Piedmont National Wildlife Refuge

1
Piedmont National Wildlife Refuge

• 15 February 2006
• Smoke Simulation Analysis
• Gary L. Achtemeier
• USDA Forest Service
• Athens, GA

2
Task Description

• The following set of slides summarize a
  PB-Piedmont post-burn analysis of smoke movement
  in the aftermath of two (426 and 643 acre)
  prescribed burns conducted by the Piedmont
  National Wildlife Refuge. These burns were
  conducted on 15 February 2006. The model
  simulations run from 1600 EST 15 February through
  0800 EST 16 February. Model smoke locations are
  compared with field observations collected by
  personnel from the Piedmont National Wildlife
  Refuge.

3
Weather

• Weather during the burn and the post-burn smoke
  movement period was variably cloudy with a
  stretch of clear skies during the middle of the
  night. A high pressure center moved from southern
  Georgia to South Carolina by the end of the
  simulation period. Winds gradually shifted from
  the southwest to the southeast. The below figure
  gives a measure of the strength of the large
  scale forcing during the analysis period. Forcing
  of 7.6 12.2 m/sec was sufficient to overcome
  drainage flow formation in northwest-southeast
  oriented valleys but to strengthen flows in
  east-west oriented valleys.

4
Modeling Information

• PB-Piedmont Version 5.0-2006
• Grid spacing 60m
• Domain size 320x374 -gt 19.2x22.4 km
• Visual domain 10.0x8.8 km
• Weather data Hourly METAR
• Elevation data USGS 30 m DEM
• Elevation contour interval 10 meters
• Elevation range is 88 208 meters

5
The PB-Piedmont domain, roads, observation points
for 15 February 2006. The yellow polygons enclose
the burn sites.
6
The modeled location of smoke on the ground
assumes smoldering is coming from random
locations within the model burn areas. Initial
winds are from the south at 5 m/sec beginning at
1630 EST. Note that this image is just the
viewing domain. The time is 1800 EST, 13 minutes
before sunset.
7
The smoke plume at 2000 15 February 2006. The
plumes have broken into patches of smoke that are
concentrated in the valleys located downwind from
the burn sites. Note that drainage flows are
transporting smoke downstream through two NE-SW
oriented valleys from the southernmost burn site.
8
The PB-Piedmont smoke plume at 0000 EST 16
February 2006. Smoke from the southern burn site
has drained into and turned northwestward in the
Okmulgee River valley (dark green strip oriented
NW-SE). Smoke from the southern burn site has
also been carried northwest to merge with smoke
from the northern burn site. Smoke has left the
area around Points 6-8 along the NE edge of the
southern burn site.
9
The PB-Piedmont smoke plume at 0600 LST 16
February 2006. Plumes from both burns have turned
to blow more toward the west leaving a scattering
of particles over large areas between the burn
sites. The red dots identify where smoke and fog
might coexist.
10
Smoke Location EvaluationInterpreting the
Observations Shown in the Next Slide

• Red symbol means smoke observed
• Green symbol means smoke not observed
• Circle means model matched observation
• Triangle means model unsuccessful

11
PB-Piedmont analysis observations 0700 EST
5 No smoke (0607)
6 No smoke (0612)
7 No smoke (0618)
8 No smoke (0622)
9 Smoke on unit (0624)
10 Slight smoke (0628)
11 Slight smoke (0630)
12 Slight smoke (0635)
13 No smoke (0639)
14 Slight smoke (0648)
15 Smoke (0651)
16 Slight smoke (0656)
17 Slight smoke (0700)
18 No smoke (0702)
19 No smoke (0705)
20 Slight smoke (0708)
21 Slight smoke (0712)
22 Slight smoke (0717)
Click left mouse button to view observation
points.
12
PB-Piedmont analysis observations 0700 EST
23 No smoke (0722)
24 No smoke (0620)
25 Thick smoke (0623)
26 Slight smoke (0625)
27 Thick smoke (0629)
28 Thick smoke (0632)
29 Smoke (0636)
30 Smoke (0638)
31 Slight smoke (0646)
32 Smoke (0651)
33 Smoke (0655)
34 Slight smoke (0658)
35 Light smoke (0700)
36 Slight smoke (0704)
37 Light smoke (0708)
38 No smoke (0713)
39 No smoke (0720)
Click left mouse button to view observation
points.
13
Summary

• PB-Piedmont was correct at 24 of the 35
  smoke/no-smoke observation points.
• Most of the 11 failures occurred near smoke plume
  boundaries or where very little or slight smoke
  was observed.

14
Smoke Matrix
Smoke Prediction
Yes No
5
20
No Yes
Smoke Observation
4
6