Modeling urban meteorology: applications to tropical coastal cities

1
Modeling urban meteorology applications to
tropical coastal cities
Bob Bornstein, R. Balmori, H. Taha Dept. of
Meteorology San Jose State University San Jose,
CA USA pblmodel_at_hotmail.com presented
at Caribbean Climate Symposium University of
Puerto Rico 24-25 April 2006
2
OVERVIEW

• Introduction to urban climate
• Meso-met modeling problem areas
• uMM5
• Basic equations
• Parameterizations
• Applications
• Urban climate
• Severe Wx
• Air quality
• Future work

3
URBAN CLIMATE CAUSESTO BE MODELED

• GRASS/SOIL ? CONCRETE/BUILDINGS ?
• ALTERED SFC FLUXES
• FUEL CONSUMPTION ?
• ATM POLLUTION, HEAT, MOISTURE
• BUILDINGS ATM POLLUTION?
• ALTERED SOLAR IR FLUXES

4
URBAN WX/CLIMATE MODIFICATIONS TO BE MODELED

• TEMP PRECIP increased or decreased
• VISIBILTITY decreased
• WIND DIRECTION con- divergence
• TURBULENCE mechanical thermal
• PBL NIGHT STABILITY neutral
• SYNOPTIC SEA BREEZE FRONTS slowed
• THUNDERSTORMS triggered or split

5
HUMAN-HEALTH IMPACTS OF URBAN CLIMATE TO BE
MODELED

• URBAN HEAT ISLANDS ?
• SUMMER THERMAL STRESS INCREASED
• WINTER WIND-CHILL ICING DECREASED
• PRECIP ENHANCEMENT ?
• FLOODS INCREASED
• ICING INCREASED
• ALTERED TRANSPORT DIFFUSION ?
• OZONE EPISODES
• EMERGENCY RESPONSES

6
DAVIS (RELATIVE) NDVI (UPPER) ALBEDO
(LOWER) WHERE (A) RED AREAS HAVE HIGH NDVI
HIGH ALBEDO (B) CITY HAS INTERMEDIATE SPOTTY
VALUES OF NDVI LOW ALBEDO
Fig. 3
Fig. 4
7
SFC RAD-T (C) DAY (UPPER) NIGHT (LOWER)

• DAY
• HOT RED DRY SOIL
• COOL BLUE WET SOIL
• URBAN AREAS IN BETWEEN T
• (WITH WARM ROOFS)
• UHI ? DEPENDS WHERE RURAL
• OBS TAKEN

Fig. 5

• NIGHT
• HOT RED URBAN ROADS
• WARM YELLOW WET SOIL
• COOL BLUE DRY SOIL
• UHI YES

8
Obs Nocturnal UHI-induced convergence
advected downwind of Manhattan
C
C
D
D
C
Obs Daytime building- barrier induced divergence

D
9
Obs Slow moving cold front air in lowest 250-m
over city was not replaced
Obs Sea-breeze front retarded by
building-barrier effect
10
(No Transcript)
11
Power plant plume
Area-source emissions
Obs weak urban-induced elevated-inversions
barriers to z-diffusion of canopy-layer
pollutants
12
Transformed-Projected MM5 Energy Eq

• where,

13
MM5 PBL Problems

• gt Monin-Obukov SBL-profiles fail in very
• Stable conditions ? new Fm,h
• Unstable conditions ? switch from u to w
• Yamada-Mellor 1.5 TKE-closure wrong ? TKE wrong
  in upper PBL
• Black carbon aerosols ? RFD ? PBL stability
• gt Wrong (poorly known) inputs
• Too small urban z0 ? too fast urban V
• Charnock mid-ocean z0 ?
• too-fast coastal V
• Coarse SST ? poor coastal T-grad
• Deep soil-temp BC ? bad Tmin
• IC soil-moisture ? bad Tmax

14
Businger Fs
unstable
unstable
stable
stable

• Notes
• Stable and unstable data fit curves
• Unstable slope not same as stable slope
• (as it must be with M-O)
• Problem stable Fs rise too rapidly ?
• model turbulence shut off too early

15
LA MM5 Domain-Average 2-m Temps 3-6 Aug 97
RUN 1

• Synoptic trend
• (red arrow)
• use analysis nudging
• Max-T
• adjust soil-misture
• Min-T
• adjust IC soil-moisture
• Bad meso-T ?
• bad meso-V

RUN 5
16
Realistic High-Resolution SSTs
control SST
model SST
Nest 4 1.33 km
Pullen Holt (NRL) Blumberg (SIT)
Observed SST
control SST
model SST
Nest 5 0.44 km
17
CORRECT PBL LENGTH SCALE Freedman Jacobson
(2002 2003, BLM)
Prog Eq. for TKE DISSIPATION RATE, e
__
__

• Where l ceE3/2/e
• Values of se sE reversed in Mellor Yamada ?
• reversed in all atm models ?
• Wrong K TKE in upper PBL

18
CALIBRATION TO NEUTRAL PBL l vs. z
Lines various ? ce2se/sE

• x COLEMAN (99) DNS
• PBL elev inversion
• New (R-panel) good-fit
• in all of PBL (? 1.3),
• as l? aloft
• gt Poor fit aloft
• no elev inversion
• Standard (L-panel) bad fit
• in upper PBL (? 2.5)

good
bad
? 2.5
? 1.3
new
old
19
Same, but for K (z)
bad
good
x COLEMAN (99) DNS New (R-panel) best-fit
? 1.3 (solid line) gt good in lower PBL
K? aloft Standard approach (L-panel) best
fit ? 2.5 gt poor in lower PBL
new
old
20
Urbanization Techniques

• Urbanize surface, SBL, PBL momentum,
• thermo, TKE Eqs
• Allows prediction within UCL
• From forest-canopy model Yamada (1982)
• Veg param replaced with urban (GIS/RS) terms
• Brown and Williams (1998)
• Masson (2000)
• Martilli et al. (2001) in TVM
• Dupont et al. (2003) in MM5 ? uMM5
• Taha et al. (2005) in uMM5

21
Within Gayno-Seaman PBL/TKE scheme
From EPA uMM5 Mason Martilli (by Dupont)
22
_________
______
3 new terms in each prog equation
? Advanced urbanization scheme from Masson (2000)
23
uMM5 inputs as f (x, y)

• land use (38 categories)
• roughness elements
• anthropogenic heat as f (z, t)
• building heights
• paved surface-fractions
• building drag-force coefficients
• building height-to-width, wall-to-plan,
• impervious-area ratios
• building frontal, building plan, rooftop
• area-densities
• e, c?, a, etc. of walls and roofs

24

• For houston
• (S. Stetson)
• Used hb
• Should use
• s of hb

25
UCP (1 km grid) Houston Ching Burian (2006)
Wall-to-Plan area ratio
Building plan area fraction
Building frontal area index
Height-to-Width ratio
26
uMM5 performance by CPU
? With 1 CPU MM5 is 10x faster than uMM5
With 96 CPU MM5 is still gaining, but MM5 has
ceased to gain at 48 CPU then it starts to
loose
? With 96 CPU MM5 is only 3x faster than uMM5
(lt 12 CPU not shown)
27
Urbanization scheme
28
Urbanized meso-met model TKE(z)
hc building top
max urban effect
_________________
29
From Martilli
MM5
?uMM5 day nite
30
ATLANTA UHI-INITIATED STORM OBS SAT PRECIP
(UPPER) MM5 W PRECIP (LOWER)
31
uMM5 Houston Runs 22-26 August 2000

• Model configuration
• Domains 108, 36, 12, 4, 1 km
• (x,y) grids 43x53, 55x55, 100x100, 136x151,
  133x141
• s-levels 29 in D1-4 49 in D5lowest ½ s-level
  7m
• 2-way feedback D1-4
• Physics options
• gt Grell Cu D1-2 gt ETA or MRF PBL D1-4
  
• gt Gayno-Seaman PBL D5 gt Simple ice
• gt Urbanization D5 gt NOAH LSM gt RRTM
  radiation
• Inputs
• gt NNRP Reanalysis ADP obs
• gt Burian LIDAR building-data D5
• gt Byun LU/LC modifications

32
Episode-day Synoptics 8/25, 12 UTC (08 DST)
Surface Dp 4 hPa
700 hPa Dz 30 m
700 hPa sfc GC Hs at weakest ? (no gradient)
over Texas ? meso-scale forcing (sea-breeze UHI
convergence) will dominate
33
Concurrent NNRP fields at 700 hPa sfc
Dz 30 m
NNRP-input to MM5 (as IC/BC) captured GC/synoptic
features, as location strength of High were
similar to NWS charts (previous slide)
34
MM5 episode day, 3 PM gt D1 well-reproduced
weak GC pressure-grad flowgt coastal L emerges
in D-2 (in weak form) gt D-3 shows (well formed)
L thus along-shore V
L
35
Domain 4 (3 PM) Note L off of Houston on O3 day
(25th)
? Episode day
L
L
36
uMM5 Domain 5 near-sfc winds at 3 PM 4
successive days

• Episode day
• gt along-coast flow
• warm-inland
• cold-core L

37
Along-shore flow, 8/25 (episode day) obs at 1500
UTC vs uMM5 (D-5) at 2000 UTC
38
O3 Interval 5 ppb (00-16 UTC) 10 ppb (gt16
UTC)
gt 02 UTC UHI Con gt O3 urban-L
weak left-over H
39
Obs speeds (D-5) sfc roughness ? speed-decrease
over city
-
40
Day UHI-Induced Convergence accel obs vs. uMM5
C
uMM5
Gridded Obs
41
Along shore flow came from Cold-Core L D-3 MM5
vs Obs Temps
H
Obs (18 UTC) gt Cold-core L (only 1 ob) gt
Urban area (blue-dot clump) retards cold-air
penetration

• MM5 produces coastal cold-core low (origin like
  in NYC?)

42
Satellite SST Over Gulf of Mx many
detailshttp//www7333.nrlssc.navy.mil/
43
D-5 uMM5-MM5 UHI at 8 PM 21 Aug
H
H

• Upper L MM5 UHI 2.0 K
• Upper R uMM5 UHI 3.5 K
• Lower L uMM5-MM5
• 1.5 K warmer
• Blob is LU/LC error

44
8/23 Day UHI obs vs uMM5 (D-5) 2-m Temps
H
45
Base-case veg (0.1s) urban min of current trees

• Modeled increases
• in trees ( 0.01s)
• urban reforestation
• rural deforestation

46
Run 14 (urban-reforestation) minus Run 13 (base
case) 4 PM 2-m ?T (K)gt reforested urban-area
cools gt surrounding deforested rural-areas warm
W
47
DUHI(t) for Base-case minus Runs 15-18

• UHI Temp in Urban Box minus Temp in Rural Box
• Runs 15-18 different urban re-forestation
  scenarios
• DUHIRun-17 UHI Run-13 UHI (max effect, green
  line)
• Reduced UHI ? lower max-O3 (not shown) ?
• EPA emission-reduction credits

48
Houston Summary

• NNRP captured large-scale forcing for MM5
• uMM5 captured observed
• Urbanization LU/LC characteristics
• UHI and urban-velocity patterns
• Ozone transport-processes
• Urban reforestation
• decreased max-daytime UHI-values
• should thus also decrease max-O3 values

49
FUTURE WORK (1)

• Current uMM5 reduced Houston-UHI ?
• reduced (?) CAMx/CMAQ O3?
• EPA emission-reduction credits
• Update uMM5 input
• Deep-soil BC-temperature to correct min-T bias
• IC soil-moisture (post rain-storm) to correct
  max-T bias
• SST as f(x,y,t)
• New MM5 sea-surface roughness

50
FUTURE WORK (2)

• SJSU/NCAR uWRF for DTRA with
• uMM5 scheme
• Zilitinkevich stable unstable SBLs
• Freedman k-e PBL
• Link with rapid ER canyon dispersion model
• NYC as first US urban test bed?
• Links with
• Urban CFD-models for ER planning 2-way
• Downscaled climate-change model-output ?
• gt increased thermal stress
• gt stronger thunderstorms ?urban flooding
• gt deteriorated air quality

51
Tropical modeling special features

• Small Coriolis ? slow convergence
• Barotropic ? pressure is important
• Highly convective ? hard to simulate
• Small islands with steep topography ?
• key slope-gradient terms missing in equations
• Fewer observations

52
Puerto Rico Collaborations

• With Prof. Gonzales
• Sacramento, CA current
• Puerto Rico planned
• Shortage of qualified US grad students in Met ?
• most US grad students foreign-born with Met,
  Phy, or Engineering BS degrees
• Most US Grad students get research assistantship
  about 18K/year tuition
• You first need to identify
• Research area of interest
• University and Prof. in that area (check UCAR web
  site)
• You then need
• good (math, phy, computer) background, grades,
  GRE, TOFEL
• someone to contact them on your behalf ? send
  them your CV

53
ACKNOWLEDGEMENTS

• D. Hitchock P. Smith, State of Texas
• D. Byun, U. of Houston
• J. Nielsen-Gammon, TAMU
• J. Ching S. Dupont, US EPA
• S. Stetson, SWS Inc.
• S. Burian, U. of Utah
• D. Nowak, US Forest Service
• Funding from USAID, NSF, and DHS

54
Houston and uMM5 References

• Balmori, R., 2006 Urbanized MM5 study of urban
  impacts on an August 2000 Houston O3 episode. MS
  Thesis, Dept. of Met., SJSU.
• Banta, R. M., et al., 2005 A bad air day in
  Houston. Bull. Amer. Meteor. Soc., 86, 657-669.
• Byun, D., et al., 2004 Modeling effects of land
  use/land cover modifications on the UHI and air
  quality in Houston, Texas. Tech. Note, University
  of Houston, 55 pp.
• Dupont, S., et al., 2003 Simulation of
  meteorological fields within and above urban and
  rural canopies with MM5. Tech. Rep., US EPA NOAA,
  ARL, NC, 67 pp.
• Martilli, A., et al., 2002 An urban surface
  exchange parameterization for mesoscale models.
  Boundary-Layer Meterol., 104, 261-304.
• Nielsen-Gammon, J. W., 2004 The surprising
  dynamics of the Houston urban sea breeze.
  Preprints, 84th AMS Conf, Seattle, WA.
• Taha, H., et al., 2005 UHI alterations from
  urban-forest changes in Houston uMM5 model of
  Aug 2000 O3 episode. SJSU Report to HARC.

55
NYC-OBS REFERENCES

• Bornstein, 1968 J. Appl. Met.., 7., 575-82.
• Born. Johnson, 1977 At. Env., 11, 597-04
• Loose Born., 1977 MWR, 105, 567-71.
• Born. Thompson, 1981 JAM, 20, 843-58
• Born., 1987 Modeling the Urban BL, AMS, 53-93.
• Gaffen Born., 1988 Met. Atmos. Phys, 38,
  185-94

56
The End Any Questions?