Leaf phenology refers to the timing of leaf life cycle events and is essential to our understanding of the earth system as it impacts the terrestrial carbon and water cycles and indirectly global climate through changes in surface roughness and albedo. Traditionally, leaf phenology is described as a response to higher temperatures in spring and lower temperatures in autumn for temperate regions. With the advent of carbon ecosystem models however, we need a better representation of seasonal cycles, one that is able to explain phenology in different areas around the globe, including tropical regions, and has the capacity to predict phenology under future climates. We propose a global phenology model based on the hypothesis that phenology is a strategy through which plants reach optimal carbon assimilation. We fit this 14 parameter model to five years of space borne data of leaf area index using a Bayesian fitting algorithm and we use it to simulate leaf seasonal cycles across the globe. We explain the observed increase in leaf area over the Amazon basin during the dry season through an increase in available direct solar radiation. Seasonal cycles in dry tropical areas are explained by the variation in water availability, while phenology at higher latitudes is driven by changes in temperature and daylength. We explore the hypothesis that phenological traits can be explained at the biome (plant functional group) level and we show that some characteristics can only be explained at the species level due to local factors such as water and nutrient availability. We anticipate that our work can be incorporated into larger earth system models and used to predict future phenological patterns.

Land ecosystems absorb about a quarter of all human emissions of carbon (C) by fossil fuel burning and land use change. This percentage varies greatly within years due to the land ecosystem response to climate variability and disturbance. Significant uncertainties remain in our knowledge of the magnitude and spatio-temporal changes in the land C sinks. The aims of my thesis are 1) to evaluate the capacity of different dynamic global vegetation models (DGVMs) to reproduce the fluxes and stocks of the land C cycle and 2) to analyse the drivers of change in the land C over the last two decades (1990-2009). In the first part of this thesis I evaluated the DGVM results over two regions: the Northern Hemisphere (NH) and the Tropics. Over the NH DGVMs tend to simulate longer growing seasons and a greater positive leaf area index trend in response to warming than that observed from satellite data. For the tropical region we found a high spatial correlation between the DGVMs and the observations for C stocks and fluxes, but the models produced higher C stocks over the non-forested areas. In the second part I studied the processes controlling the regional land C cycle. The findings can be summarized as: (1) the land CO2 sink has increased over the study period, through increases in tropical and southern regions with negligible change in northern regions; (2) globally and in most regions, the land sinks are not increasing as fast as the growth rate of excess atmospheric CO2 and (3) changes in water availability, particularly over the dry season, played a fundamental role in determining regional trends in NPP. My work seeks to improve our understanding of the relationship between the C cycle and its drivers, however considerable research is needed to understand the role of additional processes such as land use change, nitrogen deposition, to mention just a few.

Submitted by Helena Alves do Prado null (helenapradoeco@gmail.com) on 2017-12-08T19:34:27Z No. of bitstreams: 1 Dissertação_HAP.pdf: 5105198 bytes, checksum: 05421b9dc8b38cd4f75d69d9f2066044 (MD5) / Approved for entry into archive by Adriana Aparecida Puerta null (dripuerta@rc.unesp.br) on 2017-12-11T16:34:40Z (GMT) No. of bitstreams: 1 prado_ha_me_rcla.pdf: 5105198 bytes, checksum: 05421b9dc8b38cd4f75d69d9f2066044 (MD5) / Made available in DSpace on 2017-12-11T16:34:40Z (GMT). No. of bitstreams: 1 prado_ha_me_rcla.pdf: 5105198 bytes, checksum: 05421b9dc8b38cd4f75d69d9f2066044 (MD5) Previous issue date: 2017-07-27 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / A importância das florestas tropicais para o ciclo de carbono global e a incerteza de suas respostas a novas condições ambientais evidenciam a necessidade de aprofundar o conhecimento acerca de como as mudanças climáticas repercutirão nas dinâmicas florestais e nos processos fisiológicos associados. Essa dissertação, vinculada ao experimento AmazonFACE, objetiva avaliar como o incremento de CO2 atmosférico poderia repercutir na produtividade e eficiência no uso da água da Floresta Amazônica; se atributos funcionais envolvidos nesses processos (vcmax e g1) seriam modificados; e se ocorreriam mudanças na configuração funcional modelada da região. Para melhor caracterizar a diversidade funcional e as diferentes estratégias de sobrevivência adotadas pelas plantas em um ecossistema hiperdiverso, desenvolveu-se o CAETÊ (Carbon and Ecosystem Functional-Trait Evaluation Model) que, diferentemente dos modelos baseados em tipos funcionais, não parametriza a priori as possíveis combinações de atributos funcionais e simula um número superior de estratégicas de vida (PLS), sendo essas amostradas semi-aleatoriamente pelo modelo. Ao avaliar a performance do CAETÊ, verificou-se que a consideração de um número elevado de PLS – em vez de uma abordagem baseada em poucos tipos funcionais – simula maiores taxas de produtividade e condutância estomática; e que esse aumento tende à estabilização, com o aumento de estratégias consideradas. A aplicação do modelo foi feita utilizando a concentração de CO2 a ser empregada no AmazonFACE, de 600 ppm. Nas simulações, verificou-se uma redução nos valores médios de vcmax e g1, assim como uma modificação nos seus padrões de distribuição. Essas alterações associaram-se a um incremento nas taxas modeladas de produtividade e no uso da água, nas florestas tropicais, próximo a 13.5% (GPP) e 14.2% (NPP); redução de 3.8% na condutância estomática; e maior eficiência no uso da água (+21.1%). O enriquecimento da atmosfera também esteve associado à redução de ~5.5% na riqueza funcional. À medida que se avança para condições ambientais sem precedentes, torna-se de grande valia o uso de novas abordagens pelos modelos vegetacionais, que possibilitem a consideração de um número superior de estratégias de sobrevivência das plantas; e a captação de possíveis alterações nas respostas de atributos funcionais essenciais e, consequentemente, na riqueza funcional dos ecossistemas. As florestas tropicais desempenham um papel essencial nessas dinâmicas, uma vez que, além de hiperdiversas, são centrais para o ciclo de carbono global. / The importance of tropical forests in the global carbon cycle and the uncertainty of the forest’s responses to new environmental conditions evidence the necessity of deepening our knowledge about how climate changes affect forest dynamics and the associated physiological processes. This dissertation, linked to the AmazonFACE experiment, aims to evaluate how increasing atmospheric levels of CO2 could influence primary productivity and water use efficiency of the Amazon Forest; if functional traits involved in such processes (vcmax e g1) would be modified; and if changes would occur in the region’s modeled functional configuration. Aiming for a better characterization of the functional diversity and the different survival strategies adopted by plants in a hyperdiverse ecosystem, we developed the CAETÊ (Carbon and Ecosystem Functional-Trait Evaluation Model) which, unlike models based on functional types, does not perform a priori parameterization of the possible combinations of functional traits and simulates a superior amount of life strategies (PLS), that are semi-randomly sampled by the model. When evaluating CAETÊ’s performance, we verified that the consideration high number of PLS – instead of a framework based in few functional types – simulates higher amounts of productivity and stomatal conductance; and that such increase tends to stabilization when more strategies are considered. The model was applied using the CO2 concentration to be used in AmazonFACE – 600ppm. Simulations showed a reduction in the average values of de vcmax and g1, as well as a change in their patterns of distribution. Such alterations are associated to an increase in modeled values of productivity and water use of tropical forests close to 13.5% (GPP) and 14.2% (NPP), a reduction of 3.8% in stomatal conductance, and higher efficiency in water use (+21.1%). The atmospheric enrichment was also associated with a ~5.5% reduction in the functional richness. As the environment advances to unprecedented conditions, it becomes clear how worthy is the use of new approaches by vegetational models that enable consideration of higher numbers of plant survival strategies; and the uptake of possible changes in responses of essential functional traits, therefore, in the functional richness of the environment. The tropical forests influence deeply such dynamics; once beyond hyperdiverse.