Custom services
text in 
English (pdf)
Article in xml format
Article references
Send this article by e-mail
Cited by SciELO 
Similars in
SciELO 
Permalink
Introduction
The ecological intensification of sustainable agriculture, according to Tittonell (2014) and Vázquez and Funes (2014), demands technological changes and the inclusion of new knowledge, allowing to reach a balance between resource preservation and agricultural production.
The botanical composition of the grassland and its yield are attributes of vital importance to measure the “health” of the ecosystem. Senra et al. (2005) considered them as fundamental indicators in the evaluation of sustainability of cattle grazing systems.
Silvopastoral systems become an agricultural option that integrates an adequate tree density in pastures and implies the presence of trees in interaction with other components of the ecosystem, subject to an integrated management, which tends to increase its long-term productivity (Campo 2006).
In Carrizal-Chone, Ecuador, the spontaneous presence of Prosopis juliflora tree is common in grassland areas (Basurto 2016 and Roca 2017). On the other hand, farmers of this area often consider the existence of trees in the pastures as negative for their yield and botanical composition, which can be real when proper system management is not carried out.
The objective of this study was to determine changes in the botanical composition of grassland, according to the amount of Prosopis juliflora trees in an area subject to grazing dairy cattle in Carrizal-Chone region, Ecuador.
Materials and Methods
The experiment was carried out in a grassland area divided into paddocks, belonging to the teaching, research and linkage units of the Agricultural Polytechnic School of Manabí "Manuel Félix López" (ESPAM-MFL, in Spanish), located in the geographical site "El Limón", Calceta, Manabí province, Republic of Ecuador. These facilities are part of the cattle rearing systems of Carrizal-Chone rural development region. They are located at 00 ° 49’23” South and 80 ° 11’01” West, at 15 m.o.s.l.
The ecosystem is low pre-montane tropical forest (Regalado et al. 2012), with humid tropical climate and differentiated fluvisol soil. Table 1 shows climatological variables of the experimental period (2012 - 2014).
Table 1 Climatological variables of the area under study during the experimental period
| Variable | Experimental year | ||
|---|---|---|---|
| 2012 | 2013 | 2014 | |
| Mean temperature, °C | 25.0 | 25.0 | 26.0 |
| Mean relative humidity, % | 82.9 | 83.8 | 82.0 |
| Precipitation, mm | 1638.9 | 962.4 | 777.3 |
Out of the spontaneous tree vegetation of the grassland area, only Prosopis juliflora (Sw.) DC species trees were preserved, which had a height superior to 6 m and a canopy diameter greater than 4 m. Tree areas were thinned up to the desired densities.
Treatments and experimental design. Treatments consisted on three amounts of Prosopis juliflora trees: low (1-4 trees/ha); medium (5-8 trees/ha) and high (9-12 trees/ha), distributed in a completely randomized design.
Experimental procedure. There were 9 paddocks/treatment, 0.45 ha each, consisting of star grass cv. African (Cynodon nlemfuensis) and naturalized saboya (Megathyrsus maximus = Panicum maximum) with associated creeping legumes, of Centrosema (C. molle and C. acutifolium), Desmodium (D. incanum and D. scorpiurus), Macroptilium (M. atropurpureum) and Teramnus (T. labialis) genera.
Sprinkler irrigation was applied at a rate of 1,200 m3/ha every 28 d. Grazing areas were fertilized with 50 kg of N/ha/year for the three evaluated levels. The grazing technique was rotational, with an occupation time of two days. Grazing intensity of the paddocks under evaluation was 111 LAU/ha/rotation.
An amount of 25 Brown Swiss x Zebu cows, with 450 kg of liveweight were used, with a lactation range between 90 and 120 d. After grazing, the animals received chopped forage of king grass (Cenchrus purpureus) at a rate of 10 kg/cow/d, in the feeder inside the unit.
Measurements of botanical composition of the grassland were made by the dry weight rank method of t’Mannetje and Haydock (1963).
The botanical composition of the grassland was measured in each treatment, at the beginning (2012) and end (2014) of the experiment. Sampling was performed in five randomly selected paddocks in each treatment, which were maintained throughout the study. Table 2 shows the values of initial botanical composition of experimental paddocks.
Table 2 Initial botanical composition of the grassland (%) in the experimental paddocks
| Species | Tree amounts (trees/ha) | ||
|---|---|---|---|
| 1 - 4 | 5 - 8 | 9 - 12 | |
| Star grass (Cynodon nlemfuensis) | 50.3 | 27.5 | 19.3 |
| Saboya grass (Megathyrsus maximus) | 18.5 | 37.8 | 50.5 |
| Other grasses | 13.9 | 6.7 | 9.9 |
| Creeping legumes | 14.0 | 23.0 | 13.0 |
| Weed (broadleaf plants) | 3.3 | 5.0 | 7.3 |
The cover of species was assumed as a result of determining the botanical composition. Changes in cover were determined by the difference between final and initial botanical composition.
Statistical analysis. For the statistical data processing, SPSS package, version 15 (Visauta 1998) was used. Data normality was analyzed by Kolmogorov-Smirnov test and homogeneity of variances by Bartlett test. A covariance analysis (ANACOVA) was performed for the final botanical composition of the grassland, according to the mathematical model of the design used, in which the initial botanical composition was considered as covariate. The effect of tree amounts (3) was also controlled.
After maintaining the five species or group of initial species throughout the experiment, it was decided to consider species (5) as a factor and its interaction (3x5) with tree amount, so to establish a possible comparison between them.
To the data of the difference between the final botanical composition with the initial one (coverage changes), an analysis of variance was performed, in which the effects tree amount (3), species (5) and their interaction (3x5) were controlled. Differences among means were determined by the Tukey test.
Results and Discussion
Results of the covariance analysis, performed at the final botanical composition variable, when the effect of the initial botanical composition was considered as covariate (P <0.001), showed highly significant effects (P <0.001) of the interaction tree amount (3) x species (5) and of isolated factors.
Table 3 shows the results of the interaction tree amount x species in the final botanical composition of the grassland. The treatment that showed the highest population percentage (45.7%) was saboya grass (M. maximus) at the high tree amount. This grass is considered well adapted to natural intermediate shading, such as that produced by P. juliflora tree.
Table 3 Combined effect of the amount of P. juliflora trees and the species in the final botanical composition of grassland (%)
| Species | Tree amount (trees/ha) | SE ± | ||
|---|---|---|---|---|
| 1 - 4 | 5 - 8 | 9 - 12 | ||
| Star grass (Cynodon nlemfuensis) | 42.5 (0.64b) | 27.3 (0.52d) | 21.8 (0.48f) | 0.003*** |
| Saboya grass (Megathyrsus maximus) | 21.0 (0.47f) | 36.0 (0.59c) | 45.7 (0.67a) | |
| Other grasses | 11.0 (0.33h) | 6.8 (0.25k) | 8.1 (0.28i) | |
| Creeping legumes | 16.2 (0.41g) | 23.9 (0.49e) | 17.5 (0.43g) | |
| Weeds (broadleaf plants) | 9.3 (0.32j) | 6.1 (0.21l) | 6.9 (0.25k) | |
abcdefghijklDifferent superscripts indicate significant differences for P˂0.001, according to Tukey test. *** P˂0.001
Means fit to initial botanical composition by covariance
( ) Values transformed according to arcsen √(x/100)
The response obtained with saboya grass coincides with reports of Pentón (1999), who evaluated the evolution of botanical composition in a silvopastoral system in Cuba. This author found that P. maximum (today Megathyrsus maximus) showed the highest associative potential with trees, and with the shade, it was a controlling factor of plants with low nutritional value, such as naturalized grasses and weeds.
An element that should be considered in these results as favorable to saboya grass is its high competitiveness. According to Pentón (2002), this plant can be stimulated under woodland conditions, based on its ability to gain preferential access to the incident radiation through its height. In this regard, Ericksen and Whitney (1981) pointed out that forage species differed markedly in their response to shade and that P. maximum is one of the cultivated grasses with the best performance.
When the analysis is performed to the high tree amount, saboya and star grass predominated in the grassland, to the detriment of other grasses and broadleaf plants. Shelton (1996), Sánchez et al. (1997) and Ruiz et al. (2008) reported beneficial effects of leucaena and other tree shading on pastures and forages.
The result obtained with the creeping legumes, which showed a growing change with the increase of tree amount, identifies them as well adapted to the natural intermediate shade. This performance has been indicated in other tropical scenarios and in association with Brachiaria, saboya and star grass, depending on shading intensity, area rest, defoliation intensity and recovery of their reserves (Guevara 1999 and Iglesias et al. 2003).
Referring to the importance of grassland botanical composition, Ray et al. (2016) pointed out that the availability and floristic composition allow to maintain a proper intake of the offered base grass, and stressed that the presence of legumes is manifested as a basic element of grazing system. Gómez (2015) corroborated the role of legumes in the entrance of nitrogen into the grazing system.
Previously, Wright et al. (2015) reported that depending only on the quality of grazing pastures, intake, digestibility and animal response can be affected.
The magnitude of changes in the coverage of species subjected to the tree amount studied is presented in figure 1.
With respect to the initial coverage of 2012, the largest increases in 2014 occurred in saboya grass, subject to levels from 9 to 12 trees/ha (5.1%) and from 4 to 8 trees/ha (4%), without differences between them. This same species, at the low tree amount (1-4 trees/ha), achieved an increase in its coverage (2%) statistically similar to that reached by star grass (2 to 2.3%) in the three studied tree amount.
In relation to the performance of creeping legumes, it is important to state that the increase of tree amount positively influenced on the increase of their coverage, which was superior in the medium (5-8) and high (9-12) levels, with respect to low level (1-4 trees/ha). This could be due to the fact that the shadow cast by trees can bring about important and favorable changes in the environmental factors and in the attributes of the plants that grow under that shade, which is a sign that, apparently, these plants have good adaptation to the environment generated at these tree amounts.
Pentón (2000) said that shade conditions variations in quantity and quality of sunlight from the reduction of photosynthetically active radiation, causes a decrease of the temperature of air, soil and plant leaves. In addition, it promotes the increment of relative humidity of air and soil, and increases water potential and plant productivity.
This study shows that all the species indicated in the literature with the greatest contribution to animal diet (saboya and star grass, and creeping legumes) increased their coverage as tree amount increased. In contrast, the other grasses and broadleaf plants significantly reduced their coverage by statistically similar magnitudes (between 3.8 and 5.1%) at different tree amounts, with the exception of broadleaf plants at the level of 1 to 4 trees / ha, which increased by 0.6%.
This indicates that, with the presence of 9 to 12 trees/ha in grasslands, the appearance of these species with low nutritional value is better controlled. With these trees, according to Muschler (2000), light reduction levels can be marked, as a result of size and arrangement of leaves and branches, which affect lower vegetation. Particularly in this study, the involvement was marked for broadleaf plants and other grass.
It is concluded that tree amount with 9 to 12 trees/ha of Prosopis juliflora (Sw.) DC. improves botanical composition of grassland, by increasing the coverage of the most important species in cattle rearing systems of Carrizal-Chone, Manabí, Ecuador. The study of higher densities of tree plants under these conditions is suggested.
