IMPORTANCE OF BOUNDARY LAYER PROCESSES FOR SURFACE FLUXES
Erica McGrath-Spangler
PhD Defense20 October 2011
Acknowledgements
• Dr. Scott Denning• Drs. Dave Randall, Colette Heald, Dusanka
Zupanski, and Jennifer Hoeting• Denning group members• Data providers• Family and Friends
Outline• Introduction and Motivation•PBL Top Entrainment• Idealized Experiment• Case Study•PBL Depth from Space•Conclusions and Future Work
Introduction and Motivation
• Planetary boundary layer (PBL) • Turbulent layer closest to the Earth’s surface
• Determines surface fluxes of heat, moisture, momentum• Interacts with clouds, radiation, convection,
aerosols, pollutants• Transport of water vapor and momentum• Scalar quantities (CO2, H2O, heat) diluted by
depth of the PBL
Stull, 1988; Beljaars and Betts, 1993
PBL top
Free Atmosphere
Potential Temperature
CO2
Entrainment Zone
Idealized
• Response of CO2 mixing ratio to a surface flux is inversely proportional to the depth of the PBL• A deeper PBL will dilute
a flux signal relative to a shallower PBL• In models, need to get
the depth of the PBL right in order to get the CO2 mixing ratio correct
Shallow PBLHigh CO2
Deep PBLLow CO2
Another way to look at this
Full Glass => Deep PBL ½ Glass => Shallow PBL
Full Glass => Deep PBL
CO2 Flux CO2 Flux
½ Glass => Shallow PBL
Full Glass => Deep PBL ½ Glass => Shallow PBL
Small Impact Large Impact
Air Parcel
Air Parcel
Air Parcel
Inverse Modeling
Wind Wind
SourcesSinks
Sample Sample
Changes in CO2 in the airtell us about sources and sinks
But what if we don’t know the vertical
transport??
• The PBL depth is hard to observe• 1-2 km above the ground• Wind profilers and aircraft are
expensive • Limited spatially and temporally
• Radiosondes launched at the wrong times within the US (0 and 12 UTC)• Measure only one point in space/time
and may differ from average by up to 40%
Angevine et al., 1994
What happens when we try to model PBL depth?
• Sunny, midday PBL depths• June, July, and August• 2006-2010• Qualitatively similar• Very different
quantitative results!
500
1000
1500
2000
2500
3000
• Different methods exist to evaluate the PBL depth• Turbulent Kinetic Energy• Heat Diffusivity• Temperature profiles• Bulk Richardson number• PBL processes are often small-scale and not
resolvable by the models• Parameterizations exist, but these are based
on empirical data from idealized experiments
• Overshooting thermals produce entrainment of free tropospheric air into the PBL
• Parameterize the effects
PBL top
Free Atmosphere
Potential Temperature
CO2
Entrainment Zone
e.g. Stull 1988
Overshooting Thermals
Atmosphere
Land
Land-Atmosphere
Overshooting Thermals
Entrainment
Atmosphere
Land
Land-Atmosphere
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Atmosphere
Land
Land-Atmosphere
Entrainment
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Atmosphere
Land
Land-Atmosphere
Entrainment
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Dilution of Surface Fluxes
Atmosphere
Land
Land-Atmosphere
Entrainment
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Dilution of Surface Fluxes
Higher Daytime CO2
Atmosphere
Land
Land-Atmosphere
Entrainment
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Dilution of Surface Fluxes
Higher Daytime CO2
Stomatal Closing Decreased Cloud Cover
Atmosphere
Land
Land-Atmosphere
Entrainment
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Dilution of Surface Fluxes
Higher Daytime CO2
Decreased Cloud Cover
Decreased CarbonAssimilation
Atmosphere
Land
Land-Atmosphere
Decreased Latent Heat,Increased Sensible Heat
Greater Surface Heat Flux
Stomatal Closing
Entrainment
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Dilution of Surface Fluxes
Higher Daytime CO2
Decreased Cloud Cover
Decreased CarbonAssimilation
Atmosphere
Land
Land-Atmosphere
Decreased Latent Heat,Increased Sensible Heat
Greater Surface Heat Flux
Stomatal Closing
Entrainment
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Dilution of Surface Fluxes
Higher Daytime CO2
Decreased Cloud Cover
Decreased CarbonAssimilation
Atmosphere
Land
Land-Atmosphere
Decreased Latent Heat,Increased Sensible Heat
Greater Surface Heat Flux
Stomatal Closing
Entrainment
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Dilution of Surface Fluxes
Higher Daytime CO2
Decreased Cloud Cover
Decreased CarbonAssimilation
Atmosphere
Land
Land-Atmosphere
Decreased Latent Heat,Increased Sensible Heat
Greater Surface Heat Flux
Stomatal Closing
Entrainment
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Dilution of Surface Fluxes
Higher Daytime CO2
Decreased Cloud Cover
Decreased CarbonAssimilation
Changes in Surface Pressure,Precipitation Patterns, WindVelocity, etc
Atmosphere
Land
Land-Atmosphere
Decreased Latent Heat,Increased Sensible Heat
Greater Surface Heat Flux
Stomatal Closing
Entrainment
Overshooting Thermals
Weaker Capping Inversion Deeper PBL Warmer, Drier PBL
Dilution of Surface Fluxes
Higher Daytime CO2
Decreased Cloud Cover
Decreased CarbonAssimilation
Decreased Latent Heat,Increased Sensible Heat
Greater Surface Heat Flux
Changes in Surface Pressure,Precipitation Patterns, WindVelocity, etc
Changes in Large-ScaleWeather and Climate
Atmosphere
Land
Land-Atmosphere
Stomatal Closing
Entrainment
Idealized Experiment
• Simple Biosphere 3 – Regional Atmospheric Modeling System (SiB-RAMS)• Idealized experiment• Horizontally homogeneous• Cyclic boundary conditions• No cloud formation• 5x5 grid over WLEF tower in N. Wisconsin
McGrath-Spangler et al., 2009
Potential Temperature H2O Mixing Ratio
Hei
ght
Hei
ght
Kelvin g/kgMcGrath-Spangler et al., 2009
Controlα = 0.2
Controlα = 0.2
• Deeper PBL dilutes CO2 fluxes•Warmer, drier
conditions impact physiological stress factors• Shifts Bowen ratio• Changes
photosynthesis
CO2 Concentration
Local Time
ppm
McGrath-Spangler et al., 2009
Controlα = 0.2
Case Study
• July – September 1999• Fully 3D Model• Nudged lateral boundary conditions• North America domain with 40 km grid
intervals• Initialized from reanalysis
McGrath-Spangler and Denning, 2010
McGrath-Spangler and Denning, 2010
7 ppm
McGrath-Spangler and Denning, 2010
PBL Depth from Space• Cloud-Aerosol LIDAR and
Infrared Pathfinder Satellite Observations (CALIPSO)• Launched April 2006, First light
June 2006• Part of the Afternoon
(A-train) constellation of satellites
e.g. Winker et al., 2007
• CALIOP LIDAR is sensitive to aerosols and clouds• Lidar measures the scatter of a laser beam off of
objects such as aerosols, clouds, and the ground• CALIOP has a horizontal resolution near the
surface of 0.33 km and a vertical resolution of 30 m• Emits light at 1064 nm (infrared) and polarized
light at 532 nm (green)• First satellite lidar optimized for aerosol and
cloud measurements• CALIOP acquires 1.7 million laser shots every 24
hours
e.g. Winker et al., 2007; 2009
PBL from SpaceRocky MountainsOcean Mexican Plateau
Altit
ude
(km
)
•Method initially developed by Jordan et al. (2010) • Identifies a local maximum in LIDAR 532 nm
backscatter collocated with a maximum in the variance in the backscatter• Entrainment zone has mixing of aerosol-
laden PBL air mixing with clean, free tropospheric air or capped by boundary layer clouds
• Search only between 250 m and 5 km AGL• Remove surface noise• Remove unclear aerosol signatures
• Reject profiles attenuated by thick clouds• Identified by high values in the 1064 nm
backscatter• Boundary layer clouds accepted
• Accept only easiest to retrieve
Backscatter
Variance
Retrieved PBL
km-1 sr-1
Hei
ght (
km)
CALIPSO 532nm Total Attenuated Backscatter
km-1 sr-1
CALIPSO 532nm Total Attenuated Backscatter
Hei
ght (
km)
Backscatter
Variance
Latitude, Longitude
PBL DepthSurface
Yucatán Peninsula Little Rock, AR Winnipeg, Manitoba
CleanProfile
5000 10002500 7500 12500
PBL Height (m)
25 50 75
• Success decreased by cloud cover and unclear aerosol signature• Greatest success over
water• Reduced success over
highly convective regions• Florida and the Gulf
Coast• Rocky Mountains and
Mexican Plateau
• Success ranges from 15% to near 100%
1000 1500 2000 2500
• Deeper PBL over land• Shallow PBL off
California coast• Shallow PBL over
Midwest farmlands• Deeper PBL over
Rocky Mountains and Mexican Plateau• Deeper PBL over
Canada - stomatal control and longer day length
100 400 700 1000
• Lowest variability over water• Highest variability
along Rocky Mountains and boreal North America• Standard deviation of
almost 1 km here in Colorado!
• MERRA reanalysis• At times and
locations of satellite overpass• Above 55°N,
daytime PBL not yet developed
500 1000 1500 2000 2500 3000
MERRA is deeper CALIPSO is deeper25 50 75 100 125 150 175
• Compares MERRA reanalysis to CALIPSO• Over much of US,
CALIPSO and MERRA give similar results• Over SW US, MERRA
deeper• Off California coast and
boreal Canada, CALIPSO deeper• Boreal MERRA PBL not fully
developed by time of overpass
• Off CA coast, CALIPSO detects stratocumulus cloud top
Relevance to Inversions• If observed PBL is shallower than simulated, inversion
adjustment should be decreased• Model CO2 is too high, inversion increases photosynthesis• Decrease PBL depth, photosynthesis is more concentrated so in
new inversion, increase photosynthesis less• Southwestern United States
• If observed PBL is deeper than simulated, inversion adjustment should be increased• Model CO2 is too high, inversion increases photosynthesis• Decrease PBL depth, photosynthesis is more dilute so in new
inversion, increase photosynthesis more• Boreal Canada
Conclusions and Future Work
• PBL depth is important for carbon budget studies, especially inversion studies• Inaccurate PBL depths produce inaccurate
CO2 mixing ratios, even for perfect surface fluxes•Millions of satellite observations can be used
to determine PBL depth and constrain model simulations• Initial estimates are qualitatively reasonable
• Success of the retrieval is greatest over subtropical oceans• General success over land is about 50%• Decreased success over mountainous United
States and convective regions• Underprediction of PBL depths by reanalyses
over oceans• Overprediction by reanalyses over SW United
States
Zupanski et al. 2007
Atmospheric CO2
FoutFin
Future Work• Data Assimilation• Optimize entrainment fluxes using well-observed
variables (T, Td, etc.) and space-borne measurements of PBL depth• Use CO2 observations to optimize CO2 fluxes
FT(x,y,t) = βRESP(x,y)*Fin(x,y,t) – βGPP(x,y)*Fout(x,y,t)
• CALIPSO PBL Depth• Compare to and evaluate against other
observations• Extend analysis spatially and temporally• Compare CALIPSO PBL depth to aerosol
distributions in models
Thank you.
Questions?
References• Angevine, W. M., A. B. White, and S. K. Avery (1994), Boundary-layer depth and entrainment zone characterization with a boundary-layer
profiler, Boundary-Layer Meteorology, 68(4), 375-385.
• Beljaars, A. C. M., and A. K. Betts (1993), Validation of the boundary layer representation in the ECMWF model, paper presented at Seminar Proceedings on Validation of Models over Europe, ECMWF, Reading, England.
• Corbin, K. D., A. S. Denning, E. Y. Lokupitiya, A. E. Schuh, N. L. Miles, K. J. Davis, S. Richardson, and I. T. Baker (2010), Assessing the impact of crops on regional CO2 fluxes and atmospheric concentrations, Tellus B, 62(5), 521-532.
• Jordan, N. S., R. M. Hoff, and J. T. Bacmeister (2010), Validation of Goddard Earth Observing System-version 5 MERRA planetary boundary layer heights using CALIPSO, J. Geophys. Res., 115(D24), D24218.
• Lokupitiya, E., S. Denning, K. Paustian, I. Baker, K. Schaefer, S. Verma, T. Meyers, C. J. Bernacchi, A. Suyker, and M. Fischer (2009), Incorporation of crop phenology in Simple Biosphere Model (SiBcrop) to improve land-atmosphere carbon exchanges from croplands, Biogeosciences, 6(6), 969-986.
• McGrath-Spangler, E. L., A. S. Denning, K. D. Corbin, and I. T. Baker (2009), Sensitivity of land-atmosphere exchanges to overshooting PBL thermals in an idealized coupled model, Journal of Advances in Modeling Earth Systems, 1(Art. #14), 13 pp.
• McGrath-Spangler, E. L., and A. S. Denning (2010), Impact of entrainment from overshooting thermals on land–atmosphere interactions during summer 1999, Tellus B, 62(5), 441-454.
• Okamoto, H., et al. (2007), Vertical cloud structure observed from shipborne radar and lidar: Midlatitude case study during the MR01/K02 cruise of the research vessel Mirai, J. Geophys. Res., 112(D8), D08216.
• Stull, R. B. (1988), An introduction to boundary layer meteorology, 666 pp., Kluwer Academic Publishers, Norwell, MA.
• Winker, D. M., W. H. Hunt, and M. J. McGill (2007), Initial performance assessment of CALIOP, Geophys. Res. Lett., 34(19), L19803.
• Zupanski, D., A. S. Denning, M. Uliasz, M. Zupanski, A. E. Schuh, P. J. Rayner, W. Peters, and K. D. Corbin (2007), Carbon flux bias estimation employing maximum likelihood ensemble filter (MLEF), J. Geophys. Res.-Atmos., 112(D17).
Top Related