Algorithm to Infer Vertical Columns from Total Slant Columns Randall Martin Dalhousie SAO with contributions from Nick Krotkov NASA Goddard Lok Lamsal NASA Goddard ID: 777195
Download The PPT/PDF document "Development of TEMPO NO 2" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Development of TEMPO NO2 Algorithm to Infer Vertical Columns from Total Slant Columns
Randall Martin (Dalhousie, SAO)with contributions fromNick Krotkov (NASA Goddard), Lok Lamsal (NASA Goddard), Jintai Lin (Peking U), Chris McLinden (Environment Canada), Caroline Nowlan (SAO)
TEMPO Science Team MeetingHuntsville, Alabama27 May 2015
Slide2Attention Needed to Removal of Stratospheric NO2Adaptation to OMI algorithms to TEMPO will require consideration of diurnal variation and boundaries (more from Lok Lamsal)
Annual mean, from OMI (2009) Fraction of total NO2 column in the troposphere can be smallUrban/Industrial areas: 30-80%Rural/background areas: 10-30%FractionFigure from Chris McLindenTropospheric Fraction
Stratospheric Assimilation May Offer Advantages
Slide3Need to Account for Atmospheric Scattering and Vertical Profile (Air Mass Factor; AMF) in UV-Vis RetrievalsOften Largest Source of Uncertainty in LEO Retrievals
dt()Io
I
B
EARTH SURFACE
Radiative Transfer Model
Scattering weight
Atmospheric Chemistry Model
“a-priori” Shape factor
Calculate
w
(
) as function of:
solar and viewing zenith angle
surface
reflectance,
pressure
cloud pressure,
aerosol
INDIVIDUAL
TEMPO SCENES
Mixing
ratio
C(
)
() is temperature dependent cross-section
sigma (
)
Slide4Challenges in Development of AMF Calculation for TEMPO
Representing diurnal variation in surface reflectance at launch (use GOES-R spatially & TROPOMI spectrally)Diurnal variation in trace gas profile (changes rapidly in morning; need to evaluate; more from Lok Lamsal)Resolve geophysical fields at fine resolution (e.g. BEHR)Accounting for aerosols (still unresolved for LEO). Existing cloud products partially account for aerosolSurface reflectance (BRDF & snow)
Concentration
A
ltitude
Slide5Current Operational LEO AMF Algorithms Treat Aerosol as Cloud: Prone to Bias when Together Including Aerosols in Retrievals of Both Clouds and Trace Gases Can Consistently Account for Their Effects
Clouds Decrease Sensitivity to Trace Gas Below
Aerosols Increase Sensitivity to
Trace Gas
Within Haze Layer
“Average” Tends to Decrease
Sensitivity to Trace Gas
Below
Accounting for Effects of Aerosol Haze on Cloud and NO
2
Increases R
2
(0.72
0.96)
of OMI NO
2
vs Ground-based DOAS Observations in China
(Lin et al., ACP, 2014)
Slide6Should Account for Bidirectional Reflectance Distribution Function (BRDF)
Sun Behind ObserverSun Opposite Observer
Can introduce changes in retrieved NO
2
of +/- 50% (Lin et al., ACPD, 2015)
Information available from MODIS and forthcoming from GEOS-R
Will also be important over snow
AMF Sensitive to Surface
Figure from
Lok
Lamsal
Slide7Address Potential Retrieval Bias for Snow-Covered ScenesCreate Separate Reflectance Databases for Snow-Free and Snowy ScenesSnow Identification: IMS (NOAA/NESDIS) or CaLDAS (Env. Canada)
O’Byrne et al., JGR, 2010
With Cloud
Fraction
Threshold (f < 0.3)
-0.5
0
1.0
All
Cloud
Fractions
0.5
Slide8Plan for TEMPO Algorithm ImplementationBasic algorithm (default standard algorithm)empirical stratospheric removalAMF without aerosolsLambertian reflectanceCombined snow-free & snow-covered reflectance…Research algorithm (components become standard when mature)stratospheric assimilationAMF with aerosols
BRDFSeparate snow-free & snow-covered reflectance…
Slide9Slide10Snow CoverBest productsIMSCMC CaLDASProvider
NOAA/NESDISEnvironment Canada / Canadian Meteorological CentreAvailabilityNear-real timeNear-real timeSpatial ExtentNorthern HemisphereNorth America / GlobalSpatial resolution (current)4 x 4 km210 x 10 km2 / 24 x 24 km2Spatial resolution (future)1 x 1 km22.5 x 2.5 km2 / 10 x 10 km2 (~2015/2016)Temporal resolutionCurrent: daily; future: 12-hourCurrent: 12-hour; future: 6 hour or better
Field providedSnow extent (yes / no)Snow depth*Input informationsatellite imagery; derived mapped products; surface observationsCMC: analysis using surface observationsCaLDAS: Data assimilation of land-surface model, satellite imagery; surface observations
* Could be used to identify fresh snow
Figure from Chris
McLinden