Papers
Master's Thesis
"Implementation and Evaluation of the Automated Model Reduction (AMORE) Version 1.1 Isoprene Oxidation Mechanism in GEOS-Chem"
Detailed chemical mechanisms are computationally challenging to include in large-scale chemical transport models such as GEOS-Chem. Employing a graph theory-based automated model reduction (AMORE) algorithm, we developed a new reduced (12 species and 23 reactions) gas-phase isoprene oxidation mechanism. We performed GEOS-Chem simulations for a full year (June 2018 – May 2019) with the default (BASE) and AMORE version 1.1 isoprene mechanisms at 2° × 2.5° horizontal resolution globally and 0.25° × 0.3125° resolution over the eastern United States (EUS). Additionally, we conducted BASE and AMORE sensitivity simulations in which biogenic isoprene and anthropogenic emissions were sequentially set to zero in the model. For the entire year simulated, GEOS-Chem was faster by 10% in total and 25% in the chemical solver (KPP) with the AMORE mechanism. Evaluating GEOS-Chem against surface observations from the Air Quality System (AQS) and Interagency Monitoring of Protected Visual Environments (IMPROVE) networks as well as satellite columns from the Tropospheric Monitoring Instrument (TROPOMI) and Cross-track Infrared Sounder (CrIS), our results show comparable accuracy in BASE and AMORE nested-grid simulations of air pollutants, with annual mean model bias changes of 1% for both PM2.5 and O3 over the EUS. From the sensitivity simulations, we find that US biogenic isoprene contributes to 7-9% of PM2.5 and 3-4% of O3 on average in summer over the EUS. This study indicates that AMORE is an attractive option for future GEOS-Chem modeling studies, especially where detailed isoprene chemistry is not the focus.
Detailed chemical mechanisms are computationally challenging to include in large-scale chemical transport models such as GEOS-Chem. Employing a graph theory-based automated model reduction (AMORE) algorithm, we developed a new reduced (12 species and 23 reactions) gas-phase isoprene oxidation mechanism. We performed GEOS-Chem simulations for a full year (June 2018 – May 2019) with the default (BASE) and AMORE version 1.1 isoprene mechanisms at 2° × 2.5° horizontal resolution globally and 0.25° × 0.3125° resolution over the eastern United States (EUS). Additionally, we conducted BASE and AMORE sensitivity simulations in which biogenic isoprene and anthropogenic emissions were sequentially set to zero in the model. For the entire year simulated, GEOS-Chem was faster by 10% in total and 25% in the chemical solver (KPP) with the AMORE mechanism. Evaluating GEOS-Chem against surface observations from the Air Quality System (AQS) and Interagency Monitoring of Protected Visual Environments (IMPROVE) networks as well as satellite columns from the Tropospheric Monitoring Instrument (TROPOMI) and Cross-track Infrared Sounder (CrIS), our results show comparable accuracy in BASE and AMORE nested-grid simulations of air pollutants, with annual mean model bias changes of 1% for both PM2.5 and O3 over the EUS. From the sensitivity simulations, we find that US biogenic isoprene contributes to 7-9% of PM2.5 and 3-4% of O3 on average in summer over the EUS. This study indicates that AMORE is an attractive option for future GEOS-Chem modeling studies, especially where detailed isoprene chemistry is not the focus.
EESC 4924 - Atmospheric Chemistry
"Spatiotemporal Variations of PM2.5 From Reference-Grade and Low-Cost Monitors in Rwanda"
byang_eesc4924_project_paper.pdf | |
File Size: | 1101 kb |
File Type: |
EAEE 4000 - Machine Learning for Environmental Engineering and Sciences
"PM2.5 Prediction for Cities in Sub-Saharan Africa"
Benjamin_Yang_EAEE4000_Paper.pdf | |
File Size: | 2643 kb |
File Type: |
Undergraduate Honors Thesis
"Diagnosing Summertime PM2.5 Biases of the Community Multiscale Air Quality Model"
Particulate matter with a diameter of 2.5 micrometers or less (PM2.5) is one of the most harmful ambient air pollutants to human health. To improve regional air quality forecasting, it is essential to upgrade numerical weather prediction models. The Environmental Protection Agency’s (EPA) Community Multiscale Air Quality (CMAQ) model is driven by two National Oceanic and Atmospheric Administration (NOAA) numerical weather predictions: the operational North American Mesoscale (NAM) model and the experimental Finite Volume Cubed-Sphere Global Forecasting System (FV3GFS) model. PM2.5 predictions by both models were compared and evaluated over the contiguous United States (CONUS) from 1-19 June 2019, using AirNow observations. Aircraft-derived planetary boundary layer (PBL) height and surface weather station observations were compared against the corresponding predicted meteorology.
The FV3GFS-CMAQ generally predicted less PM2.5 than the NAM-CMAQ in the eastern United States. Following a cold front passage over the Southeast, the NAM-CMAQ overpredicted PM2.5, while the FV3GFS-CMAQ underpredicted PM2.5. Similar divergences in PM2.5 predictions occurred on other cold front days. Enhanced vertical mixing due to wind shear in the FV3GFS-CMAQ weakened the temperature inversion in the nocturnal boundary layer, allowing for warmer and drier air from aloft to be mixed down. Due to this enhanced vertical mixing, the FV3GFS-CMAQ likely overpredicted PBL height and dry deposition, thereby reducing surface PM2.5 concentrations. The NAM-CMAQ probably has the preferred PBL scheme and resolution in our case study, as underprediction may cause greater PM2.5 exposure. Horizontal advection and wet deposition are other important PM2.5 removal mechanisms, which should be explored more extensively in future case studies over various regions and time periods.
Particulate matter with a diameter of 2.5 micrometers or less (PM2.5) is one of the most harmful ambient air pollutants to human health. To improve regional air quality forecasting, it is essential to upgrade numerical weather prediction models. The Environmental Protection Agency’s (EPA) Community Multiscale Air Quality (CMAQ) model is driven by two National Oceanic and Atmospheric Administration (NOAA) numerical weather predictions: the operational North American Mesoscale (NAM) model and the experimental Finite Volume Cubed-Sphere Global Forecasting System (FV3GFS) model. PM2.5 predictions by both models were compared and evaluated over the contiguous United States (CONUS) from 1-19 June 2019, using AirNow observations. Aircraft-derived planetary boundary layer (PBL) height and surface weather station observations were compared against the corresponding predicted meteorology.
The FV3GFS-CMAQ generally predicted less PM2.5 than the NAM-CMAQ in the eastern United States. Following a cold front passage over the Southeast, the NAM-CMAQ overpredicted PM2.5, while the FV3GFS-CMAQ underpredicted PM2.5. Similar divergences in PM2.5 predictions occurred on other cold front days. Enhanced vertical mixing due to wind shear in the FV3GFS-CMAQ weakened the temperature inversion in the nocturnal boundary layer, allowing for warmer and drier air from aloft to be mixed down. Due to this enhanced vertical mixing, the FV3GFS-CMAQ likely overpredicted PBL height and dry deposition, thereby reducing surface PM2.5 concentrations. The NAM-CMAQ probably has the preferred PBL scheme and resolution in our case study, as underprediction may cause greater PM2.5 exposure. Horizontal advection and wet deposition are other important PM2.5 removal mechanisms, which should be explored more extensively in future case studies over various regions and time periods.
Benjamin_Yang_Thesis.pdf | |
File Size: | 1609 kb |
File Type: |
GEOG 417 - Satellite Climatology
"The El Nino-Southern Oscillation: Relationships Between Sea Surface Temperature Anomaly, Rainfall, and Outgoing Longwave Radiation in the Central-Eastern Tropical Pacific"
Reliable satellite measurements of sea surface temperature (SST), precipitation, and outgoing longwave radiation (OLR) in the central-eastern tropical Pacific are indispensable for monitoring and predicting the El Nino-Southern Oscillation (ENSO). To test the consistency of the relationships among these three variables, monthly average SST anomalies, rainfall, and OLR in the Nino 3.4 region were compared over one full ENSO cycle—April 2009 to June 2011—and linearly correlated. Datasets from the National Aeronautics and Space Administration (NASA) Earth Observations website included SST anomalies from the Advanced Microwave Scanning Radiometer - Earth Observing System (AMSR-E), OLR from the Clouds and the Earth’s Radiant Energy System (CERES) sensors, and rainfall from Tropical Rainfall Measuring Mission (TRMM) Microwave Imager and Precipitation Radar. Monthly maps, 27 for each variable, were produced using the Imager Composite Explorer (ICE) analysis tool on the NASA Earth Observations website. Using Python, time series and scatter plots were created, and correlation coefficients were calculated.
The results of this investigation verified findings from past studies, clearly showing that OLR decreases as SST anomaly and rainfall increase. OLR and rainfall were strongest correlated, while SST and OLR were weakest correlated. The delayed reaction of the atmosphere to ocean warming might have explained the deepest convection that occurred one month following the highest SST anomalies during El Nino. Nonetheless, these variables were strongest correlated during El Nino and weakest correlated during La Nina, possibly due to the more spatially organized precipitation that occurred during El Nino. This investigation suggests that applying multiple variables may be advantageous in identifying the ENSO state. Future research involving more sophisticated methods, such as using different combinations of satellite instruments, study regions, and time periods, would help climatologists disseminate more accurate information regarding the ENSO on both interannual and interdecadal timescales.
Reliable satellite measurements of sea surface temperature (SST), precipitation, and outgoing longwave radiation (OLR) in the central-eastern tropical Pacific are indispensable for monitoring and predicting the El Nino-Southern Oscillation (ENSO). To test the consistency of the relationships among these three variables, monthly average SST anomalies, rainfall, and OLR in the Nino 3.4 region were compared over one full ENSO cycle—April 2009 to June 2011—and linearly correlated. Datasets from the National Aeronautics and Space Administration (NASA) Earth Observations website included SST anomalies from the Advanced Microwave Scanning Radiometer - Earth Observing System (AMSR-E), OLR from the Clouds and the Earth’s Radiant Energy System (CERES) sensors, and rainfall from Tropical Rainfall Measuring Mission (TRMM) Microwave Imager and Precipitation Radar. Monthly maps, 27 for each variable, were produced using the Imager Composite Explorer (ICE) analysis tool on the NASA Earth Observations website. Using Python, time series and scatter plots were created, and correlation coefficients were calculated.
The results of this investigation verified findings from past studies, clearly showing that OLR decreases as SST anomaly and rainfall increase. OLR and rainfall were strongest correlated, while SST and OLR were weakest correlated. The delayed reaction of the atmosphere to ocean warming might have explained the deepest convection that occurred one month following the highest SST anomalies during El Nino. Nonetheless, these variables were strongest correlated during El Nino and weakest correlated during La Nina, possibly due to the more spatially organized precipitation that occurred during El Nino. This investigation suggests that applying multiple variables may be advantageous in identifying the ENSO state. Future research involving more sophisticated methods, such as using different combinations of satellite instruments, study regions, and time periods, would help climatologists disseminate more accurate information regarding the ENSO on both interannual and interdecadal timescales.
Benjamin_Yang_GEOG417_Paper.pdf | |
File Size: | 1258 kb |
File Type: |
METEO 440W - Principles of Atmospheric Measurements
"Impacts of Homemade Rain Gauge Design and Micrometeorology on Precipitation Measurement Accuracy"
Accurate precipitation observations are essential in the verification of weather forecasts, the study of climate patterns, and the decision-making process across many industries. Here, we investigate the impacts of both rain gauge design and micrometeorological phenomena on precipitation measurement accuracy. To test inexpensive, nonstandard materials, a homemade rain gauge was built using a plastic funnel and glass jar for the collection container, a graduated cylinder for the measurement container, and plastic bucket for the wind shield. Precipitation was collected and measured daily from 14–27 September 2018 near Walker Building in University Park, Pennsylvania using three rain gauges—our homemade gauge, a commercial gauge, and the Myers Weather Center gauge. We found that the homemade gauge had maximum errors of 25% and 31% relative to the commercial and Weather Center gauges, respectively. Since the Weather Center gauge was sited closer to Walker Building and other obstructions, we hypothesize that reduced evaporation and increased turbulence likely occurred, resulting in some of the observed errors between each gauge. While the homemade gauge’s wind shield might have aided in increased catch, the transfer of water between the separated collection and measurement containers was a major deficiency in its design. To improve the design for future studies, it is recommended that the two containers need to be physically integrated, comprised of durable plastic materials, and accompanied by a sturdier, more effective wind shield. This study suggests that a single rain gauge is not reasonably representative of an entire city or region.
Accurate precipitation observations are essential in the verification of weather forecasts, the study of climate patterns, and the decision-making process across many industries. Here, we investigate the impacts of both rain gauge design and micrometeorological phenomena on precipitation measurement accuracy. To test inexpensive, nonstandard materials, a homemade rain gauge was built using a plastic funnel and glass jar for the collection container, a graduated cylinder for the measurement container, and plastic bucket for the wind shield. Precipitation was collected and measured daily from 14–27 September 2018 near Walker Building in University Park, Pennsylvania using three rain gauges—our homemade gauge, a commercial gauge, and the Myers Weather Center gauge. We found that the homemade gauge had maximum errors of 25% and 31% relative to the commercial and Weather Center gauges, respectively. Since the Weather Center gauge was sited closer to Walker Building and other obstructions, we hypothesize that reduced evaporation and increased turbulence likely occurred, resulting in some of the observed errors between each gauge. While the homemade gauge’s wind shield might have aided in increased catch, the transfer of water between the separated collection and measurement containers was a major deficiency in its design. To improve the design for future studies, it is recommended that the two containers need to be physically integrated, comprised of durable plastic materials, and accompanied by a sturdier, more effective wind shield. This study suggests that a single rain gauge is not reasonably representative of an entire city or region.
Benjamin_Yang_METEO440W_Paper.pdf | |
File Size: | 5459 kb |
File Type: |