Abstract (eng)
Biological corridors are essential for the structural connectivity of natural landscape elements across cultural areas, promoting migratory movements and genetic exchange of various plant and animal species. To assess the state of corridors within the agricultural area of La Gamba, Costa Rica, the land cover and land use, as well as linear landscape elements were investigated. The structure of the landscape was analyzed by means of landscape metrics. Linear elements were examined in terms of their structural parameters, whereas live fences were treated more in detail and characterized by their plant species compositions.
The current land use map clearly shows that pastures were most widespread, covering 61.07% of the agricultural area. Oil palm plantations, which comprised 30.55% of the agricultural area, were the second most important cultivation type. Over the last few years this type of land use had been spreading rapidly, as nearly half of the plantations (48.75%) were younger than three years. On the other hand, the production of rice had decreased rapidly. Rice plantations covered only 1.04% of the agricultural area. Hence, agricultural economy in La Gamba mostly based on cattle breeding and oil palm industry. Primary vegetation comprised 28.97% of the study area and secondary vegetation 35.49%. The intersection of the land use map with a digital elevation model illustrated that cultivations were mostly restricted to plain areas with inclinations lesser than 5°, while steeper slopes were mainly covered by primary forest. Landscape metrics illustrated that forest and rural areas clearly differed in their structure and that the generally big and compact shaped oil palm plantations have significantly influenced landscape structure. Furthermore, these plantations were scarcely bordered by live fences and provided fewer microhabitats than pastures.
The main purpose of live fences in La Gamba is to divide pastures and to restrict cattle movement. Consequently, they were found more frequently in pasture dominated parts of the study area. Most of them were very narrow (mean width 3.71 m) and poorly structured. Compared to natural line elements their ecological value was rather low. 46.57% of the live fences were linked to other connecting line elements, but only 13.70% were linked to forests. Moreover, plant diversity within individual live fences was low. In total 92 plant species were found in live fences, but the mean species number (vascular plants with stem diameter >1 cm within a 50 m long sections of a live fence) was only 9.0. The total species number ranged from two to 19 species per site. 21 (22.83%) species were definitely planted in live fences, most of them were trees. Only five species (5.43%) were neophytes. The most important trees were Erythrina fusca (Fabaceae), Gliricidia sepium (Fabaceae) and Psidium guajava (Myrtaceae). In 24 out of 54 investigated live fences Erythrina fusca (Fabaceae) was the dominant species. Based on their species composition, live fences were classified into five groups by means of cluster analysis. Two groups were characterized by Erythrina fusca (Fabaceae), one by Gliricidia sepium (Fabaceae) and another by Tectona grandis (Lamiaceae). The fifth group consisted of live fences dominated by several species.
In order to map and visualize potential corridor routes between the forest areas, least cost paths through the farmland were computed. In those landscape parts which are poor in live fences, additional corridor elements would have to be established to make the agricultural area pervious for wildlife. Moreover, existing live fences would need to be enlarged, broadend and stocked with more ecological valuable species to improve their function as wildlife corridors. The general objective of this work is to provide an inventory of land use and live fences of the study area and to make proposals for improving the present situation.