Scholarlink

<p>This study applied Singular Spectrum Analysis (SSA) to investigate the temporal variability of carbon monoxide (CO) and tropospheric ozone (O₃) concentrations in Campo Grande, Brazil, between 2003 and 2018. By decomposing the time series into principal components, it was possible to isolate long-term trends, seasonal cycles, and residual variability. The first principal component explained 87.6% of the variance in CO and 65.2% in O₃, highlighting differences in the pollutants’ temporal structures. Seasonal cycles revealed annual peaks during the dry season (August to October), confirming the influence of biomass burning on CO and the photochemical behavior of O₃. The reconstructed series using nine components showed high fidelity, with performance metrics of R² = 0.93, RMSE = 112.4 ppb, and MAE = 84.7 ppb for CO, and R² = 0.95, RMSE = 6.8 ppb, and MAE = 5.2 ppb for O₃. These results demonstrate SSA’s effectiveness in capturing deterministic patterns and isolating high-frequency noise. The findings provide scientific support for targeted environmental policies, such as stricter traffic control, improved fire management, and integration of SSA-based monitoring into local agencies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Spectral decomposition and temporal dynamics of CO and O₃ in Campo Grande, Brazil: a singular spectrum analysis approach

  • 期刊论文

  • 发表日期：

    2025年11月5日

查看全文
  • Amaury de Souza,
  • Rafael da Silva Palácios,
  • Danielle Christine Stenner Nassarden,
  • Fernando Lucambio Pérez

摘要

This study applied Singular Spectrum Analysis (SSA) to investigate the temporal variability of carbon monoxide (CO) and tropospheric ozone (O₃) concentrations in Campo Grande, Brazil, between 2003 and 2018. By decomposing the time series into principal components, it was possible to isolate long-term trends, seasonal cycles, and residual variability. The first principal component explained 87.6% of the variance in CO and 65.2% in O₃, highlighting differences in the pollutants’ temporal structures. Seasonal cycles revealed annual peaks during the dry season (August to October), confirming the influence of biomass burning on CO and the photochemical behavior of O₃. The reconstructed series using nine components showed high fidelity, with performance metrics of R² = 0.93, RMSE = 112.4 ppb, and MAE = 84.7 ppb for CO, and R² = 0.95, RMSE = 6.8 ppb, and MAE = 5.2 ppb for O₃. These results demonstrate SSA’s effectiveness in capturing deterministic patterns and isolating high-frequency noise. The findings provide scientific support for targeted environmental policies, such as stricter traffic control, improved fire management, and integration of SSA-based monitoring into local agencies.