Impact of the implementation of a nudging upper boundary condition in AMPS

1
Impact of the implementation of a nudging upper
boundary condition in AMPS
David H. Bromwich1, Andrew J. Monaghan1, and
Kevin Manning2 1-Polar Meteorology Group, Byrd
Polar Research Center, The Ohio State
University, Columbus, Ohio 2-Mesoscale and
Microscale Meteorology Division, National Center
for Atmospheric Research, Boulder, Colorado
2
Introduction Background

• Limited area models require that the model
  boundaries have finite extents, including the top
  of the model. This makes it necessary to
  parameterize what occurs to the internally
  generated model energy that interacts with this
  boundary. This parameterization is known as the
  Upper Boundary Condition (UBC)
• In limited area modeling, the UBC was not
  addressed extensively before nonhydrostatic
  models became widely used.
• Ideally, the UBC should be imposed in such a way
  that makes upward propagating wave energy pass
  through the boundary without any reflection, as
  in the real atmosphere, which has no top
• Several UBC types have been implemented

3
Introduction UBC Type 1 Rigid Lid

• Requires that (?dp/dt0 at the model top)
• Advantages
• Numerically inexpensive
• Works well in areas with little topographic
  variation
• Disadvantages
• Upward propagating wave energy doesnt exit model
• Model top must be set very high to work well over
  steep topography (expensive and also pushes
  boundary well into stratosphere)
• Can cause biases of Ttop10oC Psfcfew hPa

Model Top
4
Introduction UBC Type 2 Radiative Condition

• (Klemp and Durran 1983)
• Advantages
• Permits internal gravity waves to propagate
  through the model top
• Disadvantages
• Must be applied in spectral space therefore,
  wave characteristics must be specified
• Also requires very high model top so that gravity
  wave energy is of secondary importance (Klemp and
  Durran 1983)

Model Top
5
Introduction UBC Type 3 Absorbing Layer

• (Klemp and Lilly 1978)
• Advantages
• Damps upward propagating wave energy via
  filtering or smoothing in the top model levels
• Ideally, does not require model top to be set
  high
• Disadvantages
• Numerically expensive
• Can cause abrupt discontinuities in model fields
  at bottom of filtering layers if not applied
  correctly (Morse 1973)
• Can still get wave reflection if filtering is
  insufficient or excessive (Pielke 1984)

Model Top
Damping Zone
6
Introduction Overview

• In this study, a new type of UBC is examined.
  The Nudging UBC nudges the model simulation
  toward a specified large-scale analysis with an
  exponential function within an absorbing upper
  boundary layer (here, the 8 highest sigma
  levels).
• The function applies smoothing and filtering of
  the simulated temperature fields toward the
  large-scale analysis fields gradually from the
  bottom of the absorbing layer to the top.
  Therefore, this method works much in the same
  manner as a lateral boundary condition.
• Three topics are discussed here
• The Nudging UBC is tested versus several other
  UBC schemes for an intense cyclone over West
  Antarctica. The results are verified against
  temperature soundings from the Global Positioning
  System / Meteorology (GPS/MET) experiment (Ware
  et al. 1996)
• The statistical performance of the new UBC as
  implemented in the Antarctic Mesoscale Prediction
  System (AMPS) is examined over a three month
  period.
• A case study of a specific forecast in October
  2003 examines more closely the impact of the new
  UBCs in AMPS.

7
1. Testing the Nudging UBC Model Description

• Use Polar MM5 modification of NCAR MM5 V2
• 
  ---- nonhydrostatic version
• The ice nuclei concentration equation (Meyers et
  al., 1992) is implemented in the explicit
  microphysics parameterization (Reisner's
  mixed-phase scheme ).
• The cloud ice and water content predicted by the
  explicit microphysics parameterization is now
  used to determine the radiative properties of
  clouds in the CCM2 radiation parameterization.
• Two additional substrate levels which increases
  the substrate depth to 1.91 m (Compared to 0.47 m
  in the unmodified version) are added to the
  multi-layer soil model proposed by Dudhia (1996).
  
• The addition of variable fraction sea ice surface
  type is added to account for the mixture of sea
  ice and open ocean in one model grid box .

8
1. Testing the Nudging UBC Experimental Design
Table 1. Eight experiments with different upper
boundary treatments
9
1. Testing the Nudging UBC Sounding Locations
121X121 60km 28 vertical layers 72-h forecasting
mode 8 GPS/Met points Cross-section line
10
1. Testing the Nudging UBC Temperature Soundings
Big Difference Over land
11
1. Testing the Nudging UBC Temperature Soundings
Not much difference over the ocean points
12
1. Testing the Nudging UBC Sea Level Pressure
6-d averaged SLP difference between simulations
and ECMWF/TOGA
Control-EC/TOGA
Rad10-EC/TOGA
4-8 hPa
13
1. Testing the Nudging UBC Sea Level Pressure
6-d averaged SLP difference between simulations
and ECMWF/TOGA
Lid10-EC/TOGA
Nudge-EC/TOGA
14
1. Testing the Nudging UBC Simulated Upper Level
Jet
00 UTC 09 October 1995
15
1. Testing the Nudging UBC Simulated Upper Level
Jet
00 UTC 09 October 1995
Rad
16
1. Testing the Nudging UBC RMSE of 500 hPa Geop.
Ht.
17
1. Testing the Nudging UBC Vertical Motion
00 UTC 09 October 1995
Control
Lid10
18
1. Testing the Nudging UBC Vertical Motion
00 UTC 09 October 1995
Nudge
Asm
19
2. Implementing the Nudging UBC in AMPS Overview

• AMPS switched from its old UBC (rigid lid at 100
  hPa) to its new UBC (nudging UBC at 50 hPa) in
  May 2003
• Here we compare three months of statistics from
  two winter periods
• JJA 2002 (Before new UBC)
• JJA 2003 (After new UBC)
• The Correlations, RMSEs, and biases of the
  following fields are examined
• Column Temperature
• Column Wind Speed
• 150-hPa Temperature
• 150-hPa Wind Speed
• Surface Pressure
• The statistics are calculated for the AMPS 30-km
  grid versus radiosonde and surface observations,
  and unless otherwise noted, are valid for the
  36-60 hr forecasts.

20
2. Implementing the Nudging UBC in AMPS
Temperature and Wind Speed Soundings (36-60 hr,
averaged over all stations)

• In general, correlations and RMS errors are
  improved throughout model column
• Temperature biases reduced (become cooler) in top
  layers of model
• Winds become stronger throughout column

21
2. Implementing the Nudging UBC in AMPS 150
hPa Temperature (36-60 hr)

• Correlations much higher
• RMS errors are improved at all stations
• Biases reduced at most stations
• Wind Speeds (not shown) also show significant
  improvement

Before
After
22
2. Implementing the Nudging UBC in AMPS
Surface Pressure (36-60 hr)

• Not much change in correlations performance
  very high before/after new UBC
• RMS errors are improved at all stations
• Biases reduced at 15 of 16 stations

Before
After
23
2. Implementing the Nudging UBC in AMPS
Performance vs. Forecast Hour 150 hPa
Temperature (averaged over all stations)

• Improvement increases with forecast hour
• RMS errors cut in ½ throughout forecast
• Similar performance for 150-hPa winds

24
2. Implementing the Nudging UBC in AMPS
Performance vs. Forecast Hour Surface
Pressure (averaged over all stations)

• RMS errors improved by about 1 hPa throughout
  forecast
• Improvement to bias in later forecast hours

25
3. AMPS Case Study Experimental Design

• Ran 2 AMPS forecasts for 12Z October 02, 2003
• Old-100 (Rigid Lid, Top 100 hPa)
• New-50 (Nudging UBC, Top 50 hPa)
• Look at differences in cross sections of vertical
  motion on AMPS 30-km domain

26
3. AMPS Case Study Synoptic Situation

• An upper level ridge approaches West Antarctica.
• Upward(red)/Downward(blue) motion becomes
  enhanced

Hour 60
Hour 66
Hour 72
27
3. AMPS Case Study Vertical Motion
Hour 60
Hour 66
Hour 72
28
Summary

• When the rigid lid and radiation UBC in MM5 is
  applied to Antarctica which has high and steep
  terrain, the model
• generates large warm biases near the model top
• produces large biases of sea level pressure
• underestimates the magnitude of the upper level
  jet and positions it below its actual location
• These problems are significantly reduced by the
  implementation of the Nudging UBC
• This bears out statistically in a 3-month
  comparison of AMPS forecasts before and after the
  new UBC was implemented. Increases in the skill
  of temperature and wind speed forecasts are noted
  throughout the atmospheric column. Surface
  pressure bias and RMSE are widely reduced as well
• The new UBC reduces unrealistically high vertical
  motions throughout the column, though the exact
  mechanisms for this are still under
  investigation.

29
Unused slides to follow
30
1. Testing the Nudging UBC The Event
31

1. Testing the Nudging UBC The Event
Land Soundings (gravity waves most pronounced)

32

1. Testing the Nudging UBC The Event
Ocean Soundings (gravity waves least pronounced)

33
2. Implementing the Nudging UBC in AMPS 150
hPa Wind Speed (36-60 hr)

• Correlations generally higher
• RMS errors are improved at all stations
• Biases reduced at most stations

Before
After