Talk to be held at the workshop on image analysis and spatial statistics in forestry on November 2, 1999 at KVL, Frederiksberg, Denmark
Wijesekara Vithanage P.R.
(renuka.wijesekara@isf.nlh.no)
Dept.of Forest Sciences, Agricultural University of Norway.
Avian populations were studied in some selected land-use systems invicinity of Delwala (DEL) and Kudumeriya (KDU) proposed forest reserves in Ratnapura district of Sri Lanka to assess their sustainability. The systems studied were Forest reserves, Pinus plantations, Tea monoculture plantations, Tea mixedplantations, Rubber plantations, Cinnamon plantations and Home gardens
The method used for collect data on bird communities in different selected land use systems was a simplified version of the variable circular plot technique (Reynolds et al. 1980) as described by Bibby et al. 1992.
In order to obtain floral diversity and density in different land use systems, systematic quadrate sampling was carried out. Five squares of 20m * 20m in size were established in each type of selected land use system with the bird counting point essentially in the centre. All plants from one meter level to canopy level were identified and counted separately in all habitats. Canopy cover estimation was carried out by using photographs. Camera with a zoom lens was focused up from ground level to infinity through the canopy to take these photographs. Five photographs for each bird counting point were taken. (one from the centre and four from the periphery of the 50m radius circle)
The availability of floral data for the various land use systems in conjunction with the large quantity of bird data makes it possible to examine inter-habitat abundance patterns of trees and birds. Vegetation data are also important to find out the reasons for differences in bird community in variety of land use systems.
One way ANOVA analysis and Two sample T-Test were applied in order to test for significant differences in bird species diversity in different land use systems. Correlation analysis and simple linear regression analysis was used to understand the pattern of relationship among plant species diversity, canopy cover and plant density vs avian species diversity in all studied land use systems.
A total of 73 bird species representing 31 families including 16 endemic species and 7 rare species were identified from DEL site. Eighty-three bird species representing 32 families including 17 endemic species and 14 rare species were recorded from KDU site. Majority of bird species recorded from both study sites are distributed in all zones of the country. Twelve species recorded from DEL site including 8 endemics and 18 bird species recorded from KDU site including 9 endemics are distributed only in Wet Zone. Out of total recorded bird species, 55 from DEL site and 66 from KDU site were forest birds.
Diversity gradient in avian communities in different land-use systems indicates that the changes in vegetation structure had strong influence on bird community structure. Diverse agro-ecosystems such as home gardens and Tea mixed cultures supported a higher bird diversity with higher endemism and more rare species and abundance of forest birds were also higher in these diverse agro-ecosystems. Bird species diversity was significantly lower in Tea monoculture plantations. Bird species abundance were significantly higher in Rubber and Cinnamon plantations. Bird population densities were very low in Pinus plantation because most of the birds were recorded when they crossed the system to use village habitats.
Tea mixed cultures and home gardens are the most sustainable agro-ecosystems. It is possible to improve Tea monocultures, Rubber and Cinnamon plantations to get high biodiversity with more economic benefits by changing vegetation structure and growing pattern.