Infusing satellite data into aerosol forecast for near real-time episode detection and diagnosis in East Asia.

A near-real-time (NRT) aerosol forecast and diagnostic approach is developed based on the system of Infusing satellite Data into Environmental Applications for East Asia, herein denoted as IDEA-EA. The design incorporates a 0.5-degree Global Forecast System (GFS) and Visible Infrared Imaging Radiometer Suite (VIIRS) aerosol and cloud retrievals for meteorological and remote sensing inputs. The primary output of IDEA-EA includes aerosol forward and backward air mass trajectory forecasts, migration visualization, and data synthesis purposed for NRT aerosol detection, monitoring, and source tracing in East Asia. Two aerosol episodes of Southeast Asia (SEA) biomass burning and Chinese haze infusion with Gobi dust are illustrated by IDEA-EA to demonstrate its forecast and source tracing capabilities. In the case of SEA biomass burning (late March 2021), forward trajectories of IDEA-EA forecasted air masses with high aerosol optical depth (AOD) from SEA affecting Taiwan. The IDEA-EA forecasts were verified by increased AOD and surface PM2.5 observations at a mountain site. In the case of the Chinese haze (October 30, 2019), backward trajectories from the northern tip of Taiwan traced air masses back to the east coast of mainland China and possibly further to the Gobi Desert. Compared with conventional numerical model simulations, the combination of the state-of-the-art aerosol remote sensing and trajectory modeling in IDEA-EA provides a cost-effective alternative for air quality management.

Taiwan ozone trend in response to reduced domestic precursors and perennial transboundary influence.

Two decades of nitrogen oxides (NOx) and volatile organic compounds (VOCs) in ambient concentrations have been continuously reduced by nearly 60% since 1994. However, the annual mean ozone (O3) concentrations in Taiwan had leveled off for more than a decade. Furthermore, the significant cuts in precursors did lead to reduction in episodic days (O3≥ 100 ppbv) over time, implying the weakened photochemical production of O3. Simultaneously, the number of low O3 days (O3 < 20 ppbv) also decreased due to weakened NO titration. Nevertheless, the reduction of both high and low O3 days resulted in an increase in the number of medium O3 days (O3 from 60-80 ppbv). While the nighttime mean O3 continued to increase, the daytime mean O3 leveled off and coincided with the marine background concentration of approximately 40 ppbv, revealing a gradual decrease in the daytime-nighttime mean O3 difference. In addition to multiple chemical reasons, transboundary O3 from Asian continental outflow is thought to be another major contributor to the prolonged leveling of domestic O3. To quantitatively illustrate the role of the transboundary effect, model simulations were conducted in two ways: one assessed the time percentage of a year affected by the transboundary influence; and the other involved decoupling the transboundary O3 from the domestically produced O3 to quantify the amount of domestic O3 that can be mitigated by prescribing scenarios; namely, the so-called sensitivity test. The sensitivity test suggests that although controlling domestic emissions is still vital to contain O3 under weak transboundary conditions, the overall O3 averages and the long-term trend at the island scale are closely associated with the transboundary influence. The gradual increased medium O3 concentrations to approximate the marine background level could imply that future O3 concentration on the island will be more influenced more by the transboundary conditions and less relevant to the domestic containment measures.

Source Apportionment of Volatile Organic Compounds (VOCs) by Positive Matrix Factorization (PMF) supported by Model Simulation and Source Markers - Using Petrochemical Emissions as a Showcase.

This study demonstrates the use of positive matrix factorization (PMF) in a region with a major Petrochemical Complex, a prominent source of volatile organic compounds (VOCs), as a showcase of PMF applications. The PMF analysis fully exploited the quality and quantity of the observation data, sufficed by a cluster of 9 monitoring sites within a 20 km radius of the petro-complex. Each site provided continuous data of 54 speciated VOCs and meteorological variables. Wind characteristics were highly seasonal and played a decisive role in the source-receptor relationship, hence the dataset was divided into three sub-sets in accordance with the prevailing wind flows. A full year of real-time data were analyzed by PMF to resolve into various distinct source types including petrochemical, urban, evaporative, long-range air parcels, etc., with some sites receiving more petro-influence than others. To minimize subjectivity in the assignment of the PMF source factors, as commonly seen in some PMF works, this study attempted to solidify PMF results by supporting with two tools of spatially/temporally resolved air-quality model simulations and observation data. By exploiting the two supporting tools, the dynamic process of individual sources to a receptor were rationalized. Percent contributions from these sources to the receptor sites were calculated by summing over the occurrence of different source types. Interestingly, although the Petro-complex is the single largest local VOC source in the 20 km radius study domain, all monitoring sites in the region received far less influence from the Petro-complex than from other emission types within or outside the region, which together add up to more than 70% of the total VOC abundance.

Air quality impacted by local pollution sources and beyond - Using a prominent petro-industrial complex as a study case.

The present study combines high-resolution measurements at various distances from a world-class gigantic petrochemical complex with model simulations to test a method to assess industrial emissions and their effect on local air quality. Due to the complexity in wind conditions which were highly seasonal, the dominant wind flow patterns in the coastal region of interest were classified into three types, namely northeast monsoonal (NEM) flows, southwest monsoonal (SEM) flows and local circulation (LC) based on six years of monitoring data. Sulfur dioxide (SO2) was chosen as an indicative pollutant for prominent industrial emissions. A high-density monitoring network of 12 air-quality stations distributed within a 20-km radius surrounding the petrochemical complex provided hourly measurements of SO2 and wind parameters. The SO2 emissions from major industrial sources registered by the monitoring network were then used to validate model simulations and to illustrate the transport of the SO2 plumes under the three typical wind patterns. It was found that the coupling of observations and modeling was able to successfully explain the transport of the industrial plumes. Although the petrochemical complex was seemingly the only major source to affect local air quality, multiple prominent sources from afar also played a significant role in local air quality. As a result, we found that a more complete and balanced assessment of the local air quality can be achieved only after taking into account the wind characteristics and emission factors of a much larger spatial scale than the initial (20 km by 20 km) study domain.

Diagnostic modeling of PAMS VOC observation.

Although a number of gas-phase chemical mechanisms, such as CBM-IV, RADM2, and SAPRC have been successful in studying gas-phase atmospheric chemical processes, they all used different combinations of lumped organic species to describe the role of organics in gas-phase chemical processes. Photochemical Assessment Monitoring Stations (PAMS) have been in use for over a decade and yet it is not clear how the detailed organic species measured by PAMS compare to the lumped modeled species. By developing a detailed mechanism specifically for the PAMS organics and embedding this diagnostic model within a regional-scale transport and chemistry model, one can then directly compare PAMS observation with regional-scale model simulations. By means of this comparison one can perhaps better evaluate model performance. The Taiwan Air Quality Model (TAQM) was modified by adding a submodel with transport processes and chemical mechanism for interactions of the 56 species observed by PAMS. It is assumed that TAQM can simulate the overall regional-scale environment including time evolution of oxidants and radicals; these results are then used to simulate the evolution of PAMS organics with species-specific source functions, meteorological transport, and chemical interactions. Model simulations of each PAMS organic were compared with PAMS hourly surface measurements. A case study with data collected at three sites in central Taiwan showed that when meteorological simulations were comparable with observations, diurnal patterns of most organics performed well with PAMS data after emissions were corrected. It is found emissions of over half of the PAMS species require correction, some by surprisingly large factors. With such correlation, simulated time evolution of ratios of ethylbenzene/m,p-xylenes and ethane/n-butane showed similar behaviors as shown by observation data. From the results of PAMS organics diurnal variations as well as indicator ratios, one can conclude that PAMS Air Quality Model (PAMS-AQM) has been successfully developed and can be applied to the study of evolution of PAMS organics in regional and urban environments. Further, one finds that an existing VOC emissions estimation procedure heavily dependent on U.S.-data based emissions speciation factors is suspect in application in Taiwan and perhaps in other countries as well. A protocol, using PAMS-AQM for testing consistency between detailed VOC emissions and PAMS observations, has been developed and demonstrated.