Effects of climate and management on biomass, species composition and phenology and their regenerative responses in a mesic grassland.

Abstract

Increasing temperatures, attributed mainly to increased greenhouse gas emissions, are a leading cause of climate change. Evidence shows that changing climatic conditions significantly affect terrestrial ecosystem structure and function. Fire, grazing, and recurrent drought are some of the more common disturbances that impact many grasslands. Two ongoing experiments (Ukulinga Grassland Fire Experiment (UGFE) and Drought-Net experiment at the Ukulinga Research Farm, University of KwaZulu-Natal, Pietermaritzburg, KwaZulu-Natal) were used to investigate the impacts of these disturbances on grassland community dynamics. This research assessed the effects of rainfall variation and warming and management factors (defoliation and fire) on phenology of grass species vis-a-vis their regenerative responses from bud banks, soil seed bank, and seed rain and aboveground species composition and biomass in the mesic grasslands of South Africa over two years. The Drought-Net experiment treatments include the combined effect of rainfall variation (drought (50%), ambient (100%), and wet (150%)) and temperature (warmed and ambient) manipulation. The warming treatment was applied using open-topped warming chambers, and the rainfall treatment was applied using rainout shelters to intercept 50% of the ambient rainfall. The intercepted rainfall was piped onto the neighboring plot to increase rainfall by 50%. In the first objective (Chapter 3), the combined effect of rainfall variation and temperature manipulation on the phenology of grass species vis-a-vis their regenerative responses from bud banks was conducted. Drought delayed the timing of budding and flowering and shortened the duration. There was no effect on the total number of buds per vegetative tiller per tuft compared to the reproductive tillers. Warming advanced the budding timing and extended the budding duration. This shows that the response of timing and duration of budding and flowering of these studied grass species was sensitive to the short-term effects of drought and warming. In the second objective (Chapter 4), the effect of burning frequency, winter defoliation type (fire/mow), summer defoliation with differing winter defoliation (defoliation time) applied over ~70 years and temperature manipulation on the phenology of grass species vis-à-vis their regenerative responses from bud banks was conducted. Across the two years of warming treatment application, the sensitivity and timing of species phenological responses and bud bank density relative to resource allocation to vegetative versus sexual reproduction were examined. Warming delayed the time of budding and flowering and increased duration slightly (p < 0.005). Annual winter mow delayed timing and extended duration of budding and flowering as compared to annual winter fire (p < 0.005). Summer defoliation frequency quickened the timing and shortened the duration of budding and flowering compared to winter defoliation frequency (p < 0.005). The flowering duration for Themeda triandra was shortened under annual burning and lengthened under the triennial burn plots (p < 0.005). In ambient plots, vegetative status had more buds per tiller and number of tillers per tuft compared to warmed plots. Annual winter defoliation by fire had the greatest number of tillers per tuft for Heteropogon contortus and the lowest recorded in the annual winter defoliation by mow plots. A greater number of buds per tiller was recorded under the summer defoliation plots as against the winter defoliation plots. Summer defoliation plots had a greater number of tillers per tuft for H. contortus than the winter defoliation plots. The results of this study indicate that the type, time, and frequency of defoliation and their interaction with the effect of warming have a notable influence on the budding and flowering periods of the species studied, ultimately impacting the length of their reproductive cycle and altering the ratio of vegetative to flowering reproductive status. The reason for the contrasting results where warming advanced the time of budding in chapter 3 and delayed in chapter 4 could be as a result of the insurance hypothesis whereby the variation of species responses to changes in environmental condition of an ecosystem can insure the system against reduction in its functioning. The third objective (Chapter 5), combined effect of management factors (defoliation and fire), and temperature manipulation on the species composition, soil seed bank, and seed rain in the Drought-Net experiment and UGFE. Plant community composition was examined to determine sensitivity to climate and management. On the Drought-Net experiment, the main and interacting effects of warming and the various rainfall regimes did not impact the soil seed bank and seed rain (p > 0.05). On the UGFE, the main effect of warming and its interactions with the burning and defoliation treatments had no significant effect on soil seed bank and seed rain (p > 0.05). The seed rain composition was only affected by defoliation frequency, while the main effects of fire and defoliation frequencies affected the soil seed bank (p < 0.05). The result of the study showed that the responses of the seedlings that germinated from the seed bank and seed rain were species-specific towards the effects of the management factors. The lack of effect of drought and warming shows that mesic grasslands are resilient to fluctuations in moisture and temperature regimes in the short term. The fourth objective (Chapter 6), examined the effect of different rainfall regimes and temperature manipulation on the species composition and biomass of a mesic grassland in the Drought-Net experiment. The results showed that the main effect of rainfall variation and their interactions with warming had an impact on the grass species composition, with just the main effect of warming being observed in the following year. No impact on the forb composition was seen in either year. While there was no evidence of an interaction impact, the main effects of rainfall variation and warming on biomass production were significant. Drought and warming reduced biomass production. This demonstrated how quickly drought reduced the grass community composition, which resulted in changes in productivity. Objective five (Chapter 7), this study examined combined effect of warming and management factors (defoliation and fire) on the aboveground species composition and biomass at UGFE. Plant community composition was examined to determine sensitivity to the impact of long-term burning and defoliation frequency under ambient and warmed conditions and whether shifts in plant community composition resulted in changes in productivity. The study showed that the main effect of warming, burning, and defoliation regimes resulted in a shift in plant community composition, which altered productivity. The overall conclusion of this study is that under different defoliation frequencies, types and timing, the species-specific seasonal patterns of species composition, phenological development and bud bank production were strongly influenced by the resultant effect of these disturbances on the availability of soil moisture and light intensity. This resulted in a reduced productivity of the grass community in our mesic grassland.