On the Co-adaptive Response of Water and Carbon Cycles

Abstract

Extreme weather events including droughts, heat waves, cold snaps, fires and large storms have the potential to interrupt normal plant development and inhibit healthy plant function during the growing season. Long-term changes in the intensity and inter-arrival times of extreme events (i.e. drought, tropical cyclones, and wildfires) have the ability to alter the regional hydroclimate of areas where incoming rainfall follows prescribed patterns. In the Southeast U.S., much of the warm season precipitation is delivered by landfalling tropical cyclones which have high inter-annual variability in number, intensity, and rainfall amounts. Conversely, in Southern Africa rainfall distributions are controlled by the interactions among large climate boundaries that dictate the timing and location of wet season rainfall. However, the impacts of growing season disturbances on carbon uptake rates by vegetation extend to drought and fire conditions that limit plant growth and leaf development; thus, restricting the maximum carbon uptake potential under favorable atmospheric and soil conditions. As such, changes in the plant life cycle, or phenology, as a result of meteorological disturbances must also be considered in a thorough investigation of water limitations on carbon assimilation rates. The overarching objective of the proposed research is to investigate the inter-cycle sensitivity of carbon and water fluxes between the land-surface and atmosphere in two distinct ecosystems by modelling how changes in the water cycle impact spatial and temporal variability in carbon assimilation rates, soil moisture, evapotranspiration, and plant phenology, specifically in the context of variability in the spatial and temporal delivery of precipitation. The research objectives are: 1) Investigate how spatial and temporal changes in precipitation alter carbon uptake by vegetation and plant phenology; and 2) Elucidate ecosystem recovery dynamics in terms of adapting energy and water budgets after extreme disturbance events (i.e. drought, tropical cyclones, fires) in the Southeast U.S. and Southern Africa, representative respectively of humid extratropical mid-latitudes and semi-arid tropical climate regions in the Atlantic basin; and 3) Quantify the impacts of large disturbances of regional and local carbon and water budgets. The Duke Coupled Hydrology Model with Vegetation (DCHM-V) is used to investigate feedbacks and inter-cycle interactions between the carbon, energy, and water cycles in the Southeast U.S. and Southern Africa. A predictive Dynamic Canopy Biophysical Properties (DCBP) model is developed and coupled to the DCHM (i.e. DCHM-PV) to dynamically estimate changes in canopy structure and development under water-limiting conditions. Findings from the Southeast U.S. show that precipitation provided by tropical cyclones can increase plant carbon uptake by 4-8% in the Southeast U.S. over the course of a drought year. Further, soil hydraulic properties alone explain most of the variability in warm season water stress in the Southeast U.S and can explain differences in carbon uptake rates when compared against available satellite data. Sensitivity tests of the DCBP model for specific locations within the Southeast U.S. reveal that selecting an inference period for the data assimilation step in the predictive phenology model amounts to imposing a plant water use strategy as a result of the non-stationarity of wet and dry periods in the assimilated data. In a study where the DCHM-PV is applied to the entire Southeast U.S. to dynamically estimate changes in canopy structure in tandem with photosynthesis rates, we find that extreme hurricane and wildfire events significantly reduce vegetation canopies with losses in potential carbon uptake rates as high as 400 g C/m2. Finally, in Southern Africa persistent wetlands correspond to less efficient water use for photosynthesis during the wet season, but higher overall photosynthesis rates because wetland vegetation has unlimited access to water in soils. Further, the persistence of these wetland well into the dry season depends on localized convective storms during the months transitioning from wet to dry seasons. Overall, this thesis contributes a quantitative understanding of the impacts of local and regional disturbances on the coupled water and carbon cycles and provides a general roadmap to evaluate ecosystem health and sustainability in light of phenologic and photosynthetic demands for water resources that can be adapted globally.