Abstract

A current concern exists in Costa Rica as to what the best way may be to return pastures to native habitat. We decided to look at three plots already in existence to determine what the effectiveness of differing treatments are in controlling African Star Grass, an introduced species that was brought into the country for cattle to graze. The plots that we looked at included a plot which is burned yearly, one that is grazed regularly, and one that has no grazing and no burning. We used a quadrat method in the burned plot, a combination transect/quadrat method in the control plot, and due to time constraints only had time to collect species in the grazed plot. The three plots were found to be quite different in terms of species present. We determined that burning is the least effective method of eliminating the star grass since the burned plot was found to contain 86% grass cover. Grazing appears to create enough of a disturbance to allow acacia growth to flourish as compared to the control plot but does appear to effectively control the star grass.

Introduction

El Parque Nacional Palo Verde (The Palo Verde National Park) and La Reserva Biologia de Loma Bardubal (the Biological Reserve of Bandubal) in Bargaces, Guanacaste are located in one of the driest ecosystems of Costa Rica. In this region an extensive history of cattle ranching exists which spans over four hundred years. During this period fire was the “tool” of choice to clear the agricultural waste at the end of each growing season. The practice of burning the plant material as a means of disposing waste ended in 1977 when Palo Verde became a wildlife refuge. Due to its nature as a deciduous tropical forest both Palo Verde and Loma Bardubal generate a great deal of leaf litter which form a thick layer of ground cover. This material serves as an excellent fuel for forest fires (Acuna-Perez, 1995). At the end of each growing season the cattle ranchers and other agricultural businesses surrounding Palo Verde burned the plant material left in the field. Often fires would go out of control, resulting in hundreds of hectacres of tropical forest vegetation and wildlife being destroyed. For instance, in 1994 a fierce fire destroyed a large area of Laguna Bocana. Along with the vegetation, many native species of fauna, such as turtles, were destroyed.

The proliferation of the exotic species African Star Grass (Jaragua), Hypharrenia rufa, into the tropical forest has added fuel to an already serious problem. Due to its invasive nature and combustibility, H. rufa acts as a catalyst for fire both in Palo Verde and Loma Barduba.

Fire prevention practices of the past had proven to be ineffective in slowing down destruction of the natural resources and wildlife due to fires. A pilot program was conducted from 1991 to 1995 to determine which method was more effective in controlling, if not eliminating, the spread of H. rufa and the high incidence of fires in Palo Verde and Loma Bardubal. The research design for this study involved the use of grazing cattle, as a method of biological control, and controlled burning to determine the effectiveness of controlling African Star Grass.

Materials and Methods

Three pre-arranged 20m x 20m plots were selected for our study. Each plot was measured for its length and width with measuring tape to determine a more exact area. The plant species count for each plot was determined by walking in and carefully scanning the ground. A sample of each different species was collected in a small ziplock bag and labeled. These samples were brought back to the lab for further analysis and identification.

The relative abundance (population density) of each plant species was determined for plots 1 and 2. In both cases a square (1.0m x 1.0m) quadrat made of steel with markings every 10cm was used.

Plot 1 was visibly abundant in grass (greater than 70%), with limited canopy cover. Six microhabitats were visible in the plot, including the grass habitat. For this reason, the microhabitat areas were measured, and the quadrat method was used to determine the population density of each species in the plot.

Plot 2 was visibly limited in ground growth, but had a relatively full canopy. Unlike Plot 1, a limited number of microhabitats could not be determined. For this reason the transect and quadrat method was used to determine the population density of the plants species there. A measuring tape was laid out from opposite corners of the plot. The quadrat was laid along the measuring tape starting at 0m, every ten meters. Since the distance was 30m from corner to opposite corner, four quadrats were taken for each transect, and a total of eight quadrats were analyzed in this plot.

Results

Plot 1: Burned yearly, no grazing

Area of plot: 391 m²

Microhabitat A: 22 m²

Grass	65.5%
Composite	33.5%
Mimosa	< 1%
Sedge	< 1%
Unknown Vine Z Fabaceae	< 1%
Open ground cover	1%

Microhabitat B: 1.25 m²

Grass	83%
Mint	5%
Composite	3%
Mimosa	1%
Nicotina	1%
Unknown X	1%
Open ground cover	6%

Microhabitat C: 5.4 m²

Grass	58%
Unknown W	35%
Passion flower	5%
Composite	2%
Sedge	< 1%
Spurge	< 1%
Unknown Z Fabaceae	< 1%
Mimosa	< 1%
Open ground cover	1%

Microhabitat D: 4.69 m²

Grass	66%
Sapindaceae	8%
Mimosa	< 1%
Composite	< 1%
Species Y Solanaceae	< 1%
Open ground cover	26%

Microhabitat E: 2.34 m²

Grass	92%
Fabaceae	6%
Species Z Fabaceae	< 1%
Open ground cover	2%

Grass Habitat area: 355 m²

	Trial 1	Trial 2	Trial 3	AVERAGE
Grass	90%	90%	83%	88%
Mimosa	2%	1%	1%	1.3%
Unknown Z Fabaceae	< 1%	< 1%	< 1%	< 1%
Sedge	< 1%	0	0	0
Open ground cover	7%	9%	16%	11%

Total Abundance of plant species in plot 1:

Grass	86.0%
Open ground cover	10.4%
Composite	1.9%
Mimosa	1.2%
All other plant species	< 1%

Plot 2: Unburned, no grazing (control)

Area of plot: 472 m²

Transect 1, Quadrat 1

2b	25%
2a	< 1%
2c	< 1%
2d	< 1%
2e	< 1%
2f	< 1%
Open ground cover	75%

Transect 1, Quadrat 2

Mimosa	< 1%
2a	< 1%
2g	< 1%
2h	< 1%
2i	< 1%
2j	< 1%
Open ground cover	99%

Transect 1, Quadrat 3

Plot 3: Unburned, low density grazing until 1995

Area:

Discussion