Results of the Town Energy Budget (TEB) coupled to the Regional Atmospheric Modeling System (RAMS) over Washington DC

1
Results of the Town Energy Budget (TEB) coupled
to the Regional Atmospheric Modeling System
(RAMS) over Washington DC

• 15 January 2009

Distribution A Approved for Public Release,
Distribution Unlimited
2
Overview

• Motivation and Effort
• Models and Data
• Model set up
• Results
• Summary and Conclusions

Distribution A Approved for Public Release,
Distribution Unlimited
3
Motivation

• UHI Effect is a well established phenomenon
• Impacts sensible weather (e.g. PBL) of urban area
• Studies have shown that the UHI can even
  interact/alter local mesoscale flow regimes
• Motivated the creation of parameterizations in
  models however, studies done at high (lt2km)
  resolution
• Current operational models use grid resolutions
  of 4-12km, thus capturing mesoscale flow
  regimes
• Primary tool used for sensible weather forecasts
• Using land surface models not designed to model
  UHI
• PBLs over urban areas likely not well
  represented
• Mesoscale flow interactions not possible

Distribution A Approved for Public Release,
Distribution Unlimited
4
Motivation

• Studies have shown Air quality, dispersion, and
  military TDAs are sensitive to PBL parameters
• Used heavily in urban areas
• Often driven by operational meteorological NWP
• In addition, in 2000 nearly 50 of the world
  population is urban and expected to grow
  significantly in years to come (Cohen,2003)
• Thus any environmental impact of UHI will be felt
  by a disproportionate part of the population
• Result Current operational forecasts in/around
  urban centers not accounting for UHI effects

4
Distribution A Approved for Public Release,
Distribution Unlimited
5
This Effort

• Primary question of my work Can an urban
  parameterization improve simulation of a PBL over
  an urban area (Wash DC) using an typical
  operational model set-up (e.g. 4-12km, fewer
  vertical layers)
• Secondary Questions (not addressed in this talk)
• What are the primary sensitivities of the coupled
  system with an eye towards issues to
  operationalization
• How detailed does morphology information need to
  be?
• How detailed does land surface information need
  to be?
• Are the differences between the simulated PBLs
  significant to follow-on applications (e.g.
  dispersion, military TDAs)

5
Distribution A Approved for Public Release,
Distribution Unlimited
6
This Effort

• Used RAMS V4.3 coupled to LEAF-2 Land Surface
  Model with a typical operational setup
• Urban Parameterization Town Energy Balance
  (TEB) Model, Masson (2000)
• Well documented/validated medium complexity
  parameterization
• Completed Rozoff (2003) coupling to LEAF-2
• Created a 1km morphology database

Description Option
Grid Structure Arakawa C Grid 3 fixed nested (80, 20, 5 km)
PBL Parameterization Mellor-Yamada
Radiation Parameterization Chen, both LW and SW
Lower Boundary LEAF-2 with Town Energy Balance
Lateral Boundary Klemp/Wilhelmson
Level of First Model Layer 23 Meters (11m for those written to center)
6
Distribution A Approved for Public Release,
Distribution Unlimited
7
Models and Data

• Performed modeling over Washington DC for three
  days in 1984
• 1984 was chosen due to a year long field campaign
  called the Metropolitan Tracer Experiment
  (METREX) that provided additional sources of met
  data
• Present work from 26 Jun 1984 a warm summer
  day/night with SSW flow on the back side of
  departing high pressure

7
Distribution A Approved for Public Release,
Distribution Unlimited
8
Models and Data

• Meteorology data was available for a few sites
  around DC
• Importance was placed on obtaining both surface
  and elevated data

8
Distribution A Approved for Public Release,
Distribution Unlimited
9
Subjective Results

• What did the addition of TEB do to the simulation
  and how does it compare with what we would
  anticipate?
• 26 Jun 84

26 Jun 2200L Temperature (C) TEB No TEB
9
Distribution A Approved for Public Release,
Distribution Unlimited
10
Subjective Results

• 26 Jun 84

27 Jun 0000L Wind Speed (m/s) TEB No TEB
26 Jun 1300L Wind Speed (m/s) TEB No TEB
10
Distribution A Approved for Public Release,
Distribution Unlimited
11
Subjective Results

• 26 Jun 84

26 Jun 2200L PBL Height TEB No TEB
11
Distribution A Approved for Public Release,
Distribution Unlimited
12
Subjective Results

• 26 Jun 84

26 Jun 2200L Streamlines No -TEB
26 Jun 2200L Streamlines TEB
12
Distribution A Approved for Public Release,
Distribution Unlimited
13
Objective Results

• What did the addition of TEB do to the simulation
  and how does it compare with what was measured?

13
Distribution A Approved for Public Release,
Distribution Unlimited
14
Objective Results

• What did the addition of TEB do to the simulation
  and how does it compare with what was measured?

14
Distribution A Approved for Public Release,
Distribution Unlimited
15
Objective Results
15
Distribution A Approved for Public Release,
Distribution Unlimited
16
Objective Results

• What did the addition of TEB do to the simulation
  and how does it compare with what was measured?

16
Distribution A Approved for Public Release,
Distribution Unlimited
17
Objective Results
Ksjflaksjdfasdfjasfjsdfkjdaslfkjdaslfjkasdflkjasl
dfkjasldkfjasldkfjasldfkjasldkfjasldfkjsadlfkjlksj
flakdfj
17
Distribution A Approved for Public Release,
Distribution Unlimited
18
Objective Results
18
Distribution A Approved for Public Release,
Distribution Unlimited
19
Conclusions

• The addition of TEB allowed the model to simulate
  many features common to a UHI and therefore seems
  to be working properly
• Comparisons against observational data suggests
• TEB is moving the simulation of the PBL closer to
  observations within the core of the urban
  environment
• Both models struggle with observations near the
  urban transition with TEB is likely over doing
  things somewhat
• possibly due to land surface representation at
  5km grid spacing
• Argue that overall TEB is moving the PBL in a
  better direction

19
Distribution A Approved for Public Release,
Distribution Unlimited
20
BACKUPS
20
Distribution A Approved for Public Release,
Distribution Unlimited
21
Motivation

• These facts have driven the research community to
  develop a number of urban parameterizations
  designed for use with NWP models
• Designed to supplement a parent LSM in urban
  areas
• Vary in sophistication from simple bulk
  approaches to fully coupled and integrated land
  and atmospheric parameterizations
• Almost all application work done with urban
  parameterizations in peer reviewed literature has
  been done with very fine resolution modeling
  (lt2km)
• Also often initialized with operational NWP

21
Distribution A Approved for Public Release,
Distribution Unlimited
22
Land Surface Model

• Each class of land surface is given different
  characteristics such as albedo, surface
  roughness, displacement height, thermal
  properties etc
• Urban is generally treated as vegetation
• No capacity for 3D geometry effects
• No ability to add anthropogenic sources
• Thermal characteristics poorly represented
• Result LSM not designed to produce a UHI or any
  of its impacts

22
Distribution A Approved for Public Release,
Distribution Unlimited
23
Models and Data

• Urban Parameterization Town Energy Balance
  (TEB) Model, Masson (2000)
• Treats the urban area as a 3D volume with each
  grid cell containing a local street canyon system
  of roofs, walls, and roads
• Models three different energy balances (road,
  roof, wall)
• Utilizes a user provided set of morphology
  characteristics that describe the urban landscape
  to perform its calculations
• All fluxes calculated in this 3D volume are
  provided as surface inputs to the parent model,
  even if the actual morphology of the urban area
  extends into the lower layers of the model
• Allows for anthropogenic fluxes both direct and
  indirect

23
Distribution A Approved for Public Release,
Distribution Unlimited
24
Models and Data

• Mesoscale Model RAMS 4.3 Fully
  non-hydrostatic mesoscale model with an extensive
  history of atmospheric simulation from 200m to
  2000km
• Land Surface Model Land Ecosystem-Atmospheric
  Feedback-2 (LEAF-2)
• Fully coupled to RAMS
• Utilizes 18 Biosphere-Atmosphere Transfer Scheme
  (BATS) vegetation classes
• Another 12 classes were defined from the
  NASA/NOAA Land Data Assimilation System (LDAS)
• LEAF-2 offers the capability to model multiple
  land patches per grid cell

24
Distribution A Approved for Public Release,
Distribution Unlimited
25
Models and Data

• TEB requires the following Morphology related
  data (all are grid cell averages)

• Average bld height
• Fractional area of bld
• Building Aspect Ratio
• Dynamic Roughness
• Albedo/Emis Roads
• Albedo/Emis Roofs
• Albedo/Emis Walls

• Number of layers roofs/roads/walls
• Thickness of layers
• Thermal cond of layers
• Heat capacity of layers
• Internal Temp Bld
• SH/LE traffic
• SH/LE Industry

• For this work I created a 1km morphology dataset
  that allowed me to vary variables in red

25
Distribution A Approved for Public Release,
Distribution Unlimited
26
Models and Data

• Morphology data over Washington DC supplied by my
  efforts and allowed to vary by gridcell all
  other locations were given a default set of fixed
  morphology
• Three different sets of land surface data sets
  were obtained for sensitivity testing
• Standard 30s (1km) AVHRR dataset provided with
  RAMS
• A Land Class/Land Use (LULC) 30m dataset
• LULC modified with Morphology vegetation
  estimates
• Designed to estimate the actual non-natural land
  cover
• AVHRR 72 LULC 60 LULC-Mod 37
• LULC-Mod was the primary dataset

26
Distribution A Approved for Public Release,
Distribution Unlimited
27
Models and Data
Description Option
Grid Structure Arakawa C Grid 3 fixed nested (80, 20, 5 km)
PBL Parameterization Mellor-Yamada
Radiation Parameterization Chen, both LW and SW
Lower Boundary LEAF-2 with Town Energy Balance
Lateral Boundary Klemp/Wilhelmson
Level of First Model Layer 23 Meters (11m for those written to center)
27
Distribution A Approved for Public Release,
Distribution Unlimited
28
Models and Data
28
Distribution A Approved for Public Release,
Distribution Unlimited
29
Subjective Results

• 26 Jun 84

26 Jun 2200L Streamlines TEB No TEB
29
Distribution A Approved for Public Release,
Distribution Unlimited
30
Subjective Results

• 26 Jun 84

27 Jun 1000L Temperature (C) TEB No TEB
30
Distribution A Approved for Public Release,
Distribution Unlimited
31
Objective Results

• Takoma Tower 60m Winds Averaged 2100-0400L
  Obs 2m/s No TEB 4.0 m/s
  TEB 3 m/s

31
Distribution A Approved for Public Release,
Distribution Unlimited
32
Objective Results?

• 26 Jun 84

26 Jun 2200L Streamlines TEB
32
Distribution A Approved for Public Release,
Distribution Unlimited