Abstract

Hardwick, K. (1999)

Tree Colonization of Abandoned Agricultural Clearings in Seasonal Tropical Montane Forest in Northern Thailand

In 1994 the Thai Government embarked upon a nationwide project to restore degraded forests. One approach to such an endeavour is to assist natural regeneration (ANR) by managing a site to counteract particular limiting factors, such as poor dispersal of tree seeds into the site or excessive competition from weeds. This thesis describes a two year project in northern Thailand which aimed to answer the following questions: (1) Can tree species be grouped according to the stages of their life-cycle that block or inhibit their colonization of weedy abandoned agricultural clearings? (2) If so, what traits characterise these groups? (3) What environmental factors block or inhibit colonization at each stage? (4) What further research is needed to develop ANR strategies to remove or overcome these limiting factors?

This study combined observation and experiment. Fruit production, seed dispersal, seedling recruitment and seedling survival of a wide range of tree species were monitored in weedy abandoned agricultural clearings. Controlled experiments were carried out on selected species to test whether seed predation, exposure to full sunlight and seasonal drought inhibited seed germination, to compare seedling establishment in a clearing and forest and to test whether weed cutting inhibited or facilitated seedling establishment in the first year.

Tree species were divided into three seed-size based groups, each characterized by different critical stages (where colonization was likely to be blocked) and inhibiting stages (where the probability of colonization was much reduced). The critical stage for small-seeded species (with seeds < 2mm long) was recruitment: seeds were dispersed prolifically to the clearings but all failed to develop into seedlings. Colonization may be restricted to the period immediately after disturbance, before competing herbs and shrubs take over, and to isolated patches of bare soil within shrubby areas. Colonization of weedy clearings was largely restricted to medium-seeded species (with seeds 2 to 14 mm long), although it was heavily dependent on the presence of fruiting trees at the clearing edge. Recruitment was an inhibiting stage for medium-seeded wind dispersed species, while dispersal and recruitment were inhibiting stages for the animal dispersed species. For large-seeded species (with seeds over 14mm long) the critical stage was dispersal: no large-seeded species were recorded in the seed rain and evidence of secondary dispersal by small mammals was mostly restricted to the edges of the clearings. The fruit production stage also inhibited colonization as over half the large-seeded species studied fruited supra-annually.

The clearing had higher air temperatures and drier soils to 30 cm depth than the forest, but levels of direct solar radiation at ground level were not higher. In experiments, seed predation and exposure of seeds to direct solar radiation reduced germination of most species. The weed canopy generally inhibited seedling survival and growth during the rainy season and facilitated survival during the hot dry season. Species' responses to these factors were not related to seed size.