Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → États-Unis → 2018 → Flashy, Patchy and Coupled : Using Spectral and Gas Exchange Approaches to Refine Dryland Carbon Uptake Predictions Across Spatial and Temporal Scales

University of Arizona (2018)

Flashy, Patchy and Coupled : Using Spectral and Gas Exchange Approaches to Refine Dryland Carbon Uptake Predictions Across Spatial and Temporal Scales

Barnes, Mallory

Titre : Flashy, Patchy and Coupled : Using Spectral and Gas Exchange Approaches to Refine Dryland Carbon Uptake Predictions Across Spatial and Temporal Scales

Auteur : Barnes, Mallory

Université de soutenance : University of Arizona

Grade : Doctor of Philosophy (PhD) 2018

Carbon dioxide, a greenhouse gas, traps heat in the atmosphere, causing the planet to warm. The rate at which plants take up atmospheric carbon dioxide depends on climatic and biophysical factors, including soil moisture, atmospheric demand, and drought. Climate models disagree on the magnitude and trend in the terrestrial carbon sink. Accurate assessment of the role of terrestrial plants in the carbon cycle is essential to predicting the degree of future climate change. Drought impacts on the carbon cycle are driven by the cascading impacts of leaf-level physiological responses to limit water loss. Dryland ecosystems like those in the Southwest (southwest United States and northwest Mexico) are an exemplary location to study drought impacts on the carbon cycle due to persistent water limitation and associated tight coupling between hydrologic and carbon cycles. Furthermore, future climate projections suggest more frequent and intense drought in the world’s water-limited regions. My dissertation research improves understanding of coupled carbon and water cycles in dryland ecosystems, and informs predictions of vegetation response to future climate conditions. Using a combination of remotely sensed and gas exchange data, I explore drought impacts on plant productivity and carbon uptake in water-limited systems across spatial and temporal scales. The studies contained in this dissertation address three key knowledge gaps : 1) the effects of drought timing on vegetation and ecosystem processes, 2) relationships between leaf-level spectral and physiological properties, and 3) impacts of climate variability on coupled carbon and hydrologic cycles and associated predictions of regional and global carbon dynamics. First, I investigate how the intra-annual timing of drought in the Southwest influences the productivity of grasslands, shrublands, and forests. This study underscores the importance of sub-annual droughts in dryland carbon uptake dynamics and identifies the critical climate period for Southwest forests, shrublands and grasslands during which climate conditions disproportionately impact annual carbon uptake. Second, I establish a link between spectral measures of productivity and photosynthetic capacity at the leaf level in the context of a field experiment. This experiment compared hyperspectral imagery with photosynthetic capacity estimated from leaf gas exchange measurements. Performed during a mid-summer period with low rainfall and associated reductions in photosynthetic capacity, the results of this experiment suggest that spectrally-derived estimates of photosynthetic capacity are robust to within-season temporal variation. Thirdly, I upscale ecosystem-level eddy covariance observations to the Southwest region to assess drought impacts on regional carbon uptake using machine learning techniques. The results of this study highlight the crucial importance of accounting for water balance and drought dynamics in studies of carbon uptake in water-limited ecosystems. The application of the derived algorithm to the global scale suggests that the inclusion of intra- and inter- annual drought metrics can improve modeled interannual variability in global carbon uptake. Soil moisture is a key control on vegetation productivity and carbon uptake in dryland ecosystems, and I use drought indices and meteorological variables as proxies for soil moisture dynamics in this research. Collectively, my work is showing how the timing and intensity of drought impacts carbon uptake and vegetation productivity in dryland ecosystems. This cross-scale approach provides new insights into drought impacts on vegetation productivity and biosphere-atmosphere interactions in drylands. Overall, improved representation of the spatial and temporal dynamics of interactions between drought and carbon in drylands will lead to better projections of future water and carbon cycling and the magnitude and speed of global climate change.


Version intégrale (4,7 Mb)

Page publiée le 29 janvier 2019