Transcript NAMSeasTransport010501

Seasonal Airmass Transport to the US
Big Bend, TX
Big Bend, TX
January
July
Prepared by: Rudolf B. Husar and Bret Schichtel
CAPITA ,Washington University, Saint Louis, Missouri 63130
Submitted to:
Angela Bandemehr
April 30, 2000, Draft
Contents
• Introduction
• Transport Climatology of North America
• Back-Trajectory Calculations to the U.S.
– Methodology
– Seasonal back-trajectories to 15 receptors
• Transcontinental Transport Events: Dust and Smoke
• Summary
Introduction
•
Anthropogenic and natural pollutants generated in one country are regularly
transported to other countries adding to their air quality burden.
•
On the average, transboundary pollutants transport to the US is small but under
favorable emission and transport conditions it may cause elevated pollutant levels.
Goal of Work:
•
Illustrate the transboundary airmass transport to the United States
•
The approach is to use backward airmass histories to 15 receptor points in the US,
located mostly at the boundaries.
•
The transport analysis was conducted over the entire calendar year 1999, aggregated
monthly to illustrate the seasonal pattern of transport to each location
•
This work is a spatial and temporal extension of the previous airmass history analysis
for Spring 1998.
Transport Climatology of North America
• To be completed
Features of Air Flow over North America
Seasonal Air Flow
over North America
January
NAVAIR, 1966
July
April
October
Back-Trajectory Calculations to the U.S.
Methodology – Airmass Histories
• An airmass history is an estimate of the 3-D
transport pathway (trajectory) of an airmass prior to
arriving at a specific receptor location and arrival
time.
• Meteorological state variables, e.g. temperature and
humidity, are saved along the airmass trajectory.
• Multiple particles are used to simulate each airmass.
Horizontal and vertical mixing is included; particles
arriving at the same time to follow different
trajectories.
• Back trajectories incorporate the transport direction,
speed over source regions and dilution
The history of an airmass
arriving at Big Bend on 8/23/99
FNL Meteorological Data Archive
The FNL data is a product of the Global Data Assimilation System (GDAS), which uses
the Global spectral Medium Range Forecast model (MRF) to assimilate multiple sources
of measured data and forecast meteorology.
• 129 x 129 Polar Stereographic Grid with ~ 190 km
resolution.
• 12 vertical layers on constant pressure surfaces
from 1000 to 50 mbar
• 6 hour time increment
• Upper Air Data: 3-D winds, Temp, RH
• Surface Data includes: pressure, 10 meter winds,
2 meter Temp & RH, Momentum and heat flux
• Data is available from 1/97 to present.
Methodology: Airmass History Analysis
For details see: Springtime Airmass Transport Pathways to the US
Airmass history (Backtrajectory) Analysis
• Backtrajectories are aggregated by counting the hours each ‘particle’ resided in a grid cell.
Methodology –Residence Time Probability Field
• The grid level residence times hours are divided by the total time the airmasses reside over the entire domain and the
area of the grid cell.
• The resulting probability density function identifies the probability of an airmass traversing a given area prior to
impacting the receptor.
• The residence time probability fields are displayed as isopleth plots where the boundary of each shaded region is
along a line of constant probability. The red shaded areas have the highest probability of airmass traversal and the light
blue areas have the smallest probability.
• The most probable pathways of airmass transport to the receptor are along the “ridges” of the isopleth plot.
The probable airmass pathway to the Seattle receptor
Residence Time Analysis: A 2 Dimensional Approach
• The residence time analysis does not account for the height of the airmass, nor
does it account for removal processes.
• Airmass above the mixing height cannot accumulate surface level emissions
• Back trajectories tend to increase in height with increasing age
Seattle, WA Particle Height Distribution
Airmass History Database
• 15 receptor sites were placed primarily along
the United States border
• Ten day airmass histories were calculated
every two hours during all of 1999.
• 25 particles were used to simulated each
airmass trajectory
•Temperature, Relative Humidity, and Precipitation rate, were also saved out along
each trajectory.
•Airmass histories were calculated using the CAPITA Monte Carlo Model driven by the
FNL global meteorological data.
• This system was previously validated for hemispheric transport by simulating the
April 1998 Chinese Dust Event.
1. Aleutian Islands, AK
January
April
July
October
Description to be completed
2. Point Barrow, AK
January
April
July
October
Description to be completed
Receptor 3: S. Oregon, OR
Receptor 4: Newport, OR
•
•
•
•
The air masses arriving to Oregon originate from the East
throughout the year
In July, the prevailing transport direction narrow, from the
northeast
During the other seasons, the transport direction is variable
between NW and SW
View monthly animation for Newport, OR
5. Seattle, WA
January
April
July
October
Description to be completed
Receptor 6: N. California, CA
Receptor 7: San Francisco, CA
Receptor 8: Santa Barbara, CA
9. San Diego, CA
January
April
July
October
Description to be completed
10. Big Bend, TX
January
April
July
October
Description to be completed
11. N. Minnesota, MN
January
April
July
October
Description to be completed
12. St. Louis, MO
January
April
July
October
Description to be completed
13: Everglades, FL
January
April
July
October
Description to be completed
14: Rochester, NY
January
April
July
October
Description to be completed
15: Burlington, VT
January
April
July
October
Description to be completed
Transcontinental Transport Events: Dust
The Asian Dust Event of April 1998
Mongolia
China
Korea
On April 19, 1998 a major dust storm occurred over the Gobi Desert
The dust cloud was seen through SeaWiFS, TOMS, GMS, AVHRR satellites
The dust transport was followed on-line by an an ad-hoc international group
Trans-Pacific Dust
Transport
The dust cloud traversed the
Pacific in 6 days at about 4
km altitude
As the dust approached N.
America, it subsided to the
ground
Asian Dust Cloud over N. America
Reg. Avg. PM10
100 mg/m3
Hourly PM10
On April 27, the dust cloud rolled into North America.
The regional average PM10 increased to 65 mg/m3
In Washington State, PM10 exceeded 100 mg/m3
Smoke from Central American Fires
May 14, 98
Smoke from Central American
Fires
DMSP – Night Light
May 15, 98
SeaWiFS, TOMS, Bext May 15,
1998
Smoke Aerosol and Ozone – Inverse Relationship
Extinction Coefficient (visibility)
Surface Ozone
Surface ozone is generally depressed under the smoke cloud
Summary of Global Air Pollution and Transport
•
The global sulfur emissions have shifted from N. America and Europe to East Asia.
•
The industrial ‘belt’, 30-60 deg N, is dominated by anthropogenic SOx, NOx and O3.
This conforms to the conventional wisdom since the 1970s.
•
Recent satellite data show that NOx, HC and aerosols are dominated by biomass
burning in the subtropics and the southern hemisphere. ??
•
The radiatively active global aerosol is dominated by smoke and dust, rather then by
industrial sulfates as we have presumed.
•
Episodic trans-continental transport of dust and smoke (ozone?) can now be detected
and modeled routinely.
•
Such extra-jurisdictional ‘pollution’ events cause significant episodic impact on the air
quality of N. America.
Source Impact of Pollution and Dust/Smoke Events
• Two key measures of source impacts are on the
concentration and dosage at the receptor
• Both depend on the source strength as well as the
atmospheric transmission probability
• Pollution emission rates are relative low (say
E=1) compared to dust/smoke events (E = 100)
but they are continuous (L= 1) while the
dust/smoke events are intermittent (L=0.01)
• Dust/smoke events produce high short-term
concentration peaks at the receptor that are easily
to detectable.
• Long-range pollution impacts are difficult to
detect because the receptor concentrations are low.
• However, the long-term dosage form the two
types of sources may be similar.
Example Concentration/Dosage Calculation
The impact of the emission from source i, Ei, on the
concentration at receptor j, Cj , is determined by
the transmission probability, Tij :Cj = Tij Ei
The dosage is the integral of the concentration over
the time length, Li, Dj = Li Cj
Emission Rate:
Transmission:
Emission Length:
Cj = 1 x 1 = 1
Dj = 1 x 1 x 1 = 1
Pollution
Ei = 1
Tij = 1
Li = 1
Dust or Smoke Event:
Emission Rate:
Ei = 100
Transmission:
Tij = 1
Emission Length:
Li = 0.01
Cj = 100 x 1 = 100
Dj = 100 x 1 x 0.01 = 1
Summary (tentative)
• At boundaries of the US the air is transported from different directions
• However, each receptor location has a climatologically well defined
seasonal pattern
• Transport to the West Coast occurs primarily from the Pacific
throughout the year
• Transport to the Southeastern US is form the north in the cold season
and from the southeast during the warm season
• The Northeast is receives air from the Arctic, Pacific and the Tropics
• The above transport pattern are consistent with the know climatological
regimes of N America
• (…..)
•
•
•
•
•
Hemispheric option – full context?
Half-hemispheric option-better proportions?
What do you think?
Note: there is a bit of a gap at the datelineproblem with the splicing.
Also, the fine features of the transport pattern
an not too meaningful
• - applicable only to 1999
• - vertically integrated, not surface
transport
Polar Stereographic Projection
• Joe, this projection always confuses me…
• It is a nice projection for meteorologists but for policy types?
• R