The effects of Chromolaena odorata on tree growth dynamics at Buffelsdraai Landfill site.

Masters Degree. University of KwaZulu-Natal, Pietermaritzburg.

Ecological restoration is a process which assists the recovery of an ecosystem which was previously disturbed or degraded. Through continuous disturbances, invasive alien plants (IAPs) are able to successfully spread and establish themselves while reducing the diversity and the abundance of native plants. The IAPs invade both human and non-human modified landscapes thus causing huge threats to biodiversity conservation and ecosystem services provision. The invasion of alien plants at Buffelsdraai Landfill Site Community Reforestation Programme is a major threat to reforestation success. Buffelsdraai is under rehabilitation through a community reforestation programme that focuses on active involvement of nearby communities in propagating seedlings and planting them on site. The study was conducted in a former sugarcane field which has been planted with diverse native tree species, but invaded by dense stands of Chromolaena odorata. This study aimed to 1) determine the influence of a woody shrub C. odorata on the growth dynamics of replanted native trees at the Buffelsdraai Landfill Site, and 2) to assess the efficacy of cutting height and frequency in mechanical control of C. odorata. Over a period of 12 months, the growth responses of three native tree species (i.e., Vachellia natalitia, Brachylaena discolor and Erythrina lysistemon) were investigated by measuring stem length and diameter, trunk basal diameter, tree height and canopy diameter, under four treatments. The treatments consisted of: (i) a control, where no clearing of C. odorata or grasses was undertaken beneath the canopy of trees; (ii) clearing of C. odorata from underneath trees; (iii) clearing of grass from underneath trees; and (iv) clearing of both grass and C. odorata from trees. Measurements were undertaken on 264 trees, 88 individual trees per species, 22 individuals per species per treatment at the beginning and the end of the study. Stem, trunk and canopy length, diameter, and tree height growth data within each tree species were analysed using generalized linear models. Tree height growth comparisons among the tree species was analyzed using a non-parametric Kruskal-Wallis test since the data did not meet the assumptions of a One-way ANOVA. There were no significant differences between all four treatments across the tree species (P > 0.05). However, E. lysistemon growing where there was grass removal had significantly higher (P = 0.033) trunk basal diameter growth, than other treatments. Results from this study showed that the removal of C. odorata at the Buffelsdraai community project had no impact on the trees due to short time period of the study. Overtime we expect that tree performamce would increase with less C. odorata competition. Trees performed equally well in the presence and absence of grass and C. odorata. Young tree saplings of pioneer species were used for this study which could be a possible influence of these results. To investigate the impact of C. odorata mechanical control, 28 plots (5 m × 5 m) were established, at least 2 m apart, with one designated as the control with no cutting of the plants. The two treatment combinations were between cutting frequencies which had: (1) cutting frequency made up of single, twice and thrice cutting; and (2) stem cutting height which had 30, 100 and 200 mm cutting heights aboveground. I used 420 C. odorata plants (i.e. 3 cutting heights × 9 replicate plots × 15 plants + 15 control plants). Plants were first cut in November 2016, the coppice regrowth was cut on plants in 18 plots in February 2017, and a third cut was made on plants in the last nine plots in May 2017. Stump basal diameter, the number, length and diameter of resprouting shoots were measured at 3-month intervals. Mean number of resprouting shoots, mean shoot length and diameter of the resprouts, mean shoot length:diameter ratio (shoot taper function) of resprouts, shoot production (shoots mm-1) and mean total shoot basal area:stump basal area ratio for all treatment combinations were analysed using the Friedman’s test. The relationship between cut stem diameter and the number of resprouting shoots, resprouting shoot length, and resprouting shoot diameter were explored using correlation analysis. There was a significant difference in the number of resprouts produced across all treatment combinations (P < 0.001). The 30 mm cutting height produced the least number of resprouts with increasing cutting cycles compared to the 100 and 200 mm cutting height. Shoot length and diameter decreased with cutting frequency, with smaller shoots produced with continued cutting cycle. There were significant difference in the shoot length and diameter of shoots produced in the 30 mm cutting height in the last cutting cycle and those that were produced on 200 mm cutting height. Shoot taper function and shoot production showed significance (P < 0.001), with smaller number of shoots produced in the 30 mm cutting height over the cutting frequencies (P < 0.001). Total shoot basal area:stump basal area ratio did not differ significantly across the cutting heights and the cutting frequencies. There was a strong positive relationship (r = 0.91; P < 0.001) between the number of resprouts and stem diameter across all repeatedly cut stems in all cutting heights and the number of resprouts produced in each cutting height decreased over each cutting frequency. Repeated cutting of C. odorata at a lower cutting height of 30 mm may deplete its energy reserves, reducing the number of resprouts produced, thereby leading to death of the plants. Repeated cuttings at short-term intervals also prevents the plants from growing to reproductive maturity and seed production, leading to no seed dispersal to increase infestations