Modelagem dos impactos do aumento da concentração de CO2 e nutrientes sobre a composição funcional na Floresta Amazônica
Carregando...
Data
2021-12-14
Autores
Darela Filho, João Paulo [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A hipótese de perda da floresta amazônica em decorrência das mudanças climáticas foi postulada no final dos anos 90, sugerindo um declínio acentuado da biomassa vegetal simulada por modelos de superfície terrestre (Land Surface Models) em resposta a retroalimentações entre o ciclo de carbono (C) e as mudanças de precipitação e temperatura projetadas para o século XXI por modelos de circulação geral (General Circulation Models). O efeito de fertilização por CO2 (dióxido de C) pode, eventualmente, manter a produtividade e a biomassa florestal nas próximas décadas, evitando ou adiando a suposta descaracterização florestal na região. A resposta da floresta amazônica aos efeitos diretos e indiretos do CO2 elevado é, contudo, incerta. Principalmente em virtude de duas características importantes, até aqui não consideradas conjuntamente em experimentos realizados com modelos de vegetação: o papel potencialmente limitante do nitrogênio e do fósforo (N e P) sobre o efeito de fertilização por CO2 e as possíveis alterações de composição funcional comunidades vegetais sob diferentes regimes de estresse e perturbação originadas pelas mudanças climáticas. Este estudo apresenta uma abordagem de modelagem de ecossistemas terrestres baseada em processos ecofisiológicos e na biodiversidade de formas e funções vegetais, representada pela alta variabilidade de atributos funcionais de plantas. Considerando os ciclos do nitrogênio e fósforo acoplados ao ciclo de carbono, numa combinação inovadora de características de modelagem dinâmica de ecossistemas terrestres. O novo modelo foi aplicado para a simulação dos ecossistemas da Pan-amazônia num período climatológico e histórico recente (1979-2016) e em um teste de sensibilidade envolvendo variações na temperatura, precipitação, concentração de CO2 e nutrientes (N e P). A simulação histórica foi comparada com dados de referência em um processo de avaliação biogeoquímica, baseada em um protocolo de amplo uso na modelagem de ecossistemas terrestres. A simulação histórica e os dados gerados no teste de sensibilidade foram a base para o desenvolvimento de um método de análise funcional multidimensional que foi utilizado aqui para identificar com precisão importantes interações entre a composição funcional da vegetação simulada e os processos ecossistêmicos sob as diferentes condições ambientais simuladas. Os resultados dos testes de sensibilidade foram comparados qualitativamente a resultados de experimentos de manipulação dos recursos e variáveis climáticas. Finalmente, sob a luz das análises anteriores, uma análise detalhada da simulação histórica foi realizada, buscando esclarecer os efeitos diversos na vegetação, que são causados pelo aumento do stress hídrico, da temperatura e da concentração atmosférica de CO2 entre 1979 e 2016. A combinação de maior diversidade funcional e dos ciclos de N e P acoplados ao ciclo de C em um modelo dinâmico de vegetação revelou interações importantes entre composição funcional da área simulada e o complexo balanço entre assimilação de C e respiração autotrófica, que em última instância compõem o principal controle sobre a biomassa em ecossistemas terrestres. Adicionalmente, um resultado relevante não previsto no plano inicial é apresentado: um conjunto de mapas e uma análise sobre as diferentes formas químicas de P nos solos da área de estudo.
The Amazon Forest dieback hypothesis was postulated in early 2000, and many studies suggested that a future climate induced dieback of the Amazon Forest was a real possibility. However, the CO2 fertilization effect may possibly enhance forest productivity and biomass for decades to come, phasing out the projected forest loss. The response of the Amazon Forest to elevated CO2 effects is however uncertain. Two important features that have not been considered in vegetation model projections: (1) the potentially limiting role of nutrients, namely, nitrogen (N) and phosphorus (P), on the CO2 fertilization effect and (2) the possibility of interspecific response variation (i.e., changes in community composition). For example, different plant growth/surviving strategies are likely to differ in their responses to climate change and elevated CO2 more than they differ in terms of effects on ecosystem processes and state. Changes at the community level can have cascading effects on ecosystems and synergistic effects with other ecosystem processes, attributes, or even environmental changes. The biogeochemical cycles, mainly nitrogen and phosphorus, are increasingly recognized as important controls of the future global carbon cycle. In addition, models based on high plant functional variability rise as a promising alternative to better characterize the dynamics of changing ecosystems. This study presents an innovative modelling approach considering the fully coupling of carbon (C), N and P biogeochemical cycles, in a modeling framework that simulates the coexistence of hundreds of Plant Life-history Strategies. A never tested combination of characteristics in terrestrial ecosystem models. The new model was applied to simulate the ecosystems of the Pan-amazon in a recent climatological and historical period (1979-2016) and in a sensitivity test involving variations in temperature, precipitation, CO2, and nutrient concentration (N and P). The historical simulation was compared with reference data in a biogeochemical evaluation process, based on a protocol of wide use in the modeling of terrestrial ecosystems. The historical simulation and the data generated in the sensitivity test were the basis for the development of a multidimensional functional analysis method that was used here to accurately identify important interactions between the functional composition of the simulated vegetation and the ecosystem processes under different simulated environmental conditions. The results of the sensitivity tests were qualitatively compared to the results of experiments involving the manipulation of nutrients and climatic variables. Finally, in the light of the previous analyses, a detailed analysis of the historical simulation was performed, aiming the elucidation of the various effects on vegetation, which are caused by increased water stress, temperature, and atmospheric CO2 concentration between 1979 and 2016. The combination of greater functional diversity and the N and P cycles coupled to the C cycle in a dynamic vegetation model revealed important interactions between functional composition of the simulated area and the complex balance between C assimilation and autotrophic respiration, which ultimately make up the main control over biomass in terrestrial ecosystems. Additionally, a relevant result not foreseen in the initial plan is presented: a set of maps and an analysis of the different chemical forms of P in the soils of the study area.
The Amazon Forest dieback hypothesis was postulated in early 2000, and many studies suggested that a future climate induced dieback of the Amazon Forest was a real possibility. However, the CO2 fertilization effect may possibly enhance forest productivity and biomass for decades to come, phasing out the projected forest loss. The response of the Amazon Forest to elevated CO2 effects is however uncertain. Two important features that have not been considered in vegetation model projections: (1) the potentially limiting role of nutrients, namely, nitrogen (N) and phosphorus (P), on the CO2 fertilization effect and (2) the possibility of interspecific response variation (i.e., changes in community composition). For example, different plant growth/surviving strategies are likely to differ in their responses to climate change and elevated CO2 more than they differ in terms of effects on ecosystem processes and state. Changes at the community level can have cascading effects on ecosystems and synergistic effects with other ecosystem processes, attributes, or even environmental changes. The biogeochemical cycles, mainly nitrogen and phosphorus, are increasingly recognized as important controls of the future global carbon cycle. In addition, models based on high plant functional variability rise as a promising alternative to better characterize the dynamics of changing ecosystems. This study presents an innovative modelling approach considering the fully coupling of carbon (C), N and P biogeochemical cycles, in a modeling framework that simulates the coexistence of hundreds of Plant Life-history Strategies. A never tested combination of characteristics in terrestrial ecosystem models. The new model was applied to simulate the ecosystems of the Pan-amazon in a recent climatological and historical period (1979-2016) and in a sensitivity test involving variations in temperature, precipitation, CO2, and nutrient concentration (N and P). The historical simulation was compared with reference data in a biogeochemical evaluation process, based on a protocol of wide use in the modeling of terrestrial ecosystems. The historical simulation and the data generated in the sensitivity test were the basis for the development of a multidimensional functional analysis method that was used here to accurately identify important interactions between the functional composition of the simulated vegetation and the ecosystem processes under different simulated environmental conditions. The results of the sensitivity tests were qualitatively compared to the results of experiments involving the manipulation of nutrients and climatic variables. Finally, in the light of the previous analyses, a detailed analysis of the historical simulation was performed, aiming the elucidation of the various effects on vegetation, which are caused by increased water stress, temperature, and atmospheric CO2 concentration between 1979 and 2016. The combination of greater functional diversity and the N and P cycles coupled to the C cycle in a dynamic vegetation model revealed important interactions between functional composition of the simulated area and the complex balance between C assimilation and autotrophic respiration, which ultimately make up the main control over biomass in terrestrial ecosystems. Additionally, a relevant result not foreseen in the initial plan is presented: a set of maps and an analysis of the different chemical forms of P in the soils of the study area.
Descrição
Palavras-chave
Ecologia, Ecossistemas, Ciclos biogeoquímicos, Produtividade primária, Simulação (Computadores), Ecology, Ecosystems, Primary productivity