Opportunity Title:

Advanced Statistical Analysis and Modeling of Data from Mobile Monitoring Designed to Map Major Population Areas in California

Research Area:

Air

EPA Lab/Center/Office:

National Exposure Research Laboratory (NERL)

Location:

Las Vegas, NV

Duration:

12 months

Brief Summary:

The student will have the opportunity to collaborate with a team that is working in an area that is truly state-of-the-science forging new directions and pushing the envelope in air pollution measurements, monitoring, data analysis, storage and communication. The research involves statistical analysis of 1 s air pollution and environmental data collected using mobile platforms in three major California cities, developing performance characteristics and conducting complex spatial and temporal analysis.

Opportunity Description:

The goal of this effort is to analyze air quality data collected in the San Francisco Bay Area (SF), the Los Angeles (LA) metropolitan area, and the Central Valley of California (CVC) during 2016 and 2017.

The overall project includes review of, and as needed application of statistics used to develop performance characteristics for reference and sensor data, statistical approaches for adjusting data for calibration factors and zero drift, statistics used to adjust sensors for T and RH effects relative to reference methods or laboratory data, statistics used to remove outliers, removal of peaks to obtain an estimate of background concentrations, and perform statistical analysis of spatial and high-time resolution temporal data obtained by the multiple mobile platforms employed in these studies. Google or Alphabet are the key supporters of this project and the mobile platforms are Google Street View cars.

Topics to address include but are not limited to:

1. Determine the analytical performance characteristic of pollutant measurements for reference and sensor instruments from laboratory and field calibrations. 
2. How do measurements from sensor nodes (mounted in the vehicles, stationary and moving) compare to measurements from reference samplers located within the vehicles? How do measurements from sensor nodes and reference instruments integrated on a mobile monitoring platform compare with stationary surface measurements while moving and stationary?
3. Spatial and temporal variation of pollutants associated with specific sources and receptors as determined from mobile and stationary monitoring. Example points of interest:
   1. Environmental Justice (EJ), non-EJ and associated neighborhoods, and industrial parks
   2. Freeways, including bridges and tunnels, and the effect of elevation off of the freeway
   3. Near roadway sites and associated neighborhoods, including schools in the area
   4. Downtown, ports (LA & SF), airports and distribution centers.
   5. Major agricultural areas (feedlot, etc), rural areas and influence from cities, Central Valley
   6. Oil production (Central Valley)
   7. Examine the influence of road barriers, including vegetative and brick walls

One goal is to analyze the data to develop multiple papers addressing similar topics as:

• Pub.1.  Study design and performance evaluation results that include, for example, comparisons among reference methods located in each car when mobile and between the cars and stationary sites, and between reference methods and sensors within each car. Laboratory evaluations conducted before and after the summer study as well as zero and span checks will be described and included in the paper. An examination of integration time in comparisons should be included in the analysis. This should address EPA’s science questions 1 and 2.
• Pub.2   Spatial analysis that addresses the near, middle, and far (within 5 km) distance from the central stationary sites to examine the spatial representativeness of the central site both after adjustment of background (what is the regional representativeness of the central site with peak values removed?) and actual representativeness of the central site when mobile data include local source emissions.  An examination of integration time in comparisons should be included in the analysis.
• Pub. 3.  Evaluate the zone of influence of identified source areas (e.g., freeways, ports, airport, industrial areas), examine pollutant and information gained by comparing EJ Areas to non-EJ areas, evaluate how the mobile monitoring provides increased data related to pollutant transport in street canyons, and others as identified in the science questions).  Also, compare near roadway site data with local mobile measurements around the within 10 m – 1 km of the near roadway site.  An examination of integration time in comparisons should be included in the analysis.

Data are unlike any mobile monitoring program to date, currently covering almost 90 days of 8 hr data collection at a 1 s sampling frequency for multiple air pollutants and environmental parameters.

The intern would have the opportunity to co-author multiple journal articles.

Opportunities for Professional Development:

The databases collected are unique in size, duration and number of parameters obtained at such a high frequency. Over 5M data points have been obtained that allow for the student to enhance skills in statistics, data mining, atmospheric chemistry, etc. The student will work directly with the EPA mentor, a senior research scientist with expertise in air chemistry measurements and analysis and with a full professor in statistics at UNLV.

There will be significant area for professional growth and experience by working with "Big Data" or huge databases that require new skills in programming to allow for statistical analysis of pollutant data that is moving in space over time - this is a challenging area where few people have expertise. The student will have access to the EPA facilities in Las Vegas, perhaps have the opportunity to visit the field locations and meet with a private company that is overseeing the mobile platforms, and have the opportunity to work in an area that is truly state-of-the-science forging new directions and pushing the envelope in air pollution measurements, monitoring, data analysis, storage and communication.

Point of Contact or Mentor:

Surender Kaushik (kaushik.surender@epa.gov)