Boundary Layer Verification

ECMWF training course

May 2010

Maike Ahlgrimm

What does the BL parameterization do?

Attempts to integrate effects of small scale turbulent motion on prognostic variables at grid resolution.

Turbulence transports temperature, moisture and momentum (+tracers).

Ultimate goal: correct model output

Which aspect of the BL can we evaluate?

1. 2m temperature/humidity2. Depth of BL3. Diurnal variability of BL height4. Structure of BL (temperature, moisture,

velocity profiles)5. Turbulent transport within BL6. Boundaries: entrainment, surface fluxes,

clouds etc.

large scale

small scale

Chandra et al., sub. to J. Climate

Part 1

Depth of the boundary layer

BL depth from radiosondes

• Problem: Define the top of the BL!

• Concept: At he top of the BL, the air motion transitions from turbulent to laminar flow.

• For an equitable comparison, apply the same criteria for identification of this transition to model profiles and radiosonde profiles.

• Alternative for convectively driven boundary layers: turbulent mixing leads to T and q gradients at the BL top. Identify these gradients in the profile.

DSE/cpFigure: Martin Köhler no

rmal

ized

BL

hei

ght

Richardson number-based approach

• Richardson number defined as:

• flow is turbulent if Ri is negative• flow is laminar if Ri above critical value• calculate Ri for model/radiosonde profile

and define BL height as level where Ri exceeds critical number

buoyancy production/consumptionshear production (usually negative)

Ri=

Difficulties with this approach

• discrete model layers -> bulk Ri number• where is the top and bottom of the bulk layer?• how much do surface fluxes increase buoyancy?

not most reliable model field• for sonde profiles, surface fluxes usually

unavailable• noise in sonde profiles can introduce uncertainties

diagnostic BLH in IFS is currently “tuned” to best agree with paramete-rization based BL height

How-to

• Need T, u,v,q,z and some constants

• Define conserved variable, e.g. virtual dry static energy:

• Apply smoothing in the vertical if necessary

• Starting at lowest model level, calculate Ri number, adding an excess to the dse to make up for missing surface fluxes

• Iterate, until Ri exceeds critical level (e.g. 0.25)

• Assign height of nearest layer as BL top height

Example: dry convective boundary layer NW Africa

2K excess

1K excess

Theta [K] profiles shiftedFigures: Martin Köhler

Example: Inversion-topped BL

• Inversion capped BLs dominate in the subtropical oceanic regions

• Identify height of jump across inversion

EPIC, October 2001southeast Pacific

Limitations of sonde measurements

• Sonde measurements are limited to populated areas

• Depend on someone to launch them (cost)• Model grid box averages are compared to point

measurements (representativity error)

Took many years to compile this map

Neiburger et al.1961

Calipso tracks

Arabic peninsula - daytimeArabic peninsula - daytime

CALIPSO tracks

BL from lidar how-to

• Easiest: use level 2 product (GLAS)

• Algorithm searches from the ground up for significant drop in backscatter signal

• Align model observations in time and space with satellite track and compare directly, or compare statistics

surface return

backscatter from BL aerosol

molecular backscatter

Figure: GLAS ATBD

Example: Lidar-derived BL depth from GLAS

Only 50 days of data yield a much more comprehensive picture than Neiburger’s map.

Ahlgrimm & Randall, 2006

Limitations to this method

• Definition of BL top is tied to aerosol concentration - will pick residual layer

• Does not work well for cloudy conditions (excluding BL clouds), or when elevated aerosol layers are present

• Overpasses only twice daily, same local time• Difficult to monitor given location

The case of marine stratocumulus

• Well mixed convective layer underneath strong inversion

• Are clouds part of the BL?• As Sc transition to trade cumulus, where is the BL

top?

Stratocumulus cloud top height

Model underestimates Sc top height

Köhler & Ahlgrimm, sub. Hannay et al. 2009

EPIC

SEP

Part 2

Diurnal cycle of boundary layer height

Diurnal cycle of convective BL from radiosonde

Example: stratocumulus-topped marine BL in the south-east Pacific: East Pacific Investigation of Climate (EPIC), 2001

Clear diurnal cycle of ~200m with minimum in early afternoon, maximum during early morning.

Bretherton et al. 2004, BAMS

Diurnal cycle from CALIPSO

Part 3

Turbulent transport

Flux towers

• Example: Cabauw, 213m mast• obtain measurements of roughness

length, drag coefficients etc.

KNMI webpageKNMI webpage

Bomex: trade cumulus regime

Stevens et al. 2001Stevens et al. 2001

Bomex - DualM

• Dual Mass Flux parameterization - example of statistical scheme mixing K-diffusion and mass flux approach

• Updraft and environmental properties are described by PDFs, based on LES

• Need to evaluate PDFs!

Neggers et al. 2009

Turbulent characteristics: humidity

Raman lidar provides high resolution (in time and space) water vapor observations

Plot: Franz Berger (DWD)

Turbulent characteristics: vertical motion

Observations from mm-wavelength cloud radar at ARM SGP, using insects as scatterers.

Chandra et al., sub. to J. Climate local time

reflectivity

reflectivity

doppler velocity

red dots: ceilometer cloud base

Turbulent characteristics: vertical motion

Variance and skewness statistics in the convective BL (cloud free) from four summer seasons at ARM SGP

Chandra et al., sub. to J. Climate

Part 4

Boundaries

Forcing

• BL turbulence driven through surface fluxes, or radiative cooling at cloud top.

• Check: albedo, soil moisture, roughness length, clouds

• BL top entrainment rate: important but elusive quantity

Entrainment rate - DYCOMS II

Example: DYCOMS II - estimate entrainment velocity

mixed layer concept:

Stevens et al. 2003

Summary & Considerations

• What parameter do you want to verify?

• What observations are most suitable?

• Define parameter in model and observations in as equitable and objective a manner as possible.

• Compare!

• Are your results representative?

• How do model errors relate to parameterization?

References (in no particular order)

• Neiburger et al.,1961: The Inversion Over the Eastern North Pacific Ocean• Bretherton et al., 2004: The EPIC Stratocumulus Study, BAMS• Stevens et al., 2001: Simulations of trade wind cumuli under a strong inversion, J.

Atmos. Sci.• Stevens et al., 2003: Dynamics and Chemistry of Marine Stratocumulus - DYCOMS

II, BAMS• Chandra, A., P. Kollias, S. Giangrande, and S. Klein: Long-term Observations of

the Convective Boundary Layer Using Insect Radar Returns at the SGP ARM Climate Research Facility, submitted to J. Climate

• Hannay et al., 2009: Evaluation of forecasted southeast Pacific stratocumulus in the NCAR, GFDL, and ECMWF models. J. Climate

• Köhler et al.: Stratocumulus in the ECMWF model. submitted to QJRMS• Ahlgrimm & Randall, 2006: Diagnosing monthly mean boundary layer properties

from reanalysis data using a bulk boundary layer model. JAS• Neggers, 2009: A dual mass flux framework for boundary layer convection. Part II:

Clouds. JAS