or the maximum amount that should be applied after each defoliation cycle. For the cultivar
Megathyrsus maximus cv. BRS Zuri, there is evidence that a dose of 647 kg ha
−1 of nitrogen promotes greater dry matter production [
18]. Under irrigated conditions, annual nitrogen rates exceeding 1200 kg ha
−1 are recommended for the BRS Tamani cultivar to enhance forage accumulation [
19].
In this context, the objective of this study was to determine the maximum nitrogen dose that can be applied per regrowth cycle for Megathyrsus maximus cv. BRS Tamani, based on its effects on forage production, nutrient uptake, bromatological composition, and in vitro degradation kinetics.
2. Materials and Methods
2.1. Local and Experimental Design
This experiment was conducted at the Federal University of Rondonópolis (latitude 16°27′50.9′′ S, longitude 54°34′50.0′′ W, and altitude 284 m). Meteorological data were obtained from a weather station located 600 m from the experimental area (
Figure 1). According to the Köppen climate classification, the region has an Aw-type tropical savanna climate, characterized by a rainy season from October to April and a dry season from May to September, in which the total precipitation measured in the experimental period was 2508 mm (
Figure 1).
The experimental area comprised 560 m
2 and was established with
Megathyrsus maximus Jacq. cv. BRS Tamani in 2015. Following establishment, the area was subjected to sheep grazing without the application of fertilizers. The soil in the experimental area was classified as Dystric Rhodic Ferralsol [
20].
On 14 October 2019, soil samples were collected for chemical and particle size characterization. For sampling, 15 soil samples were collected at a depth of 0–10 cm using a soil auger. These simple samples were homogenized into a single composite sample, which was used to estimate macronutrient contents and particle size distribution (
Table 1).
On 4 December 2019, following the onset of the rainy season, a uniformity cut of the grass was performed, leaving a residual height of 25 cm. Maintenance fertilization was carried out as recommended by Martha Junior et al. [
21]. Across the entire experimental area, 60 kg ha
−1 of P
2O
5 was applied in the form of single superphosphate (18% P
2O
5; 10% S), and 60 kg ha
−1 of K
2O was applied using potassium chloride (58% K
2O). Nitrogen fertilization was carried out according to the predetermined rates, using ammonium nitrate (29% N) as the nitrogen source. On 4 November 2020, 50 kg ha
−1 of P
2O
5 (single superphosphate, 18% P
2O
5; 10% S) and 50 kg ha
−1 of K
2O (potassium chloride, 58% K
2O) were applied. The same potassium dose was reapplied on 26 January 2021.
The experimental design was a randomized complete block design with five treatments and seven replications. Treatments consisted of nitrogen application rates following each defoliation of BRS Tamani: 0 (Control), 40, 80, 120, and 160 kg N ha−1. Experimental plots measured 16 m2 and were planted with BRS Tamani (Panicum maximum cv. BRS Tamani; syn Megathyrsus maximus cv. BRS Tamani). The experiment was evaluated over two rainy seasons: from December 2019 to April 2020 (year 1), and from November 2020 to April 2021 (year 2). During both years, in the absence of nitrogen fertilization (Control), two regrowth cycles occurred, while in the other treatments, five regrowth cycles were observed.
2.2. Measurements
From December 2019 to April 2020, canopy height and light interception were measured weekly. Canopy height was measured at four points per experimental plot using a graduated ruler. Light interception was assessed at six points per plot using an AccuPAR LP-80 ceptometer (Decagon Devices Inc., Pullman, WA, USA). The chlorophyll index was obtained utilizing a chlorophyll meter ClorofiLOG® 1030 (Falker, Porto Alegre, Brazil). When the average canopy from a given treatment intercepted 95% of the photosynthetically active radiation, evaluations of forage mass, botanical composition, and morphological composition of the pasture were conducted.
For forage mass estimation (kg DM ha
−1), a 1.0 m
2 metal square were used, and all plant material within the square was harvested, maintaining a residual height of 25 cm. The forage yield (kg DM ha
−1) was calculated as the sum of average forage mass of all regrowth cycles. The forage accumulation rate (FAR) was estimated by using the ratio of the forage mass to the regrowth interval. Three samples were collected per experimental plot. Subsequently, all plots were uniformly cut at the residual height of 25 cm using a handheld mechanical mower. After each forage harvest, nitrogen fertilization was applied according to the assigned treatment. For nitrogen application rates ranging from 40 to 160 kg ha
−1, five harvests were performed in both year 1 and year 2, whereas in the control treatment (absence of nitrogen fertilization), two harvests were carried out in year 1 and three in year 2 (
Table 2).
After forage harvesting, the total forage mass collected from each of the three samples per experimental plot was weighed. These samples were then combined, and three subsamples were separated: subsample 1 was used to estimate the air-dried sample weight, subsample 2 for botanical composition analysis (forage and weed species), and subsample 3 for morphological composition analysis (leaf blades and stem + sheath). Morphological separation was performed only for harvests in which the grass was at the reproductive stage (March 2020 and April 2021), as only leaf blades were collected for the other evaluations. In the last evaluation of the rainy season, conducted in April 2020, tiller counts were performed using a 0.25 m2 square to estimate the tiller population density.
Dried subsamples were ground using a knife mill to pass through a 1 mm screen for the quantification of dry matter (DM), ash, and crude protein (CP; Kjeldahl method), according to Detmann et al. [
22]. Neutral detergent fiber content corrected for ash and protein (aNDFom) was determined following Mertens [
23], using thermostable α-amylase and omitting sodium sulfite. Corrections for ash and protein content in NDF were applied.
The content of neutral detergent insoluble protein (NDIP) was quantified [
24] and referred to as crude protein in the cell wall (CP
cw). Using Equation (1), the crude protein content in the cellular contents (CP
cc) was estimated:
in which CP
cc is the crude protein content in the cellular content, and CP
cw is the crude protein content in the cell wall.
The content of non-fiber carbohydrate plus ethereal extract (NFC + EE) in the produced forage was estimated using the following equation [
25]:
in which CP is crude protein, and aNDFp is NDF corrected for ash and protein.
Macronutrient contents (N, P, K, Ca, Mg, and S) were quantified according to Malavolta et al. [
26], and nutrient uptake was estimated using the following equation:
The nitrogen use efficiency (NUE) was calculated as:
2.3. Gas Production and Kinetic Parameters of Ruminal Fermentation
Approximately 300 mg of sample material was weighed into 100 mL amber glass vials. Rumen fluid was collected from two rumen-cannulated steers, fed a diet of 60% corn silage and 40% concentrate. The fluid was filtered through three layers of cheesecloth, stored in a pre-warmed thermos, and immediately transported to the laboratory and kept in water bath at 39 °C under CO
2 purge. The McDougall’s buffer solution [
27], supplemented with a reducing agent containing HCl-cysteine, sodium hydroxide (0.1 mol/L) and resazurin as an indicator of oxidation–reduction was added to each vial under a CO
2 purge was mixed in a proportion of 4:1 of buffer to rumen fluid. Each vial received 30 mL of the mixed media solution under CO
2 flushing and was immediately sealed with a rubber stopper and aluminum cap. Incubations were conducted in a shaking water bath at 39 °C.
Gas pressure was measured using a pressure transducer connected to a needle at 2, 4, 6, 8, 10, 12, 24, 48, 72, and 96 h post-inoculation. Blank vials containing only buffer and rumen fluid (without substrate) were incubated in parallel to correct for background gas production. The gas volume released from the sample vials was adjusted by subtracting the corresponding blank values. The conversion of pressure (psi) to gas volume (mL) was adjusted using a regression equation. For this purpose, known volumes of air were injected into sealed amber flasks containing only distilled water, at a volume equivalent to that used in the incubations. These flasks were maintained under the same experimental conditions. Based on the relationship between the injected volumes and the resulting pressures, the following linear regression equation was determined: Gas volume (mL) = 6.4882 × gas pressure − 0.5589.
After 24 h of incubation, the flasks were removed from the water bath and placed in cold bath to stop fermentation. Residues were recovered by filtration through pre-weighed non-textile fabric filter bags and dried in a forced-air oven at 105 °C for 16 h, and the residual dry matter was weighed to calculate in vitro dry matter degradability (IVDMD). For in vitro neutral detergent degradability (IVNDFD) measurement, 50 mL of neutral detergent solution was added to the vials, sealed again, and autoclaved at 105 °C for one hour in order to extract all components that were soluble in neutral detergent [
28]. Then contents were filtered under a vacuum in non-textile fabric filter bags. The NDF residue was obtained after the material was dried in a non-ventilated oven at 105 °C for 16 h.
Cumulative gas production kinetics were modeled using the bicompartmental logistic model proposed by Schofield et al. [
28]:
where Vt is the cumulative gas volume (mL) at time T; V1 and V2 are the asymptotic gas volumes from the rapidly and slowly fermentable fractions (mL), respectively; C1 and C2 are the respective fermentation rates (h
−1); Lag is the lag phase (h); and T is the incubation time (h).
2.4. Statistical Analysis
Nitrogen application rates were considered fixed effects, while evaluation years were treated as random effects. Regression analysis (linear, linear plateau, and quadratic) was used to assess the effects of nitrogen doses (
p < 0.05). Descriptive analysis was performed for pre-grazing light interception, pre-grazing height, post-grazing height, and regrowth interval. The following model was used:
where y
ijk = expected response; μ = average, associated with the experiment; T
i = treatment effect (nitrogen dose per regrowth cycle)
i; e
ij = treatment error
i, in replicate
j, normally and independently distributed; C
k = random effect associated with year
k, normally distributed; and ɛ
ijk = experimental error associated with treatment
i, in replicate
j, in year
k, normally distributed.
3. Results
Nitrogen fertilization doses ranging from 40 to 160 kg ha
−1 resulted in light interception values close to 95% when the average canopy height reached 42 cm (
Table 3). However, grasses that did not receive nitrogen fertilization did not reach 95% light interception; therefore, evaluations were performed for these treatments when the canopy reached the average height used for the other treatments (
Table 3). Nitrogen fertilization reduced the interval between harvests compared to the absence of nitrogen application (
Table 3).
Nitrogen fertilization affected (
p < 0.05) all measured variables, except for tiller density, percentage of leaf blades, and stem + sheath (
Table 4). Forage mass increased from 1795 kg ha
−1 in the control (0 kg N ha
−1) to a maximum of 2225 kg ha
−1 at 80 kg N ha
−1, followed by a slight decline at higher doses. Similarly, forage yield increased from 4062 kg ha
−1 in the control to a peak of 9798 kg ha
−1 at 40 kg N ha
−1, with a plateau observed thereafter.
The forage yield rate increased with nitrogen application, ranging from 19 kg ha−1 day−1 at 0 kg N ha−1 to 60 kg ha−1 day−1 at 40 kg N ha−1. On the contrary, NUE decreased with increasing N doses, dropping from 91 kg DM kg−1 N at 40 kg N ha−1 to 36 kg DM kg−1 N at 160 kg N ha−1 dose.
Weed biomass was completely suppressed in all fertilized plots (0 kg DM ha−1), compared to 14 kg DM ha−1 in the control, indicating a strong suppressive effect of nitrogen application (p < 0.001). The chlorophyll index increased consistently with nitrogen dose, from 25 in the control to 38 at 160 kg N ha−1 dose (p < 0.001).
Nitrogen was the only nutrient whose uptake by BRS Tamani increased (
p < 0.001) up to the dose of 80 kg ha
−1, whereas for the other macronutrients, extraction increased only up to 40 kg ha
−1 per regrowth cycle (
Table 5). Nitrogen and potassium were the most extracted nutrients by BRS Tamani, while phosphorus and sulfur were the least extracted (
Table 5). Nitrogen fertilization resulted in an 85% increase in nitrogen uptake, while the increase in uptake of the other nutrients ranged from 68% to 78% (
Table 5).
Nitrogen fertilization resulted in increased (
p < 0.05) contents of crude protein (CP) and neutral detergent-insoluble protein (NDIP), a reduction in ash, and had a minor effect on neutral detergent fiber (NDF) content (
Table 6). Under nitrogen fertilization, the CP and NDIP contents increased by approximately 61%, while the ash content decreased by 19% (
Table 6). Cell contents (CPcc) and cell wall-associated crude protein (CPcw) both increased with nitrogen fertilization (
p < 0.001). CPcc increased from 24 g kg
−1 DM in the control to 66 g kg
−1 DM at 80 kg N ha
−1, while CPcw increased from 44 to 125 g kg
−1 DM across the nitrogen gradient (
p < 0.001). When expressed relative to total CP, the proportion of CPcc and CPcw shifted with the nitrogen fertilization rate: CPcc (g kg CP
−1) peaked at 373 g kg
−1 CP at 40 kg N ha
−1 and declined thereafter, while CPcw (g kg CP
−1) was lowest in the control (681 g kg
−1 CP) and highest at 160 kg N ha
−1 (714 g kg
−1 CP), with both responses showing significant quadratic effects (
p < 0.001).
Nitrogen doses fertilization resulted in a decreased (
p < 0.001) final volume ranging from 250 mL at the control level and decreasing up to 206 mL at dose of 80 kg ha
−1 increased (
p < 0.001) the degradation rate up to 40 kg ha
−1, and lag time up to 120 kg ha
−1 (
Table 7). The in vitro degradability of DM and NDF increased (
p < 0.001) up to the dose of 40 kg ha
−1, where the degradability ranged from 702 to 726 (g kg
−1), and 698 to 750 g kg
−1, for IVDMD and IVNDFD, respectively (
Table 7).
4. Discussion
Higher nitrogen doses are uneconomical, as they reduce nitrogen use efficiency and do not lead to additional forage yield, the key variable for increasing the growth rate of pasture stock, which directly impacts productivity per area. In a study performed by Delevatti et al. [
29] evaluating increasing nitrogen deses up to 270 kg N ha
−1 on
Urochloa brizantha cv. Marandu, a slight increase of 154 kg ha
−1 in forage mass from the dose of 90 to 270 kg N ha
−1 was observed. Furthermore, the nitrogen use efficiency decreased 106.8 g N used by kg of N applied, from the dose of 90 to 270 kg N ha
−1. Moreover, at the tested dose of up to 160 kg ha
−1, nitrogen not absorbed by the plant may be lost through leaching, potentially leading to environmental concerns [
15]. A dose of 40 kg N ha
−1 was the dose at which BRS Tamani reached its maximum forage yield, and thereafter we can assume that the excess N was not used by the plants and possibly leached into the soil.
Nitrogen fertilization significantly increased the forage accumulation rate (
Table 4), which resulted in a proportional reduction in the interval between defoliation cycles (
Table 2). This reduction in the interval can lead to an increased frequency of defoliation in grazing systems managed under intermittent stocking, potentially enhancing pasture utilization and productivity.
We hypothesize that the response of forage mass production to higher nitrogen doses is associated with the capacity of the grass to uptake nitrogen from the soil, since nitrogen uptake reached its limit at the same dose as the forage mass and yield. This response can be related to the morphological and physiological capacity of this cultivar synthesize root mass. Similarly,
Panicum maximum cv. Tanzania showed slightly increases in root mass in doses above 50 kg N ha
−1 [
30].
The increase in forage yield resulting from nitrogen fertilization led to a greater extraction of all macronutrients. Nitrogen uptake increased up to the dose of 80 kg ha
−1, whereas the extraction of the other macronutrients increased only up to 40 kg ha
−1. Therefore, exclusive nitrogen fertilization in degraded pastures on soils with low nutrient availability may exacerbate nutritional imbalances and accelerate soil nutrient depletion. Accordingly, proper planning of nitrogen fertilization must be accompanied by the replenishment of other essential nutrients, mainly potassium which has an associative effect on forage production when fertilized with nitrogen at nitrogen-to-potassium ratios from 1.3 to 2:1 [
13].
In contrast, in the absence of nitrogen fertilization, nutrient uptake was lower; however, the development of BRS Tamani was compromised, as evidenced by reductions in forage mass, forage yield, and the accumulation rate, along with an increased presence of invasive plants (
Table 4). The occurrence of invasive species is associated with the forage’s reduced ability to cover the soil surface, a consequence of the lower forage accumulation rate, which leads to slower regrowth and thereby favors the establishment of weeds [
31].
Moreover, the absence of nitrogen fertilization results in smaller leaf sizes and lower rates of leaf appearance and elongation [
11,
12], which further limit the plant’s potential for soil coverage. In the present study, this condition prevented non-fertilized plants from intercepting 95% of the photosynthetically active radiation (
Table 3). Therefore, nitrogen application at doses starting from 40 kg ha
−1 was sufficient to completely suppress the presence of weed species. The occurrence of invasive plants can increase production costs due to the need for herbicide use and may also be an animal health risks, as some weed species may be toxic to livestock [
32,
33,
34].
In addition to mitigating and delaying pasture degradation, nitrogen fertilization also influenced the chemical and bromatological composition of BRS Tamani. The observed reduction in ash content is likely due to a dilution effect, as increased forage mass leads to a lower concentration of nutrients per unit of dry matter. Despite the decline in mineral content, caution is necessary to avoid mineral antagonism. For instance, nitrogen fertilization using ammonium sulfate may induce an antagonistic interaction between sulfur and copper, potentially resulting in reduced copper concentrations in cattle diets in grazing systems, as reported by Arthington et al. [
35].
The most pronounced effect of nitrogen fertilization on the chemical and bromatological composition was observed in the increase in CP, CP
cw, and CP
cc, with increments of 154%, 159%, and 163%, respectively, compared to the absence of nitrogen fertilization. This effect of nitrogen fertilization on protein fractions has been widely documented in the literature [
10,
29]. However, it is important to note that approximately 66% of the total protein in BRS Tamani is located in the cell wall which is a nitrogen fraction characterized by slow and incomplete digestibility by ruminants [
36].
In contrast, when
Urochloa brizantha cv. Marandu was fertilized with nitrogen, around 70% of the crude protein was allocated to the cellular content, a fraction with rapid and nearly complete digestibility [
17]. Therefore, the main advantage of
Megathyrsus maximus cultivars over
Urochloa brizantha lies in their higher stocking rates due to greater forage productivity. In terms of average daily weight gain per animal, however, no significant differences have been consistently observed [
37].
Although a substantial increase in CP content was observed, the IVDMD and IVNDFD increased by only 4% and 8%, respectively. This limited improvement in degradability is likely due to an imbalance between protein and energy in the forage, as nitrogen fertilization has little effect on NDF and results in a reduced NFC content (
Table 6). This imbalance helps explain the positive effects on animal performance observed with energy supplementation during the rainy season [
38,
39,
40], as such supplementation more effectively aligns the protein and energy supply with the nutritional demands of the animals.
The extension of the lag phase is generally not associated with the absolute protein content, but rather with the form in which the protein is present in the plant. In this study, most of the protein was bound to the cell wall, and more protein is associated with fiber or compounds that are less accessible to microbial attack leading to a longer lag phase in ruminal degradation kinetics [
41]. It is noticeable that the longer lag time with the increase in nitrogen dose fertilization directly affected the total volume of gas produced.
The production of gas in in vitro cultures is associated with the digestion of nutrients in the media by the microorganisms breaking down the nutrients to volatile fatty acids [
42]. However, in this study, treatments that included nitrogen fertilization showed higher in vitro degradability and lower gas production. One hypothesis for this pattern could be the higher concentration of CP in the cell wall which creates a structure with slow and less degradable contents [
43]. Since the nitrogen fertilized treatments presented close IVDMD and IVNDFD values, the gas production observed may be related to the more degradable protein and carbohydrate fractions in the plant cell.