AFM-13 Aircraft Flux Analyses

Analysis And Interpretation Of Airborne Flux Observations Over The BOREAS Site

1. INTRODUCTION

For the BOReal Ecosystem-Atmosphere Study (BOREAS) in 1994 and 1996, the 
Airborne Fluxes and Meteorology (AFM) group was involved in measurements (from 
different platforms and heights within the atmospheric boundary layer) of key 
atmospheric variables and several surface-related parameters that can be used to 
describe the evolution of the atmospheric boundary layer and the boundary layer 
fluxes of sensible heat, latent heat, momentum, and CO2. Specifically, the AFM-
13 team was interested in analysis and interpretation of airborne flux 
observations over a 16-km by 16-km grid site in each of the BOREAS study areas. 
The primary data used in the investigation were collected using the Canadian 
Twin Otter aircraft, one among the many research aircraft flown in BOREAS.

The main objectives of the AFM-13 investigations are to use the Twin Otter-based 
data with tower flux data to map spatial and temporal variations in the fluxes 
of heat, moisture, and CO2, and to define realistic footprint functions over the 
BOREAS sites, so that airborne observations are related to the correct ground 
surface with its biological and ecological characteristics. These maps are then
compared to maps of remote sensing observations over the sites.  It is hoped 
that these studies help to develop regional scale models of fluxes of  sensible 
heat, latent heat, and CO2 for global monitoring of climate change.  This 
document presents a brief summary of the Twin Otter grid sites, the measured 
data, the type of analysis carried out, and the preliminary results from the 
1994 Intensive Field Campaigns (IFCs). 


2. PRINCIPAL INVESTIGATOR 

Peter H. Schuepp 
McGill University
Quebec, Canada
(514) 398-7935
(514) 398-7990 (fax) 
pschuepp@NRS.mcgill.ca

3. CONTACT INFORMATION

Segun O. Ogunjemiyo
McGill University
Quebec, Canada
(514) 398-7950
(514) 398-7990 (fax) 
segun@NRS.mcgill.ca



4. DESCRIPTION OF THE GRID SITES 

The Twin Otter grid site of the Northern Study Area (NSA) is shown in the file 
NSA_CLASS.TIF. The surface cover classes of the site were extracted from the 
August 20, 1988 Landsat Thematic Mapper (TM) scene of the 129-km by 86-km NSA, 
using bands 1 through 5 and 7. The spatial resolution is 30 m. Nine of the 11 
cover types defined by the National Aeronautics and Space Administration (NASA) 
classification scheme (Hall et al., 1996) can be identified. More than half of 
the area is covered by wet and dry conifers. The southern half of the site is 
dominated by burn areas, enclosing an unburned stand of primarily black spruce, 
where regeneration has been taking place in the past 7 to 15 years, and there is
a significant density of bare rock outcroppings in the southwest corner.  The 
terrain is generally flat to gently rolling, intersected by a few small 
escarpments and riverbeds. 

The Southern Study Area (SSA) site is shown in the file SSA_CLASS.TIF, which 
represents the land classification distribution based on bands 1 through 5 and 
band 7 of the 06-Aug-1990, Landsat TM scene.  All 11 surface cover types were 
identified within the grid, with wet conifers constituting about 60 percent of 
the surface cover. Effects of logging, with new regeneration, can be seen in the 
vicinity of the young jack pine tower. The area beneath the northwest-southeast 
diagonal generally contains wetter surfaces, with mixed fen and forest. Forest 
cover is often controlled by small changes in relief and soil or drainage 
characteristics. 

5. METHOD OF DATA COLLECTION 

The Twin Otter aircraft of the National Research Council (NRC) of Canada was 
used to collect all the data presented in this documentation.  A full 
description of instrumentation is given by MacPherson (1990, 1996), as detailed 
in AFM-04 documentation. The aircraft was flown in a grid pattern over the two 
study sites in either east-west or north-south trajectories. Flight patterns 
were chosen in response to the prevailing wind direction for closest approach to 
crosswind conditions. Each grid flight consisted of nine parallel straight 
lines, spaced 2 km apart, with each line sampled twice in opposite directions, 
in a sequence that ensured that all grid lines were sampled at the same mean 
time. Flight level was maintained at 30-35 m above ground level. The data were 
digitized at 16 Hz, resulting in a spatial resolution of about 4 m along the 
flight line at the aircraft speed of about 60 m/s. Where necessary, corrections 
were applied for delays between signals recorded at different sensor locations.

Details of the grid flights flown in 1994, including the weather conditions 
under which they were flown, are given in Ogunjemiyo et al. (1997) and can be 
found in the AFM-04 data. A total of 23 grid fights were flown, 14 in the NSA 
and 9 in the SSA. The grid flights were spread over the three IFCs between May 
and September 1994; i.e., covering different physiological and phenological 
stages during the growing season of the boreal forest.  Full descriptions of the 
128 different parameters measured by the aircraft are given in the AFM-04 data. 
The parameters that were used (directly or indirectly) to obtain the results 
summarized here include air temperature, potential temperature, static pressure, 
incident short wave radiation, reflected short wave radiation, the three 
orthogonal components of the wind velocity, H2O mixing ratio, C02 mixing ratio, 
aircraft position, heading, altitude (radar and pressure), radiometric surface 
temperature, and greenness index (GI) (from the Simple Ratio of reflected red to 
infrared radiation).

6. FLUX CALCULATIONS AND MAPPING PROCEDURES

Fluxes of sensible heat (H), latent heat (LE), and CO2 (C) were calculated along 
the flight lines using the eddy correlation method; i.e., estimating the 
covariance between the fluctuations of the vertical wind and the scalar of 
interest. The fluctuations were obtained from the original signals by 
detrending. The choice of the detrending method used was based on a scheme 
developed by Ogunjemiyo et al. (1997) that takes into consideration the physical 
properties of turbulent transports of scalar as observed over forests. It 
assumes that the most intensive turbulent transport occurs in the "gradient 
modes" (i.e., excess-up and deficit-down for H and LE or deficit-up and excess-
down C) during sunlit, daytime convective conditions over photosynthesizing 
areas, and that optimum detrending yields the highest gradient/countergradient 
ratio for the dominant turbulent structures.

To map fluxes, the flux events along the 16-km flight lines were block averaged 
over 2-km windows, with 1-km overlap between adjoining windows, giving 15 data 
points along each flight line. By combining the repeated passes over each grid 
line, a matrix of 135 points per grid is produced, representing a 4-km average 
per data point. Data smoothing is done by cubic spline interpolation, and 
interpolated data are used as input to Geographic Information System (GIS)-based 
IDRISI software to produce the flux maps of the grid sites. Due to the 
variability of individual maps, only composite maps representing all flights 
within the same IFC and over the given site are shown and discussed, as well as 
average maps over all three IFCs. The same procedure is used to make maps of 
surface characteristics, which are surface temperature excess Ts-Ta and GI. The 
reliability of the data to adequately represent the surface over which they were 
measured was evident from the good agreement between the observed surface 
classification at the site and those within the estimated flux 'footprint' 
developed for the sites from tracer gas release studies (Kaharabata et al., 
1997).  

7. DISCUSSIONS

The results presented in the form of composite maps of GI, Ts-Ta, LE, C, and H 
represent sampling within the same IFC, during intervals ranging from 3 to 11 
days, with the exception of IFC-3, where a last sample of the NSA grid was
added 13 days after the initial 7-day period.  The composite maps and the maps 
for each variable averaged over the three IFCs are listed in Section 8. It must 
be mentioned that GI and Ts-Ta were measured from downward looking airborne 
sensors with a narrow view angle, restricting measurements to a strip 
approximately 3 m wide beneath the aircraft.

The surface temperature patterns at the sites are tied to the surface cover 
distributions. At the NSA grid site, the largely forested area in the northern 
half of the site is cooler than the southern half of the site, which is 
dominated by regenerating burn areas. The same trend is observed in the SSA grid 
site, where the NW-SE diagonal structure on the surface classification map is
strongly reflected in the surface temperature signal. The dominant surface 
temperature patterns were persistent throughout the three IFCs, though the range 
of temperature differs significantly from one IFC to another. The GI maps show 
inverse relationship with surface temperature. This relationship is more 
pronounced in the NSA than in the SSA. 

Overall, the most clearly observed relationship between fluxes and surface 
features is that between GI and CO2 flux, with a good correlation between the CO2 
maps and GI maps. By contrast, the relationship between H and Ts-Ta is much more 
complex, with a significant degree of decoupling between
H and Ts-Ta, particularly at the NSA. In the NSA, the main features in sensible
heat flux maps are a local maximum in the NW corner, around the dry old jack 
pine site, and a broader maximum over the much cooler mature black spruce 
stands. Over the SSA grid, sensible heat fluxes are highest in the northern 
part, where the old jack pine sites and various clearings and regenerating 
stands affect the energy balance.

The high value of H over the forest and the decoupling of H from Ts-Ta have also 
been observed by others (e.g., Hall et al., 1995; Mahrt and Sun, 1996). It has 
been partly associated with viewing angle problems of the radiometers, which see 
varying fractions of shaded, cool surface between the often sparse 
trees from which heat is primarily exchanged.  While the latent heat flux shows 
the highest maximum during IFC-2 at both sites, the higher values at the SSA 
could be associated with intense precipitation at the site, prior to the start 
of IFC-2. In the NSA, no characteristic patterns appeared to emerge from the LE 
maps, which are subject to spatially and temporally varying precipitation and 
drying rates. The local maximum in LE at the SSA, in the southeastern region of 
the burn area, may be associated with a relative abundance of shallow surface 
water during most of the IFCs in that area.

The dependence of flux patterns on the relative distribution of surface covers 
is exemplified by the differences between the SSA and the NSA grids. Differences 
in flux patterns from one IFC to another are more likely caused by changes in 
soil moisture distribution and plant physiological properties than by changes in
radiant energy. By providing the maps listed below, it is assumed that the 
general patterns of averaged flux density and their persistence may provide the 
link-up point with geographically referenced models, at the high-resolution end 
of their spatial scales.  

8. LIST OF MAPS 

NORTHERN STUDY AREA GRID SITES:

Classification image of the NSA (nsa_class.tif)
CO2 flux map, IFC-1 (nsa_ifc1_carb.tif) 
CO2 flux map, IFC-2 (nsa_ifc2_carb.tif)
CO2 flux map, IFC-3 (nsa_ifc3_carb.tif) 
Average CO2 flux map for the three IFCs (nsa_avg_carb.tif) 
Greenness index map, IFC-1 (nsa_ifc1_green.tif) 
Greenness index map, IFC-2 (nsa_ifc2_green.tif) 
Greenness index map, IFC-3 (nsa_ifc3_green.tif)
Average greenness index map for the three IFCs (nsa_avg_green.tif)
Sensible heat flux map, IFC-1 (nsa_ifc1_sens.tif)
Sensible heat flux map, IFC-2 (nsa_ifc2_sens.tif)
Sensible heat flux map, IFC-3 (nsa_ifc3_sens.tif)
Average sensible heat flux map for the three IFCs (nsa_avg_sens.tif)
Latent heat flux map, IFC-1 (nsa_ifc1_latent.tif)
Latent heat flux map, IFC-2 (nsa_ifc2_latent.tif)
Latent heat flux map, IFC-3 (nsa_ifc3_latent.tif)
Average latent heat flux map for the three IFCs (nsa_avg_latent.tif)
Surface temperature excess map, IFC-1 (nsa_ifc1_tempx.tif)
Surface temperature excess map, IFC-2 (nsa_ifc2_tempx.tif)
Surface temperature excess map, IFC-3 (nsa_ifc3_tempx.tif)
Average surface temperature excess map for the three IFCs (nsa_avg_tempx.tif)

SOUTHERN STUDY AREA GRID SITE:

Classification image of the SSA (ssa_class.tif)
CO2 flux map, IFC-1 (ssa_ifc1_carb.tif) 
CO2 flux map, IFC-2 (ssa_ifc2_carb.tif)
CO2 flux map, IFC-3 (ssa_ifc3_carb.tif) 
Average CO2 flux map for the three IFCs (ssa_avg_carb.tif) 
Greenness index map, IFC-1 (ssa_ifc1_green.tif) 
Greenness index map, IFC-2 (ssa_ifc2_green.tif) 
Greenness index map, IFC-3 (ssa_ifc3_green.tif)
Average greenness index map for the three IFCs (ssa_avg_green.tif)
Sensible heat flux map, IFC-1 (ssa_ifc1_sens.tif)
Sensible heat flux map, IFC-2 (ssa_ifc2_sens.tif)
Sensible heat flux map, IFC-3 (ssa_ifc3_sens.tif)
Average sensible heat flux map for the three IFCs (ssa_avg_sens.tif)
Latent heat flux map, IFC-1 (ssa_ifc1_latent.tif)
Latent heat flux map, IFC-2 (ssa_ifc2_latent.tif)
Latent heat flux map, IFC-3 (ssa_ifc3_latent.tif)
Average latent heat flux map for the three IFCs (ssa_avg_latent.tif)
Surface temperature excess map, IFC-1 (ssa_ifc1_tempx.tif)
Surface temperature excess map, IFC-2 (ssa_ifc2_tempx.tif)
Surface temperature excess map, IFC-3 (ssa_ifc3_tempx.tif)
Average surface temperature excess map for the three IFCs (ssa_avg_tempx.tif)

The following files contain the values (in tabular form) for the various grid 
points in the maps listed above:

NSA  IFC-1
NSA_94_06-07.CSV
NSA_94-06-08A.CSV
NSA_94-06-08B.CSV
NSA_94-06-10.CSV
NSA-94-06-13.CSV

NSA IFC-2
NSA_94-07-28.CSV
NSA_94-08-01.CSV
NSA_94-08-04.CSV
NSA_94-08-08.CSV

NSA IFC-3
NSA_94-08-31.CSV
NSA_94-09-02.CSV
NSA_94-09-03.CSV
NSA_94-09-06.CSV
NSA_94-09-19.CSV

SSA IFC-1
SSA_94-05-26.CSV
SSA_94_05-31.CSV
SSA_94-06-04.CSV

SSA IFC-2
SSA_94-07-20.CSV
SSA_94-07-21.CSV
SSA_94_07-24.CSV
SSA_94-07-26.CSV

SSA IFC-3
SSA_94-09-13.CSV
SSA_94-09-16.CSV


9. Data Description

9.1 Data Characteristics

Data characteristics are defined in the companion data definition file 
(afm13afr.def).

9.2 Sample Data Record

Sample data format shown in the companion data definition file (afm13afr.def).

10. Data Organization

10.1 Data Granularity

All of the airborne flux data are contained in one dataset.

10.2 Data Format(s)

The data files contain numerical and character fields of varying length 
separated by commas. The character fields are enclosed with a single apostrophe 
marks. There are no spaces between the fields. Sample data records are shown in 
the companion data definition file (afm13afr.def).

11. Data Access 

11.1 Contact Information

Ms. Beth Nelson
NASA GSFC
Greenbelt, MD
(301) 286-4005
(301) 286-0239 (fax)
Elizabeth.Nelson@gsfc.nasa.gov

11.2 Data Center Identification

See Section 9.1.

11.3 Procedures for Obtaining Data

Users may place requests for the data on line or by telephone, electronic mail, 
or fax.

11.4 Data Center Status/Plans

The Analysis And Interpretation Of Airborne Flux Observations Over The BOREAS 
Site are available from the Earth Observing System Data and Information System 
(EOS-DIS) Oak Ridge National Laboratory (ORNL) Distributed Active Archive Center 
(DAAC). The BOREAS contact at ORNL is:

ORNL DAAC User Services
Oak Ridge National Laboratory
(865) 241-3952
ornldaac@ornl.gov
ornl@eos.nasa.gov


12. REFERENCES

Desjardins, R.L., J.I. MacPherson, P.H. Schuepp, and F. Karanja. 1989. An 
Evaluation of Airborne Eddy Flux Measurements of CO2, Water Vapour, and Sensible 
Heat. Boundary-Layer Meteorology 47:55-69.

Desjardins, R.L., P.H. Schuepp, J.I. MacPherson, and D.J. Buckley. 1992. Spatial 
and Temporal Variations of the Fluxes of Carbon Dioxide and Sensible Heat Over 
the FIFE Site. Journal of Geophysical Research 97(D17): 18,467-18,475.

Hall, F.G., K.F. Huemmrich, S.J. Goetz, P.J. Sellers, and J.E. Nickeson. 1992.  
Satellite Remote Sensing Of Surface Energy Balance: Success, Failures, and 
Unresolved Issues in FIFE.  Journal of Geophysical Research 97(D17): 19,061-
19,089.

Hall, F.G., Y.E. Shimabukuro, and K.F. Huemmrich. 1995. Remote Sensing of Forest 
Biophysical Structure Using Mixture Decomposition and Geometric Reflectance 
Models. Ecological Applications 5(4): 993-1013.

Hall, F.G. and D.E. Knapp. 1998. BOREAS TE-18 Landsat TM Maximum Likelihood 
Classification of the NSA. Oak Ridge National Laboratory Distributed Active 
Archive Center.

Hall, F.G. and D.E. Knapp. 1998. BOREAS TE-18 Landsat TM Maximum Likelihood 
Classification of the SSA. Oak Ridge National Laboratory Distributed Active 
Archive Center.

Kaharabata, S.K., P.H. Schuepp, S. Ogunjemiyo, S. Shen, M.Y. Leclerc, R.L. 
Desjardins,  and J.I. MacPherson. 1997. Footprint considerations in BOREAS. 
Journal of Geophysical Research 102(D24): 29,113-29,124.

MacPherson, J.I. 1990. Wind And Flux Calculations On The NAE Twin Otter. Rep. 
Ltr-Fr-109, National Research Council, Ottawa, Ontario, Canada.

MacPherson, J.I. 1996. NRC Twin Otter Operations in BOREAS,  Rep. Ltr-Fr-129, 
National Research Council, Ottawa, Ontario, Canada.

Mahrt, L. and M. Ek. 1993. Spatial Variability of Turbulent Fluxes and Roughness 
Lengths in HAPEX-MOBILHY. Boundary-Layer Meteorology 65: 381-400.

Mahrt, L. and Sun, J. 1996.  Formulation of Heat Flux over the Boreal Forest. 
Proc. 22nd Conference on Agric. and Forest Meteorology, American Meteorology 
Society, Atlanta, GA, Jan. 28 - Feb. 2, 1996, p. 114-117.

Mahrt, L., J.I. MacPherson, and R. Desjardins.  1994. Observations of
Fluxes over Heterogeneous Surfaces. Boundary-Layer Meteorology 67:345-367.

Ogunjemiyo, O.S., P.H. Schuepp, J.I. MacPherson, and R.L. Desjardins. 1997. 
Analysis of Flux Maps vs. Surface Characteristics from Twin Otter Grid Flights 
in BOREAS 1994. Journal of Geophysical Research 102(D24): 29,135-29,145.

Sellers, P., F. Hall, H. Margolis, B. Kelly, D. Baldocchi, G. den Hartog, J. 
Cihlar,  M.G. Ryan, B. Goodison, P. Crill, K.J. Ranson, D. Lettenmaier, and D.E. 
Wickland. 1995. The Boreal Ecosystem-Atmosphere Study (BOREAS): An Overview and 
Early Results from the 1994 Field Year. Bulletin of the American Meteorological 
Society 76(9): 1549-1577.

Sellers, P.J., F.G. Hall, G. Asrar, D.E. Strebel, and R.E. Murphy. 1992.  An 
overview of the First International Satellite Land Surface Climatology Project 
(ISLSCP) Field Experiment (FIFE). Journal of Geophysical  Research 97 (D17): 
18,345-18,371.

Sellers, P.J., F.G. Hall, R.D. Kelly, A. Black, D. Baldocchi, J. Berry, M. Ryan, 
K.J. Ranson, P.M. Crill, D.P. Lettenmaier, H. Margolis, J. Cihlar, J. Newcomer, 
D. Fitzjarrald, P.G. Jarvis, S.T. Gower, D. Halliwell, D. Williams, B. Goodison, 
D.E. Wickland, and F.E. Guertin. 1997.  BOREAS in 1997: Experiment Overview, 
Scientific Results and Future Directions.  Journal of Geophysical Research, 102 
(D24): 28,731-28,770.


13. ACRONYMS

AFM       - Airborne Flux and Meteorology
BOREAS    - BOReal Ecosystem and Atmosphere Study
GI        - Greenness Index
GIS       - Geographic Information System
IFC       - Intensive Field Campaign
NASA      - National Aeronautics and Space Administration
NRC       - National Research Council
NSA       - Northern Study Area
SSA       - Southern Study Area
TM        - Thematic Mapper 
AFM13.doc
03/03/99