Exam II Flashcards
Under which of the following circumstances do errors in model physics have the highest impact?
A) When dynamic forcing is strong
B) When physical processes are strong
C) When dynamic forcing is weak
D) When physical processes are weak
B & C
Which of the following elements help forecasters add value over NWP?
A) Continuous MetWatch
B) Use of MOS for rare events
C) Regular use of verification
D) Forecaster experience
A, C, & D
Which of the following elements help forecasters add value over NWP?
A) Continuous MetWatch
B) Use of MOS for rare events
C) Regular use of verification
D) Forecaster experience
A, C, & D
Which of the following tools help forecasters perform the continuous meteorological watch (MetWatch)?
A) Radar reflectivity and velocity
B) Medium-range NWP forecasts
C) Satellite depictions
D) Surface analyses
A, C, & D
True or False:
Model analysis and forecaster experience have a direct impact on the final forecast issued by meteorologists.
True
Which of the following situations present opportunities to improve on the NWP forecast?
A) Forecasting severe weather in the short range when convection is initiating
B) Forecasting precipitation for a landfalling hurricane
C) Forecasting temperature in the medium range when your area is in a stable block
D) Forecasting temperature in areas of rugged terrain
A,B, & C
Which of the following statements about MOS guidance are correct? (Choose True or False for each option)
A) Data sets are seasonally stratified.
B) Model guidance will lead to a good forecast if environmental conditions are typical for the event.
C) Model guidance is skillful in forecasting atypical events.
D) Regional parameterization affects MOS guidance.
E) MOS guidance corrects for systemic errors in the NWP models.
A) True B) True C) False D) False E) True
Choose the best answer.
A major snowstorm is to occur on the 10th of October at a station in Nebraska. The GFS MOS report has the following October 9-15 GFS Extended MOS forecasts for maximum and minimum temperature:
KLBF GFSX MOS GUIDANCE 10/08/xxxx 1200 UTC
Forecast Hour:
24 36 48 60 72 84 96 108 120 132
Max/Min Temperature:
30 57 26 31 17 33 27 43 33 48
Snow is rarely observed in early October at this location, and the air mass accompanying and following the storm is unusually cold. Forecast 850-hPa temperatures are expected to be at least 3 standard deviations below normal, which statistically is expected only 1% of the time. Based on this information, how would you adjust the forecast temperatures for Monday 12 October (day 4)?
A) I would use them as presented because MOS removes systematic bias.
B) I would use them as presented because MOS predicts extreme events well.
C) I would lower them because of the unusual snow cover and cold air mass.
D) I would raise them because of the unusual snow cover and cold air mass.
C
Choose all that apply.
You are a TV meteorologist on a network 10 p.m. newscast in the Philadelphia, PA area. Your main concern is the potential for snow during rush hour (7 a.m. to 9 a.m. EST, or 12 to 14 UTC) of the next morning.
At 5 p.m. EST (22 UTC) your time, you reviewed the previous hour’s radars, surface observations, and corresponding forecasts of simulated composite radar reflectivity and lowest level reflectivity from the 18 UTC NAM/WRF-NMM run that day.
Given the data in these afternoon graphics, how would you assess the NAM forecast from 18 UTC valid at 21 and 22 UTC?
A) The observations indicate that the NAM is correctly developing precipitation in south-central Virginia.
B) The composite radar reflectivity indicates that the NAM is too fast developing precipitation in south-central Virginia
C) The observations indicate that the NAM is too slow developing precipitation in south-central Virginia.
D) The composite radar reflectivity indicates that the NAM is too slow developing precipitation in eastern North Carolina
C & D
Choose all that apply.
Now you want to develop your forecast for the late evening news. The following 10 p.m. EST (03 UTC) data for observations and level 1 NAM/WRF-NMM 18 UTC run of simulated reflectivity valid at the same time are available. After viewing these graphics, answer the question below.
How do the observations now compare to the NAM/WRF-NMM observations in the mid-Atlantic states?
A) The advance of precipitation into central Maryland is too slow in the NAM.
B) The advance of precipitation into central Maryland is accurately predicted in the NAM.
C) The advance of precipitation into central Maryland is too fast in the NAM.
D) The forecast precipitation in the NAM shows signs of becoming banded.
A & D
Choose the best answer.
The graphics here show the forecast of precipitation from the 18 UTC run of the NAM/WRF-NMM for the hours covering rush hour the next morning. Assume that temperatures will support snow throughout the Philadelphia PA area. The Short-Range Ensemble Forecasts (not shown) support the precipitation forecast of the NAM/WRF-NMM.
Use this information and graphics above to answer the question.
Based on these 1-hour precipitation graphics and verification of the movement of the precipitation shield from the previous two questions, what would your 10 p.m. forecast be for the next morning rush hour (7-9 a.m. EST, 12-14 UTC) in the Philadelphia PA area?
A) Snow will start before the beginning of the rush hour in the entire Philadelphia area.
B) Snow will start after the end of rush hour in the entire Philadelphia area.
C) Snow will not reach the Philadelphia area before, during, or after the morning rush hour.
D) Snow will overspread the Philadelphia area from south to north during the rush hour.
D
Choose the best answer.
When assessing the certainty of a forecast using EFS products, which EFS limitation needs to be taken into consideration?
A) The greater uncertainty of initial conditions due to EFS initialization and perturbation methods.
B) The high resolution of EFSs and their ability to offer detailed information for high impact weather events.
C) The availability of detail for onset time of convection.
D) The probability of exceeding a forecast threshold.
A
Determine whether each of the following options would be considered an advantage of EFSs when compared to deterministic models. Choose True or False for each option.
A) Information on the probability of a weather event.
B) Details about the degree of certainty in a forecast.
C) Improved spatial resolution.
D) Meteorological details for forecasts from 0 to 72 hours.
A) True
B) True
C) False
D) False
Are each of the following an advantage of deterministic models when compared to EFSs? (Choose True or False for each option)
A) Resolution of mesoscale features.
B) Details about the location and amount of rainfall.
C) Details about wind speeds.
D) Details about the certainty of a forecast.
A) True
B) True
C) True
D) False
What are 3 primary errors in data and quality control?
- Instrument Errors
- Representativeness Errors
- Converting remotely-sensed data into high-quality observations to be integrated with other data
What are 3 types of representativeness errors?
- Vertical
- Horizontal
- Temporal
What are 2 primary model initialization problems?
- First guess can sometimes overwhelm actual data.
2. First guess may result in good observations being ignored.
What will poor analysis typical lead to in models?
Huge forecast errors.
What are some of the most common atmospheric variables that are not routinely measured?
- Longwave & Shortwave Radiation
- Cloud Water & Ice Content
- Surface roughness of the ocean
- Turbulence
What is the model’s initial impression of the atmosphere’s current condition?
The first guess (an earlier forecast)
Where is the model’s first guess most easily modified?
Where is it least easily modified?
data-rich areas (CONUS)
data-poor areas (oceans)
What are some ways to check the model’s first guess influence on the analysis?
1) Compare different model analyses to each other.
2) Compare model analyses to satellite, radar, and other real-time information.
What are some common sources of error in the model?
1) Equations of motion are incomplete
2) Errors in the numerical approximation
3) Boundary conditions
What is the horizontal resolution of an NWP model related to for grid point models?
For spectral models?
- the spacing between grid points
- the number of waves that can be resolved
What does ‘resolution’ represent in a grid point model?
In a spectral model?
- the average area depicted by each grid point
- the number of waves used
Why is it important to know the amount of area between grid points?
Because atmospheric processes and events occurring over areas near to or smaller than this size will not be included in the model.
In spectral models, what is the horizontal resolution designated by?
What does this variable indicate?
- “T” number
- It represents the number of waves used by a spectral model
What does the “T” stand for in “T-number?”
What does it indicate?
- Triangular Truncation
- The particular set of waves used by a spectral model
What equation is the wavelength of the smallest wave in a spectral model represented as?
minimum wavelength = 360 degrees/N, where N is the total number of waves (the “T” number)
What is the equation used to represent the approximate grid spacing with the same accuracy as a spectral model?
delta-X = 360 degrees/3N
Do the dynamics of spectral models or grid point models retain better wave representation?
Spectral models
Fill in the blanks.
Spectral model physics is calculation on a ________ with about _____ times as many grid lengths as number of waves used to represent data.
- grid
- three
Fill in the blanks.
Spectral models do a fine job with ____ waves in the free atmosphere, but have coarser representation of the _________, including surface properties.
- dry
- physics
True or False?
The more physics that is involved in the
evolution of the forecast, the less the advantage in
spectral model forecasts compared to comparable
resolution grid point forecasts.
True
What two factors limit model representation of orography?
- The horizontal resolution of the model
- The horizontal resolution of the terrain dataset used
What will happen to the terrain details provided in a model’s dataset if it cannot resolve terrain features?
The terrain details will be averaged out
What two factors affect the character of model terrain?
Model resolution and terrain smoothing
What are 4 common shortcomings of model terrain representation?
1) Elevations of the highest mountain peaks or ranges are
generally less than in reality
2) Valleys are often not represented or are filled
3) Mountain range elevation is spread over a too-broad
horizontal area
4) Elevation gradients are underestimated
What part of a model’s representation is a major factor in its ability to predict meteorological features induced by terrain?
Its representation of surface topography
What are some possible effects of inadequate model terrain on weather elements? (Choose all that apply.)
a) Shifting vertical motion maxima/minima away from the mountains due to insufficient terrain slope
b) Inaccuracies in forecasting the strength and location of the upper-level jet
c) Misplacing precipitation maxima and minima in complex terrain
d) Underestimating and broadening precipitation maxima
e) Difficulty in analyzing and forecasting upper-tropospheric height and temperature fields
f) Inability to depict downslope winds, valley winds, drainage winds, and other small-scale processes
g) Inability to depict mountain lee wave development and propagation
h) Difficulty in resolving and representing valley inversions and cold air damming
a, c, d, f, g, h
True or False:
Models do not have difficulty resolving features influenced or caused by the interface between land and large bodies of water.
False
True or False:
To complete one forecast, equations need to be solved for many variables at millions of grid points up to several thousands of times, totaling close to a trillion updates for each forecast variable.
True
How do computing resources affect NWP models?
They limit the resolution
Fill in the blanks:
The additional computing resources required to run a model at half its current horizontal resolution increase by a factor of _____, assuming no change to the _______ resolution or ______ size.
- eight
- vertical
- domain
Why does increased resolution increase demand on computing resources?
Because the model must determine values for more grid points.
What does the CFL criterion state?
“The speed of fastest waves in the model must be less than or equal to grid spacing divided by the time step”
What scientists created the CFL criterion?
Courant, Friedrichs, & Lewy
What would violating the CFL criterion lead to?
the “blowing up” of a finite-difference weather prediction model.
What must the CFL be in order for there to be stability?
Less than or equal to 1.
In a numerical model, why must one must look at a moving parcel often enough to keep track of its actual path?
To maintain numerical stability
True or False:
In reality, NWP models generally use much longer time steps than would be computed by the CFL equation.
False—generally use much shorter time steps….
Why do NWP models generally use much shorter time steps than would be computed by the CFL equation?
This is due to instabilities in numerical methods and use of physical parameterization
True or False:
Gradient intensity is limited by model resolution.
True
Is the relationship between the size of weather features to be predicted and the grid spacing in a gridpoint model important? (Yes or No)
Yes
Which of the features listed below can be resolved well by a 35-km resolution gridpoint model or T360 spectral model that might be used for extended prediction? For each feature, decide if it can be resolved well (yes, no, or to a limited extent).
a) Arctic plunge east of Rocky Mountains
b) Downslope windstorm
c) Cold-air damming
d) Snow band within East Coast low
e) Terrain-induced precipitation maxima/minima
f) Upper-level front
g) Synoptic surface front
h) Coastal front
i) Sea breeze circulation
j) Outflow boundary
k) MCS
l) Hurricane
a) yes
b) no
c) yes
d) limited
e) limited
f) yes
g) yes
h) limited
i) limited
j) no
k) limited
l) limited
Which of the features listed below can a mesoscale 10-km resolution model adequately resolve? For each feature, decide if it can be resolved well (yes, no, or to a limited extent).
a) Arctic plunge east of Rocky Mountains
b) Downslope windstorm
c) Cold-air damming
d) Snow band within East Coast low
e) Terrain-induced precipitation maxima/minima
f) Upper-level front
g) Synoptic surface front
h) Coastal front
i) Sea breeze circulation
j) Outflow boundary
k) MCS
l) Hurricane
a) yes
b) limited
c) yes
d) limited
e) limited
f) yes
g) yes
h) yes
i) limited
j) limited
k) limited
l) limited
Does a 4-km mesoscale model have sufficient resolution to resolve an existing mesoscale convective system (MCS)?
Yes, with limitations
Does a 4-km mesoscale model have sufficient resolution to define outflow boundaries associated with the MCS?
Yes, but with detail sensitivities
Does a 4-km mesoscale model have sufficient resolution to resolve individual cells and predict their track?
Yes, large supercells can be grossly resolved by a 4-km model, although their substructures such as the mesocyclone and wall cloud cannot.
A model’s ability to resolve features depends not
only on its horizontal resolution, but also on what? (3 things)
1) vertical resolution
2) number of vertical layers
3) physics package used to define a variety of surface and atmospheric processes.
What are limited-area models strongly constrained by?
Their boundary conditions
How do spectral models and grid point models represent the atmospheric column?
In discrete layers
What are some key features that need adequate vertical resolution?
1) Near-surface sharp inversions and superadiabatic lapse rates
2) Top of the boundary layer
3) Sloping frontal zones
4) Tropopause(s)
True or False: The ratio of the horizontal and vertical resolutions of a model must be consistent with the slope of the weather phenomena of interest.
True
