Soil and litter emission sources as important contributors to ozone production from volatile organic compounds in island tropical forests.

Journal: Environmental research
Published Date:

Abstract

While studies have confirmed that volatile organic compounds (VOCs) emitted directly by tropical island forest vegetation significantly influence ozone (O) production and climate change through atmospheric oxidation processes, the environmental effects of long-neglected soil and litter emission sources as key potential contributors to VOCs, particularly their driving mechanisms in near-surface O pollution formation, remain understudied. This investigation combines field observations with machine learning models to investigate the emission characteristics, sources, and contributions of VOCs from tropical island forests to O and secondary organic aerosol (SOA) formation. The results reveal discrepancies between traditional ozone formation potential (OFP) estimates and machine learning-based assessments. OFP calculations identified acetaldehyde and methanol as the dominant contributors to O formation, while toluene and monoterpenes were primary drivers of SOA formation. However, the XGBoost model integrated with the SHapley Additive exPlanations (SHAP) framework, which quantifies the dynamic impacts of VOCs under real-world atmospheric conditions, demonstrated that isoprene made the most significant contribution to O formation (|SHAP|=5.9), surpassing other VOCs. For SOA formation, benzene and toluene showed the highest contributions, with |SHAP| values of 1.2 and 0.8, respectively. By calculating initial VOC concentrations and applying the Positive Matrix Factorization (PMF) model, we identified four VOC sources: soil and litter emissions (41 %), oxidative formation (28.5 %), anthropogenic transport (16.6 %), and direct plant emissions (13.9 %). Photochemical reactions caused significant losses of plant-derived VOCs during transport; after accounting for photochemical losses, the contribution of direct plant emissions increased from 4.6 % to 13.9 %. SHAP analysis highlighted that soil and litter emissions contributed most significantly to O formation (|SHAP|=14.7), offering theoretical advantages over traditional OFP estimates that prioritized plant emissions. The SHAP framework, derived from observational data mining, effectively mitigated biases caused by temporal or regional variations and provided a more accurate quantification of rapidly consumed VOCs during active photochemical processes, thereby addressing limitations of conventional OFP methods. These findings indicate that VOCs from soil and litter emissions in tropical forest regions exert a substantial influence on local O formation.

Authors

  • Huayuan Zhou
    College of Geography and Environmental Sciences, Hainan Normal University, Key Laboratory of Tropical Island Land Surface Processes and Environmental Changes of Hainan Province, Haikou 571127, China; Sanya Land-Sea Interface Critical Zone Field Scientific Observation and Research Station, Sanya 572022, China.
  • Mengyang Fang
    Sanya Land-Sea Interface Critical Zone Field Scientific Observation and Research Station, Sanya 572022, China; Haikou Marine Geological Survey Center, China Geological Survey, Haikou 571127, China.
  • Yunhua Chang
    Key Laboratory of Meteorological Disaster (KLME), Ministry of Education & Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters (CIC-FEMD), Nanjing University of Information Science & Technology, Nanjing 210044, China.
  • Zhongqin Li
    State Key Laboratory of Cryospheric Sciences, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou, 730000, China.
  • Feiteng Wang
    State Key Laboratory of Cryospheric Sciences, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou, 730000, China.
  • Yuying Yu
    College of Geography and Environmental Sciences, Hainan Normal University, Key Laboratory of Tropical Island Land Surface Processes and Environmental Changes of Hainan Province, Haikou 571127, China; Sanya Land-Sea Interface Critical Zone Field Scientific Observation and Research Station, Sanya 572022, China.
  • Ming Zeng
    School of Electrical Engineering and Automation, Tianjin University, Tianjin 300072, China.
  • Huasi Zhan
    Sanya Land-Sea Interface Critical Zone Field Scientific Observation and Research Station, Sanya 572022, China; Haikou Marine Geological Survey Center, China Geological Survey, Haikou 571127, China.
  • Zhizhong Zhao
    College of Geography and Environmental Sciences, Hainan Normal University, Key Laboratory of Tropical Island Land Surface Processes and Environmental Changes of Hainan Province, Haikou 571127, China; Sanya Land-Sea Interface Critical Zone Field Scientific Observation and Research Station, Sanya 572022, China. Electronic address: zhizhong@hainnu.edu.cn.
  • Xi Zhou
    George E. Wahlen Department of Veterans Affairs Medical Center, Salt Lake City, UT, USA.

Keywords

No keywords available for this article.