Materials and methods

Location and characteristics of the study area

The work was done from October 2012 to September 2013 in five cattle ranches with cattle production system in grazing dual purpose, located in the common Punta Arena, Catazajá Municipality, Chiapas, Mexico. This community is located between the coordinates 17° 46' 50" north latitude 92° 06' 50" west longitude and 17° 43' 13" north latitude 92° 01' 03" west longitude, with warm humid climate with rain all year (Af) (57.41%) and warm humid with abundant summer rains (Aw) (42.59%), with average annual rainfall of 2 600 mm (^{Tamayo, 1985}) and annual average temperature of 26° C (^{INEGI, 2009}), with heights ranging from 9 to 10 meters. The region is characterized by three times similar to those described by ^{Escobedo (1989)} are the days of "norths" (October-January) at low temperatures and cloudy sky well defined year, the rainy season moderate to severe (June to September), and the dry season (February to May) with lack of moisture, increasing sunlight and high temperatures. The predominant forage in pastures are: Paspalum notatum (ranches 1, 3 and 4), Brachiaria humidicola (ranch 2), Brachiaria brizantha and Brachiaria hibrida (ranch 5).

Sampling

Forage samples were taken monthly from October 2012 to September 2013. To estimate the accumulation of forage every 30 days, exclusion cages were placed considering the proposed ^{Mannetje and Jones (2000)} procedure. The 30 exclusion cages 0.5 m wide x 1 m long x 1 m high made of rebar half an inch thick covered with a metal mesh used. The location of the cages considered the height variation forage selecting three layers: low, medium and high, randomly placing two cages exclusion each stratum, for a total of six cages per cattle ranch. In September 2012 (30 days before sampling), he cuts forage in the area of the cage exclusion at a height of 2 to 3 cm from the f loor (^{Pérez-Prieto and Delegarde, 2012}). The forage samples were taken using a quadrant of 0.25 m2 within the cage. First, within the quadrant 20 measurements forage height became subsequently forage at the same height (2 to 3 cm of soil) was cut, then the remaining at the same height fodder cut in and around the area exclusion cage to avoid interference from adjacent forage sunlight.

The forage samples were placed in paper bag and were weighed on a digital scale. Subsequently, the collected fodder was dried in a forced air oven at 65 °C for 48 h; dry forage

Variables analyzed

Monthly and annual accumulation forage

The monthly forage accumulation was calculated by determining the cumulative (PSA) forage dry weight and was estimated to weigh monthly dry matter (MS) accumulated in the exclusion cages, while the annual forage accumulation was estimated by determining the annual production of dry matter (PAMS) and monthly it estimated by adding the dry matter. Both are expressed in kg MS ha-1.

Forage growth rate

The growth rate (TC) forage (^{Chapman and Lemaire, 1993}) was calculated based on the amount of dry matter accumulated forage in the exclusion cages divided by the cutting range forage, and is expressed in kg MS ha-1 d-1.

Where: TC= growth rate (kg MS ha-1 d-1); PSA= accumulated forage dry weight; t= days between cuts.

Statistical analysis

The ALT, PSA, TC and PAMS data were analyzed using the general linear model (GLM) of the SAS program (2004) for a completely randomized design. The statistical model for data ALT, PSA and TC, considered the effects ranch (Ri) [1, 2, 3, 4, 5], month of the year (Mj) [october, November ... September], time of year (Ek) [norths, dry, rains], forage species (Sl) [B. brizantha, B. humidicola, B. hibrido, P. notatum] and interactions interacciones (R*Mij), (R*Eik), (R*Sil), (M*Sjl) y (E*Sks). The PAMS data only used for comparison between ranches. Tukey test (^{Steel et al., 1997}) was used to compare treatment means.

Monthly forage accumulation

There was effect (p≤ 0.05) of the interaction ranch*month of the year, ranch*season and forage species*month of the year ALT and PSA forage. For a better view of the production behavior of forages studied, in Figure 1 rainfall (PP), temperature (T°),ALT and PSAare presented throughout the year in the study area.

Figure 1 Rainfall (PP), temperature (T°), accumulated dry weight (PSA; n= 720) and height (ALT; n= 7200) of forage during the year in Humid Tropical conditions in southeastern Mexico 

As shown, the higher and forage production corresponds to the months with the highest PP (june to september) and followed by the windy season (october to january), while lower production corresponds to the dry season (february to may) when T° is greater and PP is lower.

The interaction ranch * month of the year inALT was because it was different (p≤ 0.05) in all ranches in august was higher in ranch 2 (55.05 cm), followed by ranches 4, 5, 3 and 1 with 37.55, 34.23, 27.98 and 22.33 cm, respectively; while in september ranches 2, 4 and 5, 19.93, 19.05, and 17.72 cm, respectively, had higher ALT in their fodder regarding ranches 1 and 3 with 14.63 and 15.87 cm, respectively. The Figure 2 compares ALT and PSA assessed on ranches throughout the months of the year.

Figure 2 Height (ALT; n= 7200) and cumulative (PSA; n= 720) dry weight of forage during the year in Humid Tropical conditions in southeastern Mexico. 

The interaction ranch*month of the year of PSA was because there was a higher PSA in october fodder ranches 2 and 4 (1.15 and 1.15 t ha-1) when the PSA on ranches 1, 3 and 5 (0.90, 0.77 and 0.79 t MS ha-1) were lower; while in november ranches 2 and 4 (0.93 and 0.97 t MS ha-1) had higher PSA than the rest of theranches;however, the ranch 2 had higher PSA than other ranches in December, January, February, April and May with 0.90, 0.82, 0.52 and 0.42 0.84 t ha-1, respectively; in march interaction was due to PSA on ranches 1 and 3 (0.44 and 0.41 ton MS ha-1) was lower compared to the other ranches; however, the ranch 4 had higher PSA in June and July (0.33 and 0.53 t ha-1) than the other ranches; in august the interaction was because the ranch 1 (1.26 t MS ha-1) had lower PSA than the rest of the ranches, while in September all ranches had different PSA with 0.65, 0.82, 1.41, 1.20 and 0.98 for ranches 1, 2, 3, 4 and 5, respectively. According to the results, we can consider that ALT and PSA were higher in the months of June, July, August and September (rainy season), intermediate in the months of October, November, December and January (windy season) and low in the months of February, March, April and May (dry season). This is because during the months of June, July, August and September temperatures and higher humidity were available.

^{Arteaga (2014)} also reported higher PSA from june to september with a similar pattern of seasonality in forage production in the humid tropics of Puebla. Some studies have positively correlated the level of precipitation height and dry matter production of forage (^{Ramírez et al., 2009}) as tropical grasses depend on the balance between photosynthetic rate and respiration rate of the plant for the accumulation of dry matter (^{Taiz and Zeiger, 2002}).

The ranch*time interaction ALT was due to the difference between the ranches in the rainy season, so the ranch 2 (25.18 cm) had higher ALT compared to ranch 1, 3, 4 and 5 (15.08, 16.89, 20.38 and 19.11 cm), while in the rainy and dry ALT were similar in all ranches. As shown in Figure 3, forages grow more in the rainy season, followed by the windy season, reaching the lowest level in the dry season.

Figure 3 Height (ALT) and cumulative (PSA) Forage five ranches in three seasons in Humid Tropical conditions in southeastern Mexico dry weight. 

The forages in the rainy season (19.33 cm) had 82% higher ALT in the windy season (10.44 cm) and 121% more than the time ALT and dry (6.30 cm), respectively. According to ^{Castillo et al. (2009)} mention that the height of the prairie, has a positive relationship with this fodder, concluding that can be estimated reliably dry matter present from the estimated height before grazing. ^{Hodgson (1990)} mentions that the height of the prairie, coupled forage density determines the amount of forage produced, while the diversity of species determines the quality of the dry matter available.

The ranch*time interaction PSA was because the ranch 2 had higher PSA (p< 0.05) during the windy season compared to prairie ranches 1, 3, 4 and 5 (0.95 vs 0.71, 0.56, 0.89 and 0.75 t ha-1), representing 34, 69, 7, and 26% PSA, respectively. Also in the dry season, the ranch 2 had increased production of dry forage (p< 0.05) compared ranches 1, 3, 4 and 5 (0.61 vs 0.42, 0.40, 0.48 and 0.45 t ha-1, respectively), which it represents 45, 52, 27 and 35%, respectively. Meanwhile, in the rainy season, the ranch 3 had higher PSA (p< 0.05) compared to ranches 1, 3, 4 and 5 (1.27 vs 0.78, 1.04, 0.91, 3.66 and 1.04 t ha-1, respectively) is representing 64, 23, 39 and 22%, respectively. Thus, the differences between ranches in each season in forage production (ALT and PSA) can be attributed to the conditions of precipitation, temperature of the season and pasture management at each ranch. For example, the ranch 2 had higher PSA compared to prairie ranches 1, 3, 4 and 5 (0.87 vs 0.63, 0.74, 0.76, 0.75 t ha-1, respectively), representing PSA at the ranch 2 was 36, 16, 14 and 16%, 1, 3, 4 and 5, respectively, because this ranch has records of all fertilizer to pastures periodically.

On the other hand, the average annual PSA forage was 9.00 t ha-1, which, in the months of the rainy season occurred 45% (4.03 kg ha-1), in the months of the season norths 34% (3.09 kg ha-1) and in the months of the dry season 21% (1.88 kg ha-1). Which means that in the rainy season produced 30 and 115% more PSA than in the windy season and dry, respectively. ^{Arteaga (2014}) reported higher annual PSA study of 9 331 kg ha-1, which, in the rainy season there was 53% (4 975 kg ha-1), in the windy season 26% (2 388 kg ha -1) in the dry season and 21% (1 968 kg ha-1). Although this author reported higher PSA, the seasonal behavior of forage production is similar to this study. According to ^{Villarreal-Castro (1994)}; ^{Velasco et al. (2001)} dry matter increases as age or plant growth progresses, with a higher dry matter production of tropical species when the maximum rainfall is recorded.

Thus, the increased production of dry matter in the rainy season compared to the windy season and dry, it can be attributed to the rains observed in that period allowing greater soil moisture. For the dry season, low yields of MS may be due to drought and heat causing an oxidative stress can lead to inhibition of processes photosynthesis and respiration and thus the plant growth (^{Jagtap and Bhargava, 1995}; ^{Dat et al., 1998}), as the dry season showed higher temperatures and lower rainfall.

The interaction species forager*month of the year in ALT occurs in august when B. humidicola (43.87 cm) presents higher than B. brizantha (33.03 cm), B. hibrida (35.43 cm) and P. notatum (30.84 cm ), while in the remaining months height between grass species they are similar. Figure 4 compares the height and dry matter production of four forage species through the year.

Figure 4 Height (ALT) and cumulative (PSA) of four forage species five ranches throughout the year in Humid Tropical conditions in southeastern Mexico dry weight. 

The interaction species forage*month of the year in PSA was because B. humidicola had higher PSA in october, april and may (1.07, 0.48 and 0.41 t ha-1) than the rest of the species, while P. notatum had lower PSA in december and february (0.65 and 0.50 t ha-1) compared to the rest of forage species; B. hibrida in july had increased PSA (1.26 t ha-1), but in september P. notatum (1.16 t ha-1) had PSA greater than other species. Thus, the height and PSA of the evaluated forages are higher in the months of the dry season (June, July, August and September), followed by the months of the windy season (October, November, December and January), and lower in the months of the dry season (February, March, April and May). According to ^{Juárez (2005)} these production changes in forage production of these grasses show the effect of the medium as a single grass modifies its production going from an environmental condition to another as in the case of the season and fertilization (Figure 4).

On the other hand, the average production of PSA of B. humidicola, B. hibrida, B. humidicola and P. notatum were 0.78, 0.80, 0.71, and 0.72 kg DM ha-1, respectively. These results are lower with those reported by ^{Juárez et al. (2004)} who presents a production of 2.75 t ha-1 for B. humidicola and 1.55 t ha-1 for P. notatum. However, the PSA of B. humidicola was similar to that reported by ^{Peters et al. (2003)} who obtained yields of 0.67 to 0.83 t ha-1 for this species. Regarding the least amount of dry MS produced in ranch 5 (Brachiaria hibrida and Brachiaria brizantha), compared to ranch 2 may be due to nutritional soil conditions as this ranch is no record of previous fertilizations.

Annual growth rate and production of MS

There was effect (p≤ 0.05) of the interactions ranch*month of the year, ranch*season and species*month of the year for TC for forage. The Table 1 compares the TC fodder monthly basis and PAMS five ranches evaluated.

Table 1 Monthly growth rate (kg de MS d-1) and annual production of dry matter (kg ha-1) of fodder in five ranches in southeastern Mexico. 

^abc Medias en la misma fila sin una letra en común son diferentes (p≤ 0.05); *EEM= error estándar de la media; **PAMS= producción anual de materia seca.

The TC was different (p≤ 0.0001) between the months of the year in all the ranches studied except in august when all the ranches have similar TC (p≥ 0.05). Also PAMS was different (p≤ 0.007) between ranches.

As shown in the figure above, the TC older are given in the corresponding months of the rainy season (june to september) where the availability of water for plants, the amount of solar radiation incident on the prairie and ambient temperature are environmental conditions that do not limit the growth of forage (^{McKenzie et al., 1999}; ^{Ramírez et al., 2009}; ^{Cruz et al., 2011}). Most TC and PAMS in ranch 2 in most months, can be justified by fertilization with nitrogen (N) in the grasslands of this ranch to improve productivity of forages, since the application of N improves the rate of growth (^{Quero, 2007}; ^{Améndola et al., 2011}; ^{Lopes et al., 2011}).

The interaction ranch*time of year of TC forage was because the ranch 2 had higher TC (p< 0.05) in the rainy and dry season, compared to the prairies of the other ranches. However, in the rainy season, the ranch 3 had higher forage production (p< 0.05) compared to other ranches (Table 2).

Table 2 Growth rate (kg MS ha-1 d-1) of fodder in three seasons in five ranches in southeastern Mexico. 

^abc Medias en la misma fila sin una letra en común son diferentes (p≤0.05); *EEM= error estándar de la media.

Forages in the rainy season were 30 and 115% more TC than in the windy season and dry, respectively. TC lower in the dry season corresponds to the low availability of water for plants which limits their growth. According to ^{Azcon-Bieto and Talon (2008)}, ^{McKenzie et al. (1999)} and Huerta (1997), the frequency and distribution of rainfall, the amount of radiation incident on the prairies, ambient temperature and mineral nutrition are factors that directly affect the rate of photosynthesis.

The biomass production occurs by transforming chemical energy radiation (NADPH and ATP) used to reduce carbon dioxide to sucrose in the enzymatic process of the Calvin cycle (^{Salisbury and Ross, 1992}). This process occurs in every cycle of growth in a meadow, always regulated by atmospheric conditions, causing changes in the magnitude of the rate of growth by the amount of solar radiation incident on the plot and the environmental temperature at which plants grow (^{McKenzie et al., 1999}). Photosynthesis is strongly influenced by the available moisture, the level of fertility and soil physical properties in the meadow (^{Matthew et al., 2001}; ^{Moliterno, 2002}). The interaction month*forage species was because in january B. brizantha presented higher TC (p≥ 0.05), compared with other species; in April and May was the species B. humidicola more TC; whereas in july B. hibrida showed higher (p≥ 0.05) TC other forage species (Table 3).

Table 3 Growth rate (kg MS ha-1 d-1) and annual production of dry matter (kg ha-1) four forage species in southeastern Mexico. 

^abc Medias en la misma fila sin una letra en común son diferentes (p≤0.05); *EEM= error estándar de la media; **PAMS= Producción anual de materia seca.

As shown in Table 3, a marked seasonality occurs in TC of all species and in the months with the highest rainfall (June, July, August and September) when the fastest growing fodder is presented. Other examples of the seasonality of forage production when the pastures pass the time of minimal rainfall to the time of maximum precipitation is found by Ayala and Basulto (1992), where the Andropogon gayanus increased its production of MS 2.26 to 11.34 t ha-1, in Brachiaria dictyoneura 1.99 to 10.99 t ha-1, in B. humidicola 1.91 to 7.81 t ha- 1, in B. decumbens 1.63 to 5.17 t ha-1, in B. ruziziensis 1.07 to 4.17 t ha-1, in P. maximum of 1.24 to 5.21 t ha-1, and Cynodon nlemfuensis 0.97 to 3.52 t ha-1. According to ^{Juárez (2005)}, these changes in MS production of forage of these grasses show the effect of the environment, because the same grass modifies its production going from an environmental condition to another as in the case of the time the year and fertilization.

In general, Brachiarias ranches 2 and 5 were fewer MS (9 185 kg ha-1) than the 15 182 kg ha-1 reported by ^{Arteaga (2014)}. However, the native grasses in this study had a higher annual MS yield (8 651 kg ha-1) than reported by the same author (7 707 kg ha-1).