Mudanças climáticas no estado de São Paulo: associação entre gases de efeito estufa, poluentes atmosféricos e indicadores econômicos e de saúde
Data
2023-05-12
Autores
Debone, Daniela 
Orientadores
Miraglia, Simone Georges El Khouri
Tipo
Tese de doutorado
Título da Revista
ISSN da Revista
Título de Volume
Resumo
As mudanças climáticas são reconhecidas como um dos maiores desafios do século XXI,
implicando em sérios agravos socioeconômicos e à saúde global. Os grandes centros urbanos
sofrem profundamente as consequências causadas pelas mudanças climáticas e poluição
atmosférica, bem como contribuem diretamente para a intensificação desses fenômenos.
Diante disso, esta pesquisa tem como objetivo central investigar as relações entre os gases de
efeito estufa, indicadores de saúde e econômicos do estado de São Paulo, entre 2000 e 2020.
Procurando solucionar, de maneira completa e sistematizada, cada objetivo específico
proposto, optou-se pelo formato multipaper, constituído por uma coleção de artigos
científicos independentes, porém interrelacionados. O artigo 1, a partir de uma análise
descritiva das séries temporais dos principais gases de efeito estufa, mostra que o setor de
transportes é o principal emissor de CO2 do estado. O artigo 2, valendo-se da melhora da
qualidade do ar, utiliza a função de risco relativo para associar diminuição da concentração
de poluentes com mortes prematuras evitadas. Os valores encontrados representaram uma
economia de aproximadamente 8 milhões de dólares, durante a greve dos caminhoneiros de
2018. O artigo 3 também utiliza a função de risco relativo e revela que a redução da poluição
atmosférica, no início da pandemia de COVID-19, resultou em mortes prematuras evitadas,
que representaram uma economia de 720 milhões de dólares. O artigo 4 trata-se de uma
revisão sistemática sobre os principais modelos econométricos aplicados para estabelecer a
relação entre CO2, PIB e consumo energético. O Artigo 5 compara a precisão de três
diferentes métodos, regressão linear multivariada, regressão elastic-net e redes neurais
artificiais, e demonstra a superioridade deste último em modelar as emissões de CO2, a partir
do PIB e energia (MAPE = 0,76% e R2 = 1,000). A partir da análise de diferentes estruturas
de redes neurais artificiais, o artigo 6 demonstra que, durante o verão, a temperatura máxima
é o indicador mais adequado para modelar mortalidade total. A análise da defasagem
temporal entre as variáveis revela que temperaturas elevadas podem afetar a saúde no mesmo
dia de exposição (lag 0; MAPE = 7,1%), além de ter efeito cumulativo, indicando impactos
negativos em eventos de cinco dias com temperaturas elevadas (lag-cumulativo 5; MAPE =
7,0%). Em conjunto, os resultados desta tese fortalecem o debate sobre a importância de
ações de mitigação das emissões de gases de efeito estufa e poluição atmosférica, bem como
da redução da dependência do consumo de combustível fóssil, que deveriam ser metas
priorizadas na gestão pública, em prol da minimização dos efeitos das mudanças climáticas.
Climate change is recognized as one of the greatest challenges of the 21st century, resulting in serious socioeconomic and global health losses. Large urban centers suffer deeply from the consequences caused by climate change and air pollution, as well as directly contributing to the intensification of these phenomena. Therefore, this research has as main objective to investigate the relationship between greenhouse gas emissions, health and economic indicators in the state of São Paulo, between 2000 and 2020. Seeking to solve each specific objective proposed in a complete and systematized manner, the multipaper format was chosen, consisting of a collection of independent, but interrelated, scientific articles. Article 1 refers to a descriptive analysis of the time series of the main greenhouse gases and shows that the transport sector is the main emitter of CO2 in the state. Article 2, considering the improvement in air quality, uses the relative risk function to associate the decrease in the concentration of pollutants with avoided premature deaths. The values found represented a saving of approximately 8 million dollars, during the truckers’ strike of 2018. Article 3 also uses the relative risk function and reveals that the reduction of atmospheric pollution, during the first 90 days of the COVID-19 pandemic, resulted in avoided premature deaths, which represented savings of US $ 720 million. Article 4 is a systematic review of the main econometric models applied to establish the relationship between CO2, GDP and energy consumption. Article 5 compares the precision of three different methods - multivariate linear regression, elastic-net regression and artificial neural networks - and demonstrates the latter's superiority in modeling CO2 emissions from GDP and energy (MAPE = 0,76% and R2 = 1,000). Based on the analysis of different structures of artificial neural networks, article 6 demonstrates that, during the warm season, the maximum temperature is the most suitable indicator to modelling total mortality. The analysis of the time-lag effects reveals high temperatures can affect health on the same day of exposure (lag 0; MAPE = 7.1%), besides having a cumulative effect, indicating negative impacts in hot five-day events (cumulative-lag 5; MAPE = 7.0%).Taken together, the results of this research strengthen the debate about the importance of actions to mitigate emissions of greenhouse gases and atmospheric pollution, as well as reducing the dependence on fossil fuel consumption, which should be prioritized goals in public management, to minimize the effects of climate change.
Climate change is recognized as one of the greatest challenges of the 21st century, resulting in serious socioeconomic and global health losses. Large urban centers suffer deeply from the consequences caused by climate change and air pollution, as well as directly contributing to the intensification of these phenomena. Therefore, this research has as main objective to investigate the relationship between greenhouse gas emissions, health and economic indicators in the state of São Paulo, between 2000 and 2020. Seeking to solve each specific objective proposed in a complete and systematized manner, the multipaper format was chosen, consisting of a collection of independent, but interrelated, scientific articles. Article 1 refers to a descriptive analysis of the time series of the main greenhouse gases and shows that the transport sector is the main emitter of CO2 in the state. Article 2, considering the improvement in air quality, uses the relative risk function to associate the decrease in the concentration of pollutants with avoided premature deaths. The values found represented a saving of approximately 8 million dollars, during the truckers’ strike of 2018. Article 3 also uses the relative risk function and reveals that the reduction of atmospheric pollution, during the first 90 days of the COVID-19 pandemic, resulted in avoided premature deaths, which represented savings of US $ 720 million. Article 4 is a systematic review of the main econometric models applied to establish the relationship between CO2, GDP and energy consumption. Article 5 compares the precision of three different methods - multivariate linear regression, elastic-net regression and artificial neural networks - and demonstrates the latter's superiority in modeling CO2 emissions from GDP and energy (MAPE = 0,76% and R2 = 1,000). Based on the analysis of different structures of artificial neural networks, article 6 demonstrates that, during the warm season, the maximum temperature is the most suitable indicator to modelling total mortality. The analysis of the time-lag effects reveals high temperatures can affect health on the same day of exposure (lag 0; MAPE = 7.1%), besides having a cumulative effect, indicating negative impacts in hot five-day events (cumulative-lag 5; MAPE = 7.0%).Taken together, the results of this research strengthen the debate about the importance of actions to mitigate emissions of greenhouse gases and atmospheric pollution, as well as reducing the dependence on fossil fuel consumption, which should be prioritized goals in public management, to minimize the effects of climate change.