1

Jim Crawford1, Ken Pickering2, Lok Lamsal2, Bruce Anderson1, Andreas Beyersdorf1, Gao Chen1, Richard Clark3, Ron Cohen4, Glenn Diskin1, Rich Ferrare1, Alan Fried5, Brent Holben2, Jay Herman6, Ray Hoff6, Chris Hostetler1, Scott Janz2 , Mary Kleb1, Jim Szykman7, Anne Thompson2, Andy Weinheimer8, Armin Wisthaler9, Melissa Yang1, Jay Al-Saadi1

1 NASA Langley Research Center, 2 NASA Goddard Space Flight Center, 3 Millersville University, 4 University of California-Berkeley, 5 University of Colorado-Boulder, 6 University of Maryland-Baltimore County, 7 Environmental Protection Agency, 8 National Center for Atmospheric Research, 9 University of Innsbruck

Challenges and opportunities for remote sensing of air quality: Insights from DISCOVER-AQ

http://discover-aq.larc.nasa.gov/

Maryland Department of the Environment (MDE)San Joaquin Valley Air Pollution Control District (SJV APCD)California Air Resource Board (CARB)Bay Area Air Quality Management District (BAAQMD)Texas Commission on Environmental Quality (TCEQ)Colorado Department of Public Health and Environment (CDPHE)

Environmental Protection Agency, Office of Res. and Dev.National Center for Atmospheric ResearchNational Science FoundationNational Oceanic and Atmospheric AdministrationNational Park Service

University of Maryland, College Park; Howard UniversityUniversity of California, Davis; University of California, IrvineUniversity of Houston; Rice University; University of Texas; Baylor University; PrincetonUniversity of Colorado-Boulder; Colorado State University

Thanks to Partners

DDeriving eriving IInformation on nformation on SSurface Conditions from urface Conditions from CoColumnlumn and and VERVERtically Resolved Observations Relevant to tically Resolved Observations Relevant to AAir ir QQualityuality

A NASA Earth Venture campaign intended to improve the interpretation of satellite observations to diagnose near-surface conditions relating to air quality

Objectives:

1. Relate column observations to surface conditions for aerosols and key trace gases O3, NO2, and CH2O

2. Characterize differences in diurnal variation of surface and column observations for key trace gases and aerosols

3. Examine horizontal scales of variability affecting satellites and model calculations

Investigation Overview

3

Deployment Strategy

Systematic and concurrent observation of column-integrated, surface, and vertically-resolved distributions of aerosols and trace gases relevant to air quality as they evolve throughout the day.

4

NASA UC-12 (Remote sensing)Continuous mapping of aerosols with HSRL and trace gas columns with ACAM

NASA P-3B (in situ meas.)In situ profiling of aerosols and trace gases over surface measurement sites

Ground sitesIn situ trace gases and aerosolsRemote sensing of trace gas and aerosol columnsOzonesondesAerosol lidar observations

Three major observational components:

5

Deployment Locations

Maryland, July 2011 Houston, September 2013

California, Jan-Feb 2013 Colorado, Jul-Aug 2014

6Taken from Fishman et al., BAMS, 2008

Predicted NO2 Column Behavior

7Taken from Boersma et al., JGR, 2008

Emissions

Photochemical Loss

NO2 ColumnNO2:NOx

Predicted NO2 Column Behavior

8

1 x 1015 mol/cm2 = 0.037 DU

Pandora Statistics-Maryland

Median and inner quartile values plotted

9

P-3B Average Profiles-Maryland

10

Pandora Statistics-Houston

1 x 1015 mol/cm2 = 0.037 DU

Median and inner quartile values plotted

11

Pandora Statistics-California

1 x 1015 mol/cm2 = 0.037 DU

Median and inner quartile values plotted

12

P-3B Profile Statistics-CaliforniaUrban sites: Bakersfield+Fresno

13

P-3B Profile Statistics-CaliforniaSub-Urban sites

14

Pandora Statistics-Colorado

1 x 1015 mol/cm2 = 0.037 DU

Median and inner quartile values plotted

15

Summary

1. DISCOVER-AQ has collected a dataset of unprecedented detail on the diurnal trends in air quality as it is discerned from in situ and remote sensing methods.

2. NO2 columns exhibit both unexpected and diverse diurnal trends that are consistent with vertically resolved profiles.

3. NO2 tropospheric column retrievals are highly sensitive to diurnal variation in a-priori profile shapes.

4. Next analysis steps include looking beyond median statistics.

16

Implications for TEMPO (1 of 2)

• An airborne validation campaign is probably beyond TEMPO’s budget

• D-AQ core budget $30M for 5-year, 4-campaign study; partner contributions added ~$10M

• Typical R&A-directed campaign budgets $15-$20M (3-yr total)• 1-month GeoTASO/GCAS deployment ~$1M for flights & data

processing

• There’s no guarantee that an R&A airborne campaign will take place over North America during TEMPO prime mission

• NASA R&A led airborne campaigns for atmospheric composition historically occur every 2-4 years; have to take turns with other science focus areas

• Competing resource demands, within atmospheric composition and with EV-S

17

Implications for TEMPO (2 of 2)

• So what do we do?• Science studies related to TEMPO observations are a logical

priority for tropospheric chemistry programs post-launch• Analysis of DISCOVER-AQ, KORUS-AQ, TROPOMI and other

data sets will help clarify priorities for pre- & post-launch airborne campaigns related to TEMPO

• Spatial representativeness, diurnal influences on products, vertical profile shapes, …

• Get in line and build advocacy: NASA campaign concepts typically start as community grass-roots efforts leading to development of white papers

• Consider Earth Venture proposal?

• Continue preparatory activities as funding permits (GEO-CAPE, HQ R&A, possibly HQ Applied Science, possibly TEMPO)

18

Thoughts from Jim Crawford on Possible Approaches

• There are at least two initial approaches to take.  

• One is to leverage what we have learned from DISCOVER-AQ, KORUS-AQ, etc. to define a field campaign to support TEMPO.  In this case, you would hope to be able to get airborne as soon as possible after TEMPO launches.  

• Another approach would be to get the right surface measurements in place and use them to identify the locations where TEMPO needs the most help. This would delay a field campaign in favor of getting a chance to evaluate TEMPO performance using ground obs, sondes, Pandora, TROPOMI, etc.  Such a campaign would hopefully be more targeted on TEMPO performance, but would also hope to see TEMPO observations continue beyond the initial 2-years to take advantage of what is learned.

Backups

20

P-3B Profile Statistics-CaliforniaUrban sites: Bakersfield+Fresno

21

Pandora vs Surface-Colorado

22Taken from Tzortziou et al., JGR, 2014

Observed NO2 Column Behavior

CMAQ model: Three simulationsHorizontal resolution 4 km x 4 km

Vertical levels 45 (surface-100 hPa)

Domain Washington-Baltimore

Chemical mechanism CB05

Aerosols AE5

Dry deposition M3DRY

Vertical diffusion ACM2

Chemical and initial boundary condition

RAQMS; 12 km x 12 km

Biogenic emissions Calculated within CMAQ with BEIS

Biomass burning emissions FINNv1

Lightning emissions Calculated within CMAQ

Anthropogenic emissions NEI-2005 projected to 2012

1. ACM2 Base 2. ACM2 Mod 3. YSU ModPBL scheme ACM2 ACM2 YSU

Mobile emissions Standard Reduced 50% Reduced 50%

Alkyl nitrate photolysis Standard 10 times faster 10 times faster

► Two PBL schemes selected based on the study by Clare Flynn

► Emissions and photolysis rate changed based on Anderson et al., 2014 and Canty

et al., 2014

Remote Sensing Column Air Mass Factor Sensitivity:Observations and Methods

► Location: Padonia, Maryland

► Observation period: 3-4 spirals for 14 days in July 2011 (Hours covered 6

AM – 5 PM, local time)

► NO2 observations:

Aircraft (P3B) measurements (200 m - ~4 km) NCAR data (accuracy

better than10%) Surface measurements by photolytic converter instrument (accuracy

better than 10%) Spatial resolution comparable between model (4x4km) and spiral

(radius ~4km)

► Observed PBL heights : Estimation based on temperature, water vapor,

O3 mixing ratios, and RH (Donald Lenschow)

► Methods: Model and surface measurements sampled for the days and time of

aircraft measurements Spiral data sampled at model vertical grids

Comparison of NO2 profiles and shape factors

i

i

ifFilled circles: observed grid

average

Open circles: linear interpolation in log space

Error bars: standard deviation

h

surface

ii fwAMF

fi NO2 shape factors

Ωi partial columnwi scattering weights (VLIDORT)

AMF (Observation) = 1.94AMF (Base) = 2.16AMF (ACM2 Mod) = 2.3AMF (YSU Mod) = 1.8

Errors in AMFs/retrievals from a-priori NO2 profiles

km

surface

ii fwAMF8

wi scattering weightsfi NO2 shape factors

► AMF calculated for ACAM (air-borne spectrometer

located at ~8km) but is also relevant for tropospheric

NO2 column retrievals

Surface reflectivities: 0.1 to 0.14 at 0.01

steps

Solar zenith angles: 10° to 80° at 10°

steps

Aerosol optical depths: 0.1 to 0.9 at 0.1

steps

NO2 profiles and AMFs (11 AM)

AMF (Observation) = 1.94AMF (Base) = 2.16AMF (ACM2 Mod) = 2.3AMF (YSU Mod) = 1.8