PHENOLOGICAL PATTERNS OF WOODY VEGETATION AT TINIGUA PARK, COLOMBIA: METHODOLOGICAL COMPARISONS WITH EMPHASIS ON FRUIT PRODUCTION

Abstract

Phenological patterns have a great influence on community ecology. I have quantified fruit abundance using fruit traps, phenological transects, and observations on marked trees of selected species. All methods showed the main peaks of fleshy fruit production between the dry period and the middle of the rainy season. Wind, water, and mechanically dispersed seeds showed a peak of production at the dry period. According to the number of individuals on transects, flower abundance showed predominantly two peaks, one at the end of the rainy period and one in the dry period. This pattern was positively correlated with solar irradiance estimates and differences between maximum and minimum temperatures, which tend to occur on dry days when the activity and abundance of pollinating insects seemed to be high. There was also a good correlation between relative day length and the density of flowering individuals, which suggest that this variable could trigger flowering patterns. Although many species lose their leaves in the dry period, the production of new leaves does not seem to be associated with any particular season. I used a new method to quantify fruit production for each species (in kg/ha), based on phenological transects, corrected visual estimates, and dry fruit weights. The mature forest produced more fruits than the other forests, and the flooded forest showed less fruit production during the dry season and the beginning of the rainy period. In contrast, average density of fruiting trees was similar or slightly greater in the flooded forests than in terra-firme forests during the middle part of the rainy period. Finally, I estimated how long the transects should be in order to get adequate estimates of fruit production, and I found that in general at least 7 km are necessary to get good estimates for terra firme forests at Tinigua.