Open-Path Eddy-Covariance
System Operator’s Manual
IRGASON, KH20, and FW05
Revision: 4/13
Copyright © 2004-2013
Campbell Scientific, Inc.
Warranty
“PRODUCTS MANUFACTURED BY CAMPBELL SCIENTIFIC, INC. are
warranted by Campbell Scientific, Inc. (“Campbell”) to be free from defects in
materials and workmanship under normal use and service for twelve (12)
months from date of shipment unless otherwise specified in the corresponding
Campbell pricelist or product manual. Products not manufactured, but that are
re-sold by Campbell, are warranted only to the limits extended by the original
manufacturer. Batteries, fine-wire thermocouples, desiccant, and other
consumables have no warranty. Campbell’s obligation under this warranty is
limited to repairing or replacing (at Campbell’s option) defective products,
which shall be the sole and exclusive remedy under this warranty. The
customer shall assume all costs of removing, reinstalling, and shipping
defective products to Campbell. Campbell will return such products by surface
carrier prepaid within the continental United States of America. To all other
locations, Campbell will return such products best way CIP (Port of Entry)
INCOTERM® 2010, prepaid. This warranty shall not apply to any products
which have been subjected to modification, misuse, neglect, improper service,
accidents of nature, or shipping damage. This warranty is in lieu of all other
warranties, expressed or implied. The warranty for installation services
performed by Campbell such as programming to customer specifications,
electrical connections to products manufactured by Campbell, and product
specific training, is part of Campbell’s product warranty. CAMPBELL
EXPRESSLY DISCLAIMS AND EXCLUDES ANY IMPLIED
WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A
PARTICULAR PURPOSE. Campbell is not liable for any special, indirect,
incidental, and/or consequential damages.”
Assistance
Products may not be returned without prior authorization. The following
contact information is for US and international customers residing in countries
served by Campbell Scientific, Inc. directly. Affiliate companies handle
repairs for customers within their territories. Please visit
www.campbellsci.com to determine which Campbell Scientific company serves
your country.
To obtain a Returned Materials Authorization (RMA), contact CAMPBELL
SCIENTIFIC, INC., phone (435) 227-9000. After an applications engineer
determines the nature of the problem, an RMA number will be issued. Please
write this number clearly on the outside of the shipping container. Campbell
Scientific’s shipping address is:
CAMPBELL SCIENTIFIC, INC.
RMA#_____
815 West 1800 North
Logan, Utah 84321-1784
For all returns, the customer must fill out a “Statement of Product Cleanliness
and Decontamination” form and comply with the requirements specified in it.
The form is available from our web site at www.campbellsci.com/repair. A
completed form must be either emailed to repair@campbellsci.com or faxed to
(435) 227-9106. Campbell Scientific is unable to process any returns until we
receive this form. If the form is not received within three days of product
receipt or is incomplete, the product will be returned to the customer at the
customer’s expense. Campbell Scientific reserves the right to refuse service on
products that were exposed to contaminants that may cause health or safety
concerns for our employees.
Table of Contents
PDF viewers: These page numbers refer to the printed version of this document. Use the
PDF reader bookmarks tab for links to specific sections.
1. System Description ....................................................1
1.1 OPEC (CSAT3 Only)...........................................................................1
1.2 Basic OPEC..........................................................................................1
1.3 Extended OPEC ...................................................................................2
1.4 Additional Fast Response Sensors .......................................................2
2. Installation and Mounting ..........................................2
2.1 Fetch and Sensor Height ......................................................................4
2.2 Mounting..............................................................................................4
2.2.1 Measure Sonic Azimuth................................................................4
2.3 Wiring ..................................................................................................4
2.4 Power ...................................................................................................4
3. System Datalogger Program......................................5
3.1 Program Configuration.........................................................................6
3.1.1 Sonic Azimuth ..............................................................................6
3.1.2 Sensor Configuration ....................................................................8
3.2 Loading a Program to the Datalogger ..................................................9
3.2.1 Direct Connection via LoggerNet .................................................9
3.2.2 Remote via PC/CF Card................................................................9
3.3 System Operation...............................................................................10
3.3.1 Monitoring Data..........................................................................10
3.3.2 Status Table.................................................................................10
4. Data ............................................................................ 11
4.1 Data Retrieval ....................................................................................12
4.1.1 Direct Connection Data Retrieval via LoggerNet .......................13
4.1.2 File Management with Baler.......................................................15
4.1.3 Remote Data Retrieval via a PC/CF Card...................................17
4.1.4 File Management with CardConvert...........................................18
4.1.4.1 Collecting Data with One Card ........................................21
4.1.4.2 Collecting Data with Two Cards ......................................22
4.2 Data Processing..................................................................................24
4.2.1 Online Processing .......................................................................24
4.2.2 Off-line Processing with EdiRe ..................................................24
4.2.2.1 Creating Raw File Format and Processing Lists ..............25
4.2.2.2 Example EdiRe Raw File Format and Processing Lists ...27
5. Basic Eddy-Covariance Theory...............................31
i
Table of Contents
Appendices
CSAT3 Orientation ..................................................A-1
A.
A.1 Determining True North and Sensor Orientation ............................ A-1
A.2 Online Magnetic Declination Calculator ......................................... A-3
B. References...............................................................B-1
C. OPEC200 Open Path Eddy Covariance System
Quickstart Guide................................................... C-1
Figures
2-1. Open-Path Eddy-Covariance (IRGASON) System ............................. 3
3-1. IRGASON right hand coordinate system, horizontal wind vector
angle is 0 degrees............................................................................. 7
3-2. Compass coordinate system, compass wind direction is 140
degrees ............................................................................................. 8
4-1. LoggerNet station setup for the “flux” table...................................... 13
4-2. LoggerNet station setup for the “ts_data” table................................. 14
4-3. LoggerNet station data collection schedule....................................... 14
4-4. LoggerNet station status monitor ...................................................... 15
4-5. Baler station setup for the “flux” table .............................................. 16
4-6. Baler station setup for the “ts_data” table ......................................... 16
4-7. File format flow................................................................................. 18
4-8. Destination File Option screen .......................................................... 19
4-9. Fully configured CardConvert start up screen................................... 20
4-10. List of files created by CardConvert.................................................. 21
4-11. List of files created by CardConvert with duplicates ........................ 22
4-12. List of files where the duplicate files are renamed to *.bak .............. 22
4-13. List of files collected the first time using two cards .......................... 23
4-14. List of files collected the second time using two cards ..................... 23
4-15. Interpreter settings to read a Campbell Scientific TOB1 data file..... 26
4-16. Folder that contains the raw TOB1 time series data files .................. 26
4-17. Completed Interpreter screen............................................................. 27
4-18. Estimated sample frequency and correct sample frequency .............. 27
4-19. Default EdiRe processing list created by the Interpreter ................... 28
4-20. Output file location as part of the processing list .............................. 29
4-21. Output file location as part of the processing list .............................. 29
4-22. Processing Uz and CO2 with 1 Chn Statistics instruction .................. 30
4-23. Computing CO2 flux with 2 Chn Statistics and graphing Uz
statistic with Plot Value instruction ............................................... 30
5-1. Ideal vertical profiles of virtual potential temperature and specific
humidity depicting all the layers of the atmospheric boundary
layer ............................................................................................... 31
A-1. Magnetic declination for the conterminous United States (2004) ... A-1
A-2. A declination angle East of True North (positive) is subtracted
from 360 (0) degrees to find True North...................................... A-2
A-3. A declination angle West of True North (negative) is subtracted
from 0 (360) degrees to find True North...................................... A-2
A-4. Online magnetic declination calculator with inputs and output
for Longmont, CO........................................................................ A-3
ii
Tables
Table of Contents
2-1. Nominal Sensor Power Requirements..................................................5
2-2. Nominal Datalogger Power Requirements with the Display Off
and No RS-232 Communications .....................................................5
2-3. Nominal Datalogger Power Requirements with the Display and
Backlight On, and No RS-232 Communications..............................5
iii
Table of Contents
iv
OPEC Open-Path Eddy-Covariance
System
This document will serve as a guide to properly install and operate a Campbell Scientific
Open-Path Eddy-Covariance System (OPEC). The OPEC is composed of various
products, e.g., dataloggers, fast response turbulence sensors, slow response meteorological
sensors, and software. These products are manufactured by Campbell Scientific and other
vendors. Manuals for each of these sensors shipped with the system. It is time well spent
reviewing these documents.
Before deploying the equipment in the field, Campbell Scientific Inc. recommends that it be
installed in the lab or outside near the office. Doing so will give the station operator an
opportunity practice installing the equipment, confirming that all the required parts are
available, and practice retrieving data from the system. It is much more convent and less
expensive to trouble shoot the station near the office.
The literature contains information that spans 50 years on Eddy-Covariance theory and
measurements. Section 5 briefly touches Eddy-Covariance theory. For more details on
Eddy-Covariance measurements and data analysis, see the literature.
1. System Description
The Campbell Scientific Open-Path Eddy-Covariance (OPEC) system
measures sonic sensible heat flux, momentum flux, and the flux of other scalars
between the atmosphere and earth’s surface. The system consists of a
datalogger, fast response three-dimensional sonic anemometer, and fast
response scalar sensors. Horizontal wind speed and direction are computed by
the datalogger from the three-dimensional measurements of wind made by the
sonic anemometer.
1.1 OPEC (CSAT3 Only)
The minimum components required to make eddy-covariance measurements
are a datalogger, a CSAT3 three-dimensional sonic anemometer, and a
temperature and humidity probe. This system configuration measures sonic
sensible heat flux, momentum flux, temperature, humidity, horizontal wind
speed, and wind direction. This system configuration is used to compute eddy
diffusivity required to compute fluxes of trace gases measured with a gradient
system (Warland, et al., 2001).
1.2 Basic OPEC
A more typical eddy-covariance system consists of a datalogger, an IRGASON
(integrated gas analyzer and sonic anemometer). With this configuration, the
system can measure carbon dioxide flux, latent heat flux, sonic sensible heat
flux, momentum flux, a sensible heat flux (using sonic temperature corrected
for humidity), temperature, humidity, horizontal wind speed, and wind
direction.
1
OPEC Open-Path Eddy-Covariance System
1.3 Extended OPEC
Energy balance sensors can be added to a basic OPEC system to also measure
the net radiation, soil heat flux, soil temperature, and soil water content. The
sensors required for these additional measurements are a NR-LITE, CNR2,
NR01 or CNR4 net radiometer; two or four HFP01 or HFP01-SC soil heat flux
plates, one or two TCAV averaging soil temperature probes, and one or two
CS616 or CS650 soil moisture reflectometers. The number of soil sensors will
vary with the site. A heterogeneous site will require more sensors to get
adequate spatial representation of the site. The datalogger can accommodate
the additional measurements with a multiplexer. Energy balance sensors no
longer carried by Campbell Scientific are also supported by the extended
OPEC datalogger program, including the Q7.1 and CNR1 net radiometers, and
HFT3 soil heat flux plates.
1.4 Additional Fast Response Sensors
If the application requires a direct measurement of sensible heat flux, a FW05
can be added to the system. In the absence of a FW05 or if the FW05 breaks,
the sensible heat flux can be found using the sonic temperature corrected for
humidity.
A KH20 Krypton hygrometer, instead of the IRGASON gas analyzer, can be
used to measure the latent heat flux. The KH20 can not be used to measure an
absolute concentration of water vapor, because of scaling on the source tube
windows caused by disassociation of atmospheric continuants by the ultra
violet photons (Campbell and Tanner, 1985 and Buck, 1976). A slow response
humidity sensor is used to measure the absolute humidity required to compute
the air density. Air density is needed to scale the covariances into sensible heat
flux and momentum flux. The rate of scaling on the KH20 windows is a
function of the atmospheric humidity. In high humidity environments, scaling
can occur within a few hours. That scaling attenuates the signal and can cause
shifts in the calibration. However, the scaling over a typical flux averaging
period is small. Thus, water vapor fluctuation measurements can still be made
with the hygrometer. The effects of the scaling can be easily reversed by
wiping the windows with a moist swab.
2. Installation and Mounting
When making eddy-covariance measurements near the surface (less than 3
meters), mount the datalogger enclosure between the legs of the CM106 tripod,
on a separate tripod, or user-supplied drive stake. Also, mount any sensor
electronics boxes as far from the fast response sensors as possible and always
use the tripod guy kit. This will minimize potential flow distortions and tower
sway caused by wind blowing against the fiberglass enclosure. See the tripod
manuals for detailed installation instructions.
FIGURE 2-1 depicts a typical Open-Path Eddy-Covariance station. Point the
Eddy-Covariance sensors into the prevailing wind to minimize the flow
distortion from the tower, mounting hardware, and other sensors.
2
OPEC Open-Path Eddy-Covariance System
TIP
Keep a log book for each station. Record the date and personnel
name for all site visits, as well as all maintenance and work that
is performed during the site visit. Document the condition of the
sensors and site with a digital camera.
FIGURE 2-1. Open-Path Eddy-Covariance (IRGASON) System
3
OPEC Open-Path Eddy-Covariance System
2.1 Fetch and
2.2 Mounting
2.2.1 Measure So
Sensor Height
The eddy-covariance sensors must be mounted at some height to ensure that
the measurements are made within the
layer grows at a rate of approximately 1 vertical meter per 100 horizontal
meters. Thus, a height to fetch (horizontal distance travelled) ratio of 1:100
may be used as an absolute bare minimum rough rule of thumb for determini
the measurement height.
The fetch should be homogenous and flat, and no abrupt changes in vegeta
height should exist (Tanner, 1988). Consider two adjacent fields, the first
planted with 1 m tall cor
Covariance sensors mounted at 2 m above the corn field should have a
minimum of 200 m of fetch in all the directions that the data is of interest,
particularly between the eddy-covariance sensors and the interface between the
corn and soybean field.
The IRGASON is mounted to a tripod or tower using a horizontal mounting
arm (CM20x) and
IRGASON manua
EC100 electronics.
n and the second with 0.5 m soybean. Eddy-
a leveling mounting kit (CM250). Section 3 of the
l contains detailed information on mounting the sensors and
nic Azimuth
local surface layer. The local surface
tion
ng
NOTE
2.3 Wiring
2.4 Power
To compute the correct compass wind direction, th
the negative x-axis azimuth of the sonic into the program variable
“sonic_azimuth”. If the sonic is installed such that it points into the
wind, the negative x-axis is pointing into the prevailing wind. Take
azimuth of the negative x-ax
Don’t forget to account for the magnetic declination at the site;
see Appendix A for details.
A Campbell Scie
configurations and utilize several different dataloggers. It is impractical to
document the different wiring schemes in this manual. However, each custom
datalogger program (p/n 18442 or 18443) ships with a complete and detailed
wiring diagram.
The system requires about 1.5 to 6 W continuous power, depending on the
atalogger and sensor configuration. The approximate power requirements of
d
various key components, for a system running at 10 Hz, are listed in TABLE
2-1, TABL
ntific Open-Path Eddy-Covariance system can take on several
E 2-2, and TABLE 2-3.
is (prevailing wind).
e station operator must enter
prevailing
a compass
4
OPEC Open-Path Eddy-Covariance System
TABLE 2-1. Nominal Sensor Power Requirements
Sensor Power (mW)
IRGASON 400 mA @ 12.5 Vdc
IRGASON (gas head powered 240 mA @ 12.5 Vdc
down)
KH20 10 - 20 mA @ 12.5 Vdc
HMP45C <3.8 mA @ 12.5 Vdc
TABLE 2-2. Nominal Data r Requirements wit
the Display Off a ommunications
Datalogger Power (mW)
CR1000 w/ CFM100 & CF card 8 mA @ 12.5 Vdc
CR3000 w/ CFM100 & CF card 39 mA @ 12.5 Vdc
CR5000 w/ PC/CF card 63 mA @ 12.5 Vdc
TABLE 2-3. Nominal Data r Requirements wit
t t On, and No R ns
he Display and Backligh S-232 Communicatio
Datalogger Power (mW)
CR1000 w/ CFM100 & CF card 109 mA @ 12.5 Vdc
CR3000 w/ CFM100 & CF card 75 mA @ 12.5 Vdc
CR5000 w/ PC/CF card 170 mA @ 12.5 Vdc
The OPEC system is powered by an external power supply system. The power
supply can be a Campbell Scientific PS84, a third party solar power supply
system, or a user supplied power system.
With a Campbell Scientific PS84 solar power system, connect the OPEC
sensor power cables to the positive and negative terminals of the power su
For a third party power system, see the specific power supply system manual
for installation and connection instructions.
logger Powe h
nd No RS-232 C
logger Powe h
pply.
3. System Data
Power connections are listed in the text files shipped with the datalogger
programs (p/n 18442 or 18443). Be sure the datalogger has a good earth
ground to provide maximum protection against voltage surges due to prim
and secondary lightning strikes. Cam
dataloggers in the field be earth grounded. All components of the system
(datalogger, sensors, external power s
referenced to one common earth gro
has frequent lightning strikes, spark gaps may be required to protect the
datalogger from transient voltages.
pbell Scientific recommends that all
upplies, mounts, housing, etc.) must be
und. When long cables are used or a site
ary
logger Program
The open-path eddy-covariance datalogger program integrates all of the
in an open-path eddy-covariance station into a single system. Part number
18442 is a program for a basic system and p/n 18443 is a program for an
extended system (with energy balance sensors). Each program is custom
written for the order. If your order did not include either p/n 18442 or 18443
sensors
5
OPEC Open-Path Eddy-Covariance System
as a line item on the order, contact Campbell Scientific to purchase the
appropriate datalogger program for your open-path eddy-covariance system.
The program covers a variety of sensors and is continuously growing. If your
system uses sensors that are not currently supported by the program, add the
appropriate measurement and processing instructions to the program or contact
Campbell Scientific for assistance. Campbell Scientific charges for custom
datalogger programming with a one hour minimum and one hour resolution;
call for current Application Engineering time rates.
The datalogger program aligns the measurements in time from the CSAT3,
IRGASON, KH20, and FW05 before computing online fluxes. The IRGASO
delay is a function of the user programmed bandwidth setting. The delay as a
function of bandwidth is
are removed before the data is used to compute the online fluxes. These fixed
delays are computed automatically by the datalogger program, based in the
IRGASON bandwidth setting. The default IRGASON bandwidth is 20 Hz.
The sensor delays are not removed from the high frequency time series data
saved to Final Storage.
Depending on the system configuration, the datalogger programs compute
carbon dioxide flux, latent heat flux, sonic sensible heat flux, s
flux, momentum flux, and friction velocity, along with all the second moment
covariances, standard d
sensible heat flux usin
vapor. This sensible heat flux can also be found from the sonic sensible heat
flux and latent heat flux in post processing (see Appendix B).
N
given in the IRGASON manual. The sensor delays
ensible heat
eviations, and means. The program will also compute a
g the sonic temperature corrected for the effects of water
3.1 Program C
3.1.1 Sonic Azimu
onfiguration
The site attendant must enter unique calibrati
information into the datalogger program. This information includes, but is not
limit
ed to, the calibration coefficients for the net radiometer and soil heat flux
plate
s, or site pressure. To find the section of the program where these changes
are made, search for the text “Unique”. These values are added to the program
before it is downloaded into the datalogger.
th
The example programs report the wind direction in both the sonic coordin
system (a right-handed coordinate system) and in the compass coordinate
system (a left-handed coordinate system). The sonic coordinate system is
relative to the sonic itself and does not depend on the sonic’s orientation
(azimuth of the negative x-axis). The compass coordinate s
the earth. In order for the program to compute the correct compass wind
direction, the azimuth of the sonic negative x-axis must be entered into the
program. The program default value for the variable “sonic_azimuth” is 0.
This assumes that the prevailing wind is from the North, e.g., the sonic is
mounted such that the negative x-axis points to the North.
The variable “sonic_azimuth” can be changed on the fly using the CR3000
keyboard display or via LoggerNet. Once a “sonic_azimuth” is entered, the
value is saved in the CPU. The program default “sonic_azimuth” is used on
if a site specific value has not been entered using the keyboard or through
LoggerNet.
on coefficients or site-specific
ate
ystem is fixed to
ly
6
OPEC Open-Path Eddy-Covariance System
To enter the “sonic_azimuth” using the datalogger keyboard, press the <Esc>
key until the Campbell Scientific Inc. logo is visible in the upper left hand
corner of the display. Press the <Enter> key. Select the menu “Sonic_Azmth”
and press the <Enter> key. Enter the azimuth of the negative x-axis and press
the <Enter> key. Press the <Esc> several times so that the Campb
ell Scientific
logo if visible.
To enter the “sonic_azimuth” using LoggerNet, connect to the datalogger with
LoggerNet and display the Public data table. Locate the variable
“sonic_azimuth” and click on it. Double click on the number field,
negative x-axis azimuth and
press the <enter> key.
enter the
NOTE
Don’t forget to account for the magnetic declination at the site;
see Appendix A for details.
FIGURE 3-1 and FIGURE 3-2 show the orientation of an IRGASON azimuth
bearing of 140 degrees, e.g., the negative x-axis i s pointing into 140 degrees.
If the wind is blowing into the sonic, from the negative x-axis to the positive xaxis (from the transducers to the block), the horizontal wind vector angle
(wind_dir_csat3) is 0 degrees (wind vector) and the compass wind direction
(wnd_dir_compass) is 140 degrees (wind vane).
Y
X
Wind Vane
Wind Vector
North
FIGURE 3-1. IRGASON right hand coordinate system, horizontal wind
vector angle is 0 degrees
7
OPEC Open-Path Eddy-Covariance System
X
Y
FIGURE 3-2. Compass coordinate system, compass wind direction is
140 degrees
North
Wind Vane
Wind Vector
3.1.2 Sensor Configuration
The IRGASON communicates with Campbell Scientific dataloggers using a
digital communication protocol called Synchronous Device for Measurement
(SDM). Each sensor connected to the SDM bus has a unique address. A
maximum of 15 addresses are allowed by the protocol and each sensor must
have a unique SDM address.
CAUTION
Do not use SDM address F (15) because it is reserved for
use with the Group Trigger instruction.
The IRGASON is shipped from the factory is set with a default SDM address
of 1. This address is set in software and can be changed using the software
DevConfig or ECMon and a USB cable. DevConfig is sensor/peripheral
support software shipped with LoggerNet. ECMon is PC support software
shipped with the IRGASON and is also available in the downloads section
online at www.campbellsci.com.
The program sets the IRGASON bandwidth to 20 Hz. The instrument delay is
a function the bandwidth setting and is automatically computed by the
datalogger program. To use a bandwidth other than 20 Hz, change the program
constant “BANDWIDTH” to the appropriate value. For more information on
IRGASON bandwidth and delay see the IRGASON manual.
NOTE
The settings are sent to the IRGASON each time it is powered on
and only if connected to the datalogger via SDM. The settings
are not saved onto the IRGASON’s non-volatile memory.
8
OPEC Open-Path Eddy-Covariance System
CAUTION
If you loan an IRGASON to a colleague, verify that the
address is 1 before deploying the system into the field.
3.2 Loading a Program to the Datalogger
Before the datalogger can begin to make measurements, a program must be
transferred into its CPU. The program can be transferred using a PC and
LoggerNet or using PC/CF cards.
3.2.1 Direct Connection via LoggerNet
The datalogger program can be transferred to a datalogger using a PC,
LoggerNet, and some sort of interface. In the eddy-covariance application, the
most common interface is the RS-232 cable.
TIP
To avoid potential problems (mixed data and connectivity
issues), configure LoggerNet with one station file per datalogger,
even if the data is retrieved using cards and the computer is
connected to multiple dataloggers for monitoring purposes only.
See the LoggerNet manual Section 4.2.
Set up a station in the LoggerNet network map; see Section 4.1 in the
LoggerNet manual for details. To transfer a program to the datalogger, start
LoggerNet. Click on the “Connect” button in the Toolbar. Select the station
and click on the “Connect” button. Check the datalogger time, it is located in
the upper right hand side of the Connect Screen. If the datalogger and PC time
differ by more than a few seconds, set the datalogger time by clicking on the
“Set Station’s Clock” button. To download a program, click on the “Send”
button. Navigate to the folder where the program is saved, select it, and click
on “OK”.
3.2.2 Remote via PC/CF Card
The datalogger program can be transferred to a datalogger using a PC/CF card.
The CRBasic dataloggers reserve 10% or 80 Kbytes of space, whichever is
smaller, on cards to store programs. Copy the program onto the card using
Windows Explorer.
CAUTION
Use only PC/CF Cards sold by Campbell Scientific Inc.
Although consumer versions of PC/CF cards will fit into the
card slot and operate, only the PC/CF cards carried by
Campbell Scientific have passed our temperature and
electrostatic discharge (ESD) testing.
The following instructions assume a basic familiarity in the operation of a
CRBasic datalogger keyboard; see the datalogger manual for details on using
the keyboard.
Insert the card into the datalogger card slot. Press the <Enter> key. If
necessary, select the “System Menu” menu and press the <Enter> key. Select
the “File” menu and press the <Enter> key. Select the “Copy” menu and press
the <Enter> key. The cursor should be at the “From” line. Press the <Enter>
9
OPEC Open-Path Eddy-Covariance System
key, select “CRD:” and again press the <Enter> key. The datalogger will now
display a list of program files on the card.
Navigate to the eddy-covariance program and press the <Enter> key. The
cursor will jump to the “To” line. Press the <Enter> key and then the right
arrow, select CPU:filename. Press the <Esc> key and down arrow. Finally,
press the <Enter> key to execute the copy from the card to the CPU.
To start the program, press the <Esc> key until the main menu appears and
Campbell Scientific logo appear in the upper left hand side. Press the <Enter>
key, if necessary, select the “System Menu” and press the <Enter> key. Select
the “Run/Stop Program” menu and press the <Enter> key. Tag the “Restart,
Delete Data” option, navigate to “Execute” and press the <Enter> key. Select
“Yes” and press the <Enter> key.
CAUTION
Do not run a program from the PC/CF card if the card will
be removed and replaced with another.
3.3 System Operation
Once the program is downloaded to the datalogger, compiled, and the PC/CF
card has been formatted, the datalogger will begin to communicate with the
SDM sensors and measure the analog sensors. At the top of the next minute, it
will start writing the raw time series data to the PC/CF card (the LED will
begin to pulse at a regular interval). Verify that all the sensors are wired and
measured correctly by monitoring the Public data table. Check the Status data
table for compile and runtime errors (see Section 3.3.2).
3.3.1 Monitoring Data
Monitor the instantaneous measurements and diagnostics by viewing the
“Public” table. The “Public” tables can be monitored using either LoggerNet
(see the LoggerNet manual for details) or via the keyboard and display (see the
datalogger manual for details). The online flux computations can be viewed by
monitoring the “Flux” table with either LoggerNet or the keyboard and display.
The historical online “Flux” data can be monitored in “Final Storage Data”
menu.
3.3.2 Status Table
The “Status” table contains useful information about the performance and
status of the system (see Appendix A in the datalogger manual). To view the
“Status” table with LoggerNet, go to the Connect Screen and select the Tools |
Status Table … menu (see Section 5.1.10 in the LoggerNet manual). With the
keyboard display, follow this menu path System Menu | Status. Useful
information in the “Status” table includes, but is not limited to:
OSVersion: Version and revision of the CRBasic datalogger operating
system. As of the printing of this document, Campbell Scientific recommends
using, CR3000 OS v25.0, CR1000 OS v25.0, or CR5000 OS v6.0 or greater.
WatchdogErrors: Number of times the watch dog timer reset the datalogger.
Normally, this count should be 0.
10
OPEC Open-Path Eddy-Covariance System
CompileResults: Reports compile errors.
VarOutOfBound: An element of an array was referenced that does not exist,
e.g., a VarOutOfBound is reported if element wind(9) is referenced in the
program, but the array was defined with 5 elements.
SkippedScan: If the maximum number of buffered scans is exceeded (defined
in the third parameter of the Scan () instruction), the Processing and
Measurement tasks are resynchronized. The resynchronization results in a
number of SkippedScan equal to the buffer depth.
DataFillDays(): The number of days to fill a table in both the CPU and CRD
memory. The number of days to fill a table is reported in the same order they
are found in the “Status” table, e.g., the same order they are defined in the
program.
CardStatus: Indicates if a card is used by the program or not.
ProcessTime: Time, in microseconds, it takes the datalogger to complete all
the processing. In the Sequential mode, this value must always be less than the
Scan Interval. In the PipeLine mode, this value can occasionally be greater
than the Scan Interval, e.g., when the 30 minute fluxes are computed. With the
Eddy-Covariance systems, always use the Pipeline mode.
4. Data
NOTE
BuffDepth: The current number of Scans that the datalogger Processing tasks
have fallen behind the Measurement tasks. In the Pipeline Mode, the
datalogger processing task can periodically fall behind the Scan Interval.
MaxBuffDepth: The maximum number of Scans that the Processing tasks
have fallen behind the Measurement tasks.
Each system requires an Open-Path Eddy-Covariance program (p/n 18442 or
18443). The data table outputs of the programs are described in a Microsoft
Excel document (shipped with the program). The description includes the
variable name and units. There is one sheet (tab) per output.
The datalogger programs are protected under U.S. and
International copyright laws. Do not distribute the datalogger
programs. See the End User License Agreement (EULA) shipped
with the programs for more information on your rights and
obligations.
Data is saved on the PC/CF card as a binary file in a Campbell Scientific
format called Table Oriented Binary Format 3 (TOB3). TOB3 incorporates
features to improve reliability of the card and allows for the accurate
determination of each record’s time without the space required for individual
time stamps.
®
Before the data can be used, it must be converted to a Table Oriented Binary
Format 1 (TOB1), Table Oriented ASCII Format 5 (TOA5), or Array
Compatible CSV file. In all of these formats, the time stamp is written to each
record. In the TOB1 format, the time stamp is reported as the number of
11
OPEC Open-Path Eddy-Covariance System
seconds and nanoseconds since 0000 hrs, 1 Jan 1990. In the TOA5 format, the
time stamp is a quoted string, similar to that used in Microsoft
Array Compatible CSV, time is reported as elements of the array, e.g., in the
same format as the Campbell Scientific mixed array dataloggers.
NOTE
Campbell Scientific recommends the use of TOB1 file format.
This file format can be readily read into third party post
processing software, e.g., EdiRe or MatLab. If required, a TOB1
flux file can be converted into a TOA5 flux file using
CardConvert. A TOA5 file is comma separated values and
easily read into Microsoft
Both the TOB1 and TOA5 formats contain an ASCII header. This header
contains information about the datalogger used to collect the data. The header
also describes the data with variable names and units. For more information on
the file formats, see Section 2.4 in the datalogger manual.
For backwards compatibility with mixed array dataloggers, comma separated
values, without header information, is also supported with LoggerNet and
CardConvert. Elements that contain an array ID, day of year, hour and
minutes, can be added to the first few columns of the data file. This format is
supported in LoggerNet version 3.3 and CardConvert version 1.2 or greater.
4.1 Data Retrieval
®
Excel as a CSV file.
®
Excel. In the
TIP
Data can be collected from a datalogger using both LoggerNet and a direct
connection, or by physically moving the card from the datalogger to the
computer. Using LoggerNet and a PC as a data retrieval option is only
practical if the PC will be located at the site and is continuously polling the
datalogger because the volume of data is such that it takes about 8% to 25% of
real time to download 10 Hz time series data.
Transferring data using the PC/CF card is relatively fast; however, it does
require manual intervention to manage the files. A 1024 Mbyte file (about 28
days of 10 Hz time series from a basic open-path eddy-covariance system) will
take about 15 minutes to copy from the card to a PC hard drive.
A basic Open-Path Eddy-Covariance system will collect about 40 Mbytes of 10
Hz time series data per day. A strategy for maintaining manageable file sizes
is to break them up into smaller files that cover time periods ranging from 1
hour to 1 day. This can be achieved with LoggerNet and the Baler, or using
CardConvert. Both the Baler and CardConvert support naming these smaller
files using the time stamp of the first record in the file.
A group file renaming utility can be useful to manage files from
multiple stations. These utilities are commonly used in the
photography industry to manage image files and are readily
available on the web, ranging from freeware, shareware, and
commercial versions. Any one of these will meet the data file
management needs, as long as it can find and replace text within
a file name.
12
OPEC Open-Path Eddy-Covariance System
4.1.1 Direct Connection Data Retrieval via LoggerNet
LoggerNet can be used to automatically collect and organize data from an
Open-Path Eddy-Covariance system. This approach is some what limited
because of the slow throughput into the PC’s RS-232 port. For all practical
purposes, the PC must remain continuously connected to the datalogger. It will
take about 8% to 25% of real time to collect 10 Hz time series data, e.g., it will
take between 5 to 15 minutes to download 60 minutes of 10 Hz time series
data. Using a PC and LoggerNet to “milk” data from remote sites is not
practical.
For sites where a PC is continuously connected to the datalogger via RS-232,
MD-485 short haul modems, Freewave spread spectrum radios, Ethernet
connection, or other “direct connection”, LoggerNet and the Baler can be used
to collect and manage the data.
To create a new station file see Section 4.2 of the LoggerNet manual.
Configure the station’s “Data Files” tab as shown in FIGURE 4-1 and FIGURE
4-2. Note that the “File Output Option” is “No Output File”. LoggerNet will
collect the data from the data table and place it in the cache. The cache is
where data collected from the datalogger is stored and made available for use
by client software, e.g., the Baler. Client software is run on the host PC or a
remote PC. For more information, see Section 13.1 of the LoggerNet manual.
FIGURE 4-1. LoggerNet station setup for the “flux” table
13
OPEC Open-Path Eddy-Covariance System
FIGURE 4-2. LoggerNet station setup for the “ts_data” table
In the “Schedule” tab, configure the station’s data collection as shown in
FIGURE 4-3. Modify the Primary Retry Interval, Number of Primary Retries,
and the Secondary Retry interval as needed for the specific telecommunications
option used at the station. For example, a site using Freewave spread spectrum
radios or an Ethernet connection may require a long interval between retries or
more frequent retry attempts because of other background traffic.
FIGURE 4-3. LoggerNet station data collection schedule
14
OPEC Open-Path Eddy-Covariance System
To enable LoggerNet’s data collection, click on the Status Monitor button in
the Toolbar. Select the station and click on the “Toggle On/Off” button. The
time of the next data collection will appear in the “Next Data Call” column (see
FIGURE 4-4). To display additional station information, right click on any
column header and then Select “Column…”. For more information, see
Section 6.1 Status in the LoggerNet manual.
FIGURE 4-4. LoggerNet station status monitor
4.1.2 File Management with Baler
The data is collected by LoggerNet and placed in the LoggerNet cache. Two
instances of the Baler, one for the time series data and the other for the flux
data, will pull the data from the cache and “bale” it into a user specified file
size (see FIGURE 4-5 and FIGURE 4-6). Campbell Scientific recommends
using a one hour bale size for time series data that will be post processed by
EdiRe and a 1 day bale size for the flux data. To run more than one instance of
the Baler, use the “/WorkDir=pathname” switch and two shortcuts, one for the
time series data and the other for the flux data (see Section 6.5 in the Baler
manual).
TIP
While the system is set up in the lab or outside near the lab,
practice not only operating the system and data collection, but
also file management and data processing. Only through this
type of experimentation will you determine the file sizes and
procedures that work best for you.
15
OPEC Open-Path Eddy-Covariance System
FIGURE 4-5. Baler station setup for the “flux” table
FIGURE 4-6. Baler station setup for the “ts_data” table
16
OPEC Open-Path Eddy-Covariance System
4.1.3 Remote Data Retrieval via a PC/CF Card
To transfer data manually from the PC/CF card to the PC, remove the card
from the datalogger following the proper card removal procedure (see the
appropriate datalogger manual for details). Insert a fresh card into the
datalogger or copy the data from the card to a working directory on the
computer. Transferring data using the PC/CF card is relatively fast. A 1024
Mbyte file (about 28 days of 10 Hz time series data) will take about 15 minutes
to copy from the card to a PC hard drive.
While the PC/CF card is not in the datalogger, the datalogger will continue to
collect and save the data in the CPU. When the card is returned to the
datalogger, the data will be transferred automatically from the CPU to the card.
NOTE
CAUTION
When a card is inserted into a CRBasic datalogger, the
datalogger will transfer all the data that was collected since the
card was removed. It is good practice to avoid keyboard and RS232 activity during this time. After the datalogger is finished
transferring the data to the card, the LED will blink at a steady
frequency.
The amount of CPU storage, for time series data, varies from data logger to
datalogger. The Status Table reports the time it takes to fill the CPU and CRD
DataTables in the “DataFillDays” fields. The third DataFillDays field is the
fill time for the CPU time series data table.
Never remove a PC/CF card from the datalogger or turn off
power to the datalogger without first shutting down the
card. With a CR3000 or CR1000, press the “Removal
Button” on the CFM100 Compact Flash/NL115
Ethernet/Compact Flash module. With a CR5000 select
the PCCard|Remove Card menu.
Data can be retrieved from a datalogger using one or two cards. If a single card
is used, and data on the card is not deleted after copying it to the PC, the data
from the CPU will be appended to the file on the card when it is returned to the
datalogger. Using a single card in this fashion allows the site attendant to
create files of uniform size.
The disadvantage of this approach, if the time between site visits is
significantly less than the DataTable file time, is that there will be redundant
data copied, along with the new data. This approach is not practical if the PC
can not be brought to the site or the time it takes to copy the data from the card
to the PC is greater than the time it takes to fill the CPU time series DataTable.
As an alternative, two cards can be used. The site attendant will remove the
card in the datalogger and replace it with a fresh one. This method is fast and
does not require a PC at the site. The disadvantage of this approach is that it
will create non-uniform file sizes around the time that the cards are exchanged.
If the time series data is processed by EdiRe, nonuniform file sizes are not an
issue. The files could be spliced together using the Append option in
CardConvert.
17
OPEC Open-Path Eddy-Covariance System
4.1.4 File Management with CardConvert
CardConvert is a utility that is shipped with LoggerNet and is used to convert
the files on a card from the compact file format TOB3 to a TOB1, TOA5, or
Array Compatible CSV (see FIGURE 4-7). This utility is installed on the PC
under the Start menu in the Programs | Campbell Scientific | LoggerNet |
Utilities group. To take advantage of all the functionality described in this
section, use CardConvert version 1.2 or later.
Copy the data from the PC/CF card onto the computer using Windows
Explorer. Place the file in a temporary working directory, e.g., C:\Temporary.
It will take about 15 minutes to copy a 1024 Mbyte file (about 28 days of
10 Hz time series from a Basic Open-Path Eddy-Covariance system).
TIP
It is good practice to archive the raw data as copied from the
PC/CF card. Use a zip utility to compress the file, before
archiving it to a CD. Compressing a 1024 Mbyte time series file
will take about 2 minutes using the built-in Windows XP zip
utility.
TOB1
Binary File
TOB3
File from
PC/CF Card
FIGURE 4-7. File format flow
18
TOA5
ASCII file
with header
Array
Compatible
ASCI
Arrays without
header
OPEC Open-Path Eddy-Covariance System
Start the CardConvert utility. When CardConvert is started, a summary of its
current settings is displayed in the lower right hand screen. There are four
screens/parameters to setup and configure in CardConvert. Once configured,
CardConvert will save these settings in the Windows Registry. In subsequent
runs, only the default Output File name must be changed.
• Click on the “Select Card Drive” button and navigate to the temporary
directory that contains the raw time series and flux data files. Select
the drive and click on the “OK” button.
TIP
Since it is likely that the online flux and time series files will be
broken down into different sizes based on time, process the time
series and flux data with multiple passes through CardConvert.
• Click on the “Change Output Dir” button and navigate to the folder to
store the processed data. Select the drive and click on the “OK”
button.
• Click on the “Destination File Options” button, the Destination File
Option screen will appear. Configure the screen as shown in FIGURE
4-8 and then click on the “OK” button.
TIP
FIGURE 4-8. Destination File Option screen
To create an EdiRe raw format file list using the Interpreter, save
the data in the TOB1 file format. If another format is needed at a
later date, convert the TOB1 files into the other file format (see
FIGURE 4-7).
19
OPEC Open-Path Eddy-Covariance System
• Set the file size (Bale size) in the “Time Settings” screen. For the
time series files, use a one hour file size and for the online flux data,
use a one day file size.
TIP
NOTE
Practice converting the data from the card and processing it with
your off-line tools. Only through this type of experimentation
will you determine the file sizes that work best for you.
• Finally, change the output file name. The default output file name is
TOB1_stationname.ts_data.dat, where stationname is the datalogger’s
station name and can be set using DevConfig.exe (part of LoggerNet).
The factory default station name is the datalogger serial number.
Right click on the stationname.ts_data file and select “Change Output
File”.
If any other parameters are later changed, CardConvert will
return to the default output file name.
• Click on the “Start Conversion” to start the conversion process
(FIGURE 4-9).
FIGURE 4-9. Fully configured CardConvert start up screen
The converted files will look something like the 20 Hz files shown in FIGURE
4-10. Each file name will consist of a user specified file name and the time
stamp of the first record in each file with a one minute resolution. Note that
the first file and last file in the folder are smaller than the rest. This is because
they do not contain a complete hour of data.
20
OPEC Open-Path Eddy-Covariance System
FIGURE 4-10. List of files created by CardConvert
4.1.4.1 Collecting Data with One Card
If a single card is used to collect the data and the data on the card is not deleted
after copying it to the PC, the datalogger will append the data collected since
the card was removed to the data file on the card. This means that the next
time data is collected and converted from the PC/CF card, the last file in
FIGURE 4-10 will contain a complete hour of data.
Before collecting data for the second and subsequent time, rename the last
partial file from …_0700.dat to …_0700_1.dat. This will ensure that the
complete version of this file is saved to the folder.
CardConvert will not overwrite existing files within a folder. Instead, it will
append a “_n” to the filename, where n is a numeral. FIGURE 4-11 shows the
name change to duplicate files. Note that the duplicate files get mixed in with
the original files within the folder. Rename the duplicate file using a group file
renaming utility. Specifically, find and replace the text “_1.dat” with “.bak”.
Sort the files in the folder by extension and delete the “bak” files (see FIGURE
4-12).
21
OPEC Open-Path Eddy-Covariance System
FIGURE 4-11. List of files created by CardConvert with duplicates
FIGURE 4-12. List of files where the duplicate files are renamed to
4.1.4.2 Collecting Data with Two Cards
Data can be collected using two cards. The card in the datalogger is removed
and replaced with an empty one. The data from the original card is copied to
the computer at some later time.
22
*.bak
OPEC Open-Path Eddy-Covariance System
NOTE
Using two or more cards is a fast and efficient method of
collecting data. Only new data is collected; however, two partial
files within a “baling” period are created. Variable file lengths
are not an issue with EdiRe.
Again, note that the first and last files (FIGURE 4-13) are smaller than the rest
of the files. The next time data is retrieved from the datalogger, the file
Ts_data_2000_02_25_0700.dat will be continued in another file (FIGURE
4-14).
FIGURE 4-13. List of files collected the first time using two cards
FIGURE 4-14. List of files collected the second time using two cards
23
OPEC Open-Path Eddy-Covariance System
4.2 Data Processing
There are several ways to process the raw time series data collected by the
system. Some of these are considered standard by the community and others
are subject to much debate. To gain a better understanding of the issues,
review the literature.
4.2.1 Online Processing
The Open-Path Eddy-Covariance datalogger programs perform the processing
listed below online and in real time. In many cases, no further processing of
the data is required.
• Thirty minute statistics for the turbulence and meteorological
variables, these include fluxes, means, and standard deviations. The
output period can be changed by changing the Constant
OUTPUT_PERIOD.
• Compute the mean moist air density, temperature, and vapor pressure.
Air density is used to compute the sensible heat and momentum flux.
• Webb et al. (1980) term for latent heat flux and CO
separated into its two components, e.g., one term for temperature and
the other for vapor density.
• Wind direction and wind speed from the sonic anemometer.
• Compute the KH20 Krypton Hygrometer oxygen correction.
• Cross-products (second moments) required for an off-line coordinate
rotation following Kaimal and Finnigan (1994) and Tanner and
Thurtell (1969).
NOTE
The datalogger programs do not apply a coordinate rotation to
the data.
4.2.2 Off-line Processing with EdiRe
EdiRe is a powerful and flexible software package for processing EddyCovariance data collected by a Campbell Scientific system. It is available at no
charge from the University of Edinburgh’s ftp site (http://www.geos.ed.ac.uk/
abs/research/micromet/EdiRe/Downloads.html). EdiRe can, using the
Interpreter, read TOB1 data files. EdiRe will sort the input data files by date
and time, and can deal with files that are not the same size, e.g., collected using
two cards.
flux. This term is
2
Some of the available EdiRe processing functions are:
24
• Statistics for the turbulence and meteorological variables, these
include means, standard deviations, and covariances.
• Filtering the data (detrending, recursive, FIR, exponential).
OPEC Open-Path Eddy-Covariance System
• Spatial separation correction or maximum covariance.
• Coordinate rotation (mean vertical wind = 0 or planar fit).
• Spectral analysis.
• Wavelet analysis.
NOTE
• Webb et al. (1980) term for latent heat flux and CO
flux. This term is
2
separated into its components due to temperature and to vapor density
fluctuations.
• Wind direction and wind speed from the sonic anemometer.
• KH20 Krypton Hygrometer oxygen correction.
There is a certain level of complexity inherent in any software package with
EdiRe’s flexibility. At the University of Edinburgh ftp site you will find an
online user forum (http://www.geos.ed.ac.uk/abs/research/micromet/
EdiRe/EdiReForum/), FAQ (http://www.geos.ed.ac.uk/abs/research/
micromet/EdiRe/Help_information.html), and tutorials (http://www.geos.ed.
ac.uk/abs/research/micromet/EdiRe/Tutorials/).
Along with the information on the University of Edinburgh’s ftp site, see the
EdiRe online help and the Campbell Scientific Application Note on using
EdiRe. EdiRe lends itself to experimentation. The best way to learn how to
use EdiRe is to experiment with it.
EdiRe is not a Campbell Scientific product. Campbell Scientific
provides limited support of EdiRe, specifically in reading in
TOB1 data files. If further support is required, contact the
support personnel listed at http://www.geos.ed.ac.uk/
abs/research/micromet/edisol/edicontact.htm.
4.2.2.1 Creating Raw File Format and Processing Lists
Start EdiRe, select the Processing | Interpreter menu. Fill out the Interpreter
screen as shown in FIGURE 4-15. In the “Sample file” pull down menu,
navigate to the folder that contains the raw time series data and select one of
the TOB1 files (FIGURE 4-16). Do the same for the “Format list file name”
and “Processing List file name”. After the Interpreter screen is filled out, it
should look something like FIGURE 4-17. Now click on the “Create” button.
EdiRe will estimate the data collection frequency and display this estimate in a
pop-up window. The estimate will be wrong because EdiRe includes the
header information as data records in the file. Enter the correct data collection
frequency (FIGURE 4-18). Now, click on “OK” and then “OK” again. EdiRe
will create and load the format and processing lists.
25
OPEC Open-Path Eddy-Covariance System
FIGURE 4-15. Interpreter settings to read a Campbell Scientific TOB1
data file
FIGURE 4-16. Folder that contains the raw TOB1 time series data files
26
OPEC Open-Path Eddy-Covariance System
FIGURE 4-17. Completed Interpreter screen
FIGURE 4-18. Estimated sample frequency and correct sample
frequency
4.2.2.2 Example EdiRe Raw File Format and Processing Lists
To view the Lists in EdiRe, select the Processing | Options menu. The “Raw
File Format” list. The default “Processing Steps” list must be modified for
your specific application.
27
OPEC Open-Path Eddy-Covariance System
The Processing Steps list has the default instructions needed to extract the data
from the TOB1 file (FIGURE 4-19). Additional instructions can be found in
the “Processing Items” list. To configure the instruction, select the instruction
parameter and enter values either from the “Processing Item Parameter” pull
down menu, or enter output variable names in the “Processing Item Parameter”
field using the keyboard.
The location of the output files, as comma separated values, can be specified in
the “Output Files” tab or as part of the processing list with the “Location
Output Files” instruction (FIGURE 4-20).
TIP
Campbell Scientific recommends incorporating the output file
destination as part of the processing list rather than changing the
output file name in the “Output File” tab. This will prevent the
accidental merger of results from multiple stations.
The mean and standard deviation of a variable is found using the “1 Chn
Statistics” instruction, covariances and fluxes are found using the “2 Chn
Statistics” instruction. The results of the processing can be plotted with the
“Plot Value” instruction, see FIGURE 4-21, FIGURE 4-22, and FIGURE 4-23.
Other useful instructions are: Rotation Coefficients, Rotation, Remove Lag,
Cross Correlate, Wind Direction, Raw Subset, Set Value, Comments, User
Defined, User Defined Fast, and Webb Correction. For additional information
on EdiRe and its use, see the online help and the EdiRe application note
published by Campbell Scientific
FIGURE 4-19. Default EdiRe processing list created by the Interpreter
28
OPEC Open-Path Eddy-Covariance System
FIGURE 4-20. Output file location as part of the processing list
FIGURE 4-21. Output file location as part of the processing list
29
OPEC Open-Path Eddy-Covariance System
FIGURE 4-22. Processing U
and CO2 with 1 Chn Statistics instruction
z
FIGURE 4-23. Computing CO
30
graphing U
flux with 2 Chn Statistics and
2
statistic with Plot Value instruction
z
OPEC Open-Path Eddy-Covariance System
5. Basic Eddy-Covariance Theory
Appendix B lists several references on eddy-covariance flux measurements.
The following text books are particular informative:
Atmospheric Boundary Layer Flows, Their Structure and Measurement
(Kaimal and Finnigan, 1994)
A Brief Practical Guide to Eddy Covariance Flux Measurements:
Principles and Workflow Examples for Scientific and Industrial
Applications (Burba and Anderson, 2010)
Eddy Covariance, A Practical Guide to Measurement and Data Analysis
(Abuinet et al., 2012)
Handbook of Micrometeorology. A Guide for Surface Flux Measurements
and Analysis (Lee et al., 2004)
Workshop on Micrometeorology (Haugen, 1973)
FIGURE 5-1. Ideal vertical profiles of virtual potential temperature and
specific humidity depicting all the layers of the atmospheric
boundary layer
31
OPEC Open-Path Eddy-Covariance System
′
+
The surface layer (FIGURE 5-1) is comprised of approximately the lower 10%
of the atmospheric boundary layer (ABL). The fluxes of water vapor and heat
within this layer are nearly constant with height when the following criteria are
met: the surface has approximate horizontal homogeneity; and the relationship
z/h << 1 << z/z
height of the ABL, and z
above conditions are met, the flux of carbon dioxide, water vapor and heat,
within the surface layer, may be written as:
UF′′=
ρ
(1)
zcc
ULLE′′=
ρ
ρ
is true, where z
om
(2)
zvv
UTCH′′=
(3)
zpa
is the roughness length of momentum. When the
om
is the height of the surface layer, h is the
sfc
where F
carbon dioxide density from the mean,
vertical wind speed from the mean, LE is the latent heat flux, L
heat of vaporization,
density from the mean, H is the sensible heat flux,
the heat capacity of air at a constant pressure, T
is the carbon dioxide flux,
c
ρ′
is the instantaneous deviation of the water vapor
v
ρ′
is the instantaneous deviation of the
c
U
is the instantaneous deviation of
z
is the latent
v
ρ
is the density of air, Cp is
a
′
is the instantaneous deviation
of air temperature from the mean (Stull, 1988; Massman et al., 2004).
The quantities
′′
U
ρ
zc
′′
,
U
ρ
, and,
zv
′′
UT
are the covariances between vertical
z
wind speed and carbon dioxide density; vertical wind speed and vapor density;
and vertical wind speed and temperature. These quantities are computed online
by the CRBasic datalogger as are the cross products (second order moments)
required to apply a post processing coordinate rotation following Kaimal and
Finnigan (1994) and Tanner and Thurtell (1969).
The Open-Path Eddy-Covariance system directly measures latent and sensible
heat flux. If net radiation and soil heat flux are also measured, energy balance
closure may be examined using the surface energy balance equation:
LEHGR
=− (4)
n
where R
defined as positive away from the surface and R
F
is the net radiation and G is the total soil heat flux. H, LE, and G are
n
is positive toward the surface;
is defined, following the micrometeorological sign convention, as positive
c
n
away from the surface.
If a FW05 finewire thermocouple is not part of the system, the sensible heat
flux (H
temperature.
32
) can be computed from the vertical wind and humidity corrected sonic
c
Appendix A. CSAT3 Orientation
A.1 Determining True North and Sensor Orientation
The orientation of the CSAT3 negative x-axis is found by reading a magnetic
compass and applying the site-specific correction for magnetic declination;
where the magnetic declination is the number of degrees between True North
and Magnetic North. Magnetic declination for a specific site can be obtained
from a USGS map, local airport, or through a NOAA web calculator (Section
A.2). A general map showing magnetic declination for the Conterminous
United States in 2004 is shown in FIGURE A-1.
FIGURE A-1. Magnetic declination for the conterminous United States
(2004)
A-1
Appendix A. CSAT3 Orientation
Declination angles are always subtracted from the compass reading to find
True North. A declination angle East of True North is reported as positive a
value and is subtracted from 360 (0) degrees to find True North as shown
FIGURE A-2. A declination angle West of True North is reported as a negative
value and is also subtracted from 0 (360) degrees to find True North as shown
in FIGURE A-3. Note that when a negative number is subtracted from a
positive number, the resulting arithmetic operation is addition.
For example, the declination for Longmont, CO (10 June 2006) is 9.67
True North is 360
declination for Mc Henry, IL (10 June 2006) is -2.68
° - (-2.68°), or 2.68° as read on a compass.
0
° - 9.67°, or 350.33° as read on a compass. Likewise, the
°, and True North is
FIGURE A-2. A declination angle East of True North (positive) is
subtracted from 360 (0) degrees to find True North
°, thus
FIGURE A-3. A declination angle West of True North (negative) is
A-2
subtracted from 0 (360) degrees to find True North
Appendix A. CSAT3 Orientation
A.2 Online Magnetic Declination Calculator
The magnetic declination calculator web calculator published by NOAA’s
Geophysical Data Center is available at the following url
http://www.ngdc.noaa.gov/seg/geomag/jsp/struts/calcDeclination. After the
web page loads, enter the site zip code, or longitude and latitude, then click on
the “Compute Declination” button (FIGURE A-4).
FIGURE A-4. Online magnetic declination calculator with inputs and
output for Longmont, CO
The declination for Longmont, CO is 9.67 degrees (10 June 2006). As shown
in FIGURE A-4, the declination for Colorado is positive (east of north), so true
north for this site is 360 - 9.67, or 350.33 degrees. The annual change is -8
minutes/year or 8 minutes west per year.
A-3
Appendix A. CSAT3 Orientation
A-4
Appendix B. References
Abuinet, M., Vesala, T., and Papale, D. (eds), 2012: Eddy Covariance, A
Practical Guide to Measurement and Data Analysis, Springer, New York,
438 pp.
Brutsaert, W.: 1982, Evaporation into the Atmosphere, D. Reidel Publishing
Co., Dordrecht, Holland, 300.
Buck, A. L.: 1976, “The Variable-Path Lyman-Alpha Hygrometer and its
Operating Characteristics”, Bull. Amer. Meteorol. Soc., 57, 1113-1118.
Burba, G. G., and Anderson, D. J., 2010: A Brief Practical Guide to Eddy
Covariance Flux Measurements: Principles and Workflow Examples for
Scientific and Industrial Applications, LI-COR Biosciences, Lincoln, 211
pp.
Campbell, G. S., and Tanner, B. D.: 1985, “A Krypton Hygrometer for
Measurement of Atmospheric Water Vapor Concentration” Moisture and
Humidity, ISA, Research Triangle Park, North Carolina.
Dyer, A. J. and Pruitt, W. O.: 1962, “Eddy Flux Measurements Over a Small
Irrigated Area”, J. Applied Meteorol., 1, 471-473.
Fleagle, R. G. and J. A. Businger, 1980: An Introduction to Atmospheric
Physics. Academic Press, 432 pp.
Gash, J. H. C.: 1986, “A Note on Estimating the Effect of a Limited Fetch On
Micrometeorological Evaporation Measurements”, Boundary-Layer
Meteorol., 35, 409-413.
Haugen, D. A. (ed), 1973: Workshop on Micrometeorology, American
Meteorological Society, Boston, 392 pp.
Kaimal, J. C. and J. A. Businger, 1963: A continuous wave sonic anemometer-
thermometer, J. Appl. Meteorol., 2, 156-164.
Kaimal, J. C. and Finnigan, J. J.: 1994, Atmospheric Boundary Layer Flows,
Their Structure and Measurement, Oxford University Press, New York,
289 pages.
Kaimal, J. C. and J. E. Gaynor, 1991: Another look at sonic thermometry,
Bound.-Layer Meteorol., 56, 401-410.
Mortensen, N. G., 1994: Flow-response characteristic of the Kaijo Denki
omni-direction sonic anemometer (TR-61V), Riso-R-704(EN), Riso
National Laboratory, Roskilde, Denmark.
Lee, X., Massman, W., and Law, B. E. (eds.), 2004: Handbook of
Micrometeorology. A Guide for Surface Flux Measurements and Analysis,
Kluwer Academic Publishers, Boston, 250 pp.
B-1
Appendix B. References
Schotanus, P., F. T. M. Nieuwstadt, and H. A. R. de Bruin, 1983: Temperature
measurement with a sonic anemometer and its application to heat and
moisture fluxes, Bound.-Layer Meteorol., 26, 81-93.
Schuepp, P. H., Leclerc, M. Y., MacPherson, J. I., and Desjardins, R. L.: 1990,
“Footprint Prediction of Scalar Fluxes from Analytical Solutions of the
Diffusion Equation”, Bound.-Layer Meteorol., 50, 355-373.
Shuttleworth, W. J.: 1992, “Evaporation” (Chapter 4), in Maidment (ed),
Handbook of Hydrology, Mc Graw-Hill, New York, 4.1-4.53.
Stull, R. B.: 1988, An Introduction to Boundary Layer Meteorology, Kluwer
Academic Publishers, Boston.
Tanner, C. B. and Thurtell, G. W.: 1969, Anemoclinometer measurements of
Reynolds stress and heat transport in the atmospheric surface layer, Final
Report, United States Army Electronics Command, Atmospheric Sciences
Laboratory, Fort Huachuca, Arizona.
Tanner, B. D.: 1988, “Use Requirements for Bowen Ratio and Eddy
Correlation Determination of Evapotranspiration”, Proceedings of the 1988
Specialty Conference of the Irrigation and Drainage Division, ASCE,
Lincoln, Nebraska, 19-21 July 1988.
Wallace, J. M. and P. V. Hobbs, 1977: Atmospheric Science: An Introductory
Survey. Academic Press, 350 pp.
Warland, J. S., Dias, G. M., and Thurtell, G. W.: 2001, “A Tunable Diode
Laser System for Ammonia Flux Measurements over Multiple Plots”
Environ. Pollut., 114, 215-221.
Webb, E.K., Pearman, G. I., and Leuning, R.: 1980, “Correction of Flux
Measurement for Density Effects due to Heat and Water Vapor Transfer”,
Quart. J. Roy. Meteor. Soc., 106, 85-100.
B-2
Appendix C. OPEC200 Open Path Eddy
Covariance System Quickstart Guide
1. Setup tripod and mount enclosures.
Datalogger Enclosure
EC100 Electronics
C-1
Appendix C. OPEC200 Open Path Eddy Covariance System Quickstart Guide
2. Ground tripod and enclosures.
Grounding Rod
Ground Lug
C-2
Appendix C. OPEC200 Open Path Eddy Covariance System Quickstart Guide
Crossarm-to-Pole
Bracket
3. Setup tripod or mast with CM20X Crossarm.
CM20X Crossarm
C-3
Appendix C. OPEC200 Open Path Eddy Covariance System Quickstart Guide
4. Mount the IRGASON into the prevailing wind.
IRGASON
IRGASON
Mounting Bracket
Leveling Bubble
IRGASON Boom Adapter
CM20X Crossarm
C-4
CM20X
Crossarm
Appendix C. OPEC200 Open Path Eddy Covariance System Quickstart Guide
5. Mount the radiation shield and IRGASON Temperature Probe.
Radiation
Shield
IRGASON
Temperature
Probe
C-5
Appendix C. OPEC200 Open Path Eddy Covariance System Quickstart Guide
6. Connect gas analyzer, sonic anemometer, and temperature probe
cables to EC100 Electronics.
Gas
Analyzer
Cable
Temperature
Probe Cable
Sonic
Anemometer
Cable
C-6
Appendix C. OPEC200 Open Path Eddy Covariance System Quickstart Guide
7. Connect system wiring and insert compact flash (CF) card.
Insert CF Card
Datalogger
Enclosure
EC100 Power and SDM Cables
Power Cable
SDM Cable
Power Cable (can connect here or
directly to +12Vdc power supply)
SDM Cable
If datalogger has rechargeable
base, connect cable from +17
to 28 Vdc or +18 Vac power
Otherwise connect
cable from +12Vdc
power supply here
supply here
C-7
Appendix C. OPEC200 Open Path Eddy Covariance System Quickstart Guide
p
8. Turn on the power supply and use the datalogger keyboard display to
set settings. (Follow the steps from left to right, top to bottom).
Press <Esc> to activate the display.
Press <Enter> to display th
e System
Control Menu.
Campbell
Scientific
CR3000 Datalogger
01/01/2012, 14:00:08
CPU: OPEC_basic_vx_x
Running.
.4
.cr3
Enter in the azimuth (the angle in degrees between
true north and the direction the IRGASON points),
and press <ENTER>.
Select “Sonic Azmth”.
System Control
Sonic Azmth :0.00000
Change Press Source >
On Site Zero & S
an
Press <Esc> to exit.
Modify Value
Sonic Azmth
Current Value:
0.00000
New Value:
150
NUM
C-8
Campbell
Scientific
CR3000 Datalogger
01/01/2012, 14:00:08.4
CPU: OPEC_basic_vx_x.cr3
Campbell Scientific Companies
Campbell Scientific, Inc. (CSI)
815 West 1800 North
Logan, Utah 84321
UNITED STATES
www.campbellsci.com • info@campbellsci.com
Campbell Scientific Africa Pty. Ltd. (CSAf)
PO Box 2450
Somerset West 7129
SOUTH AFRICA
www.csafrica.co.za • cleroux@csafrica.co.za
Campbell Scientific Australia Pty. Ltd. (CSA)
PO Box 8108
Garbutt Post Shop QLD 4814
AUSTRALIA
www.campbellsci.com.au • info@campbellsci.com.au
Campbell Scientific do Brasil Ltda. (CSB)
Rua Apinagés, nbr. 2018 ─ Perdizes
CEP: 01258-00 ─ São Paulo ─ SP
BRASIL
www.campbellsci.com.br • vendas@campbellsci.com.br
Campbell Scientific Canada Corp. (CSC)
11564 - 149th Street NW
Edmonton, Alberta T5M 1W7
CANADA
www.campbellsci.ca • dataloggers@campbellsci.ca
Campbell Scientific Centro Caribe S.A. (CSCC)
300 N Cementerio, Edificio Breller
Santo Domingo, Heredia 40305
COSTA RICA
www.campbellsci.cc • info@campbellsci.cc
Campbell Scientific Ltd. (CSL)
Campbell Park
80 Hathern Road
Shepshed, Loughborough LE12 9GX
UNITED KINGDOM
www.campbellsci.co.uk • sales@campbellsci.co.uk
Campbell Scientific Ltd. (France)
3 Avenue de la Division Leclerc
92160 ANTONY
FRANCE
www.campbellsci.fr • info@campbellsci.fr
Campbell Scientific Spain, S. L.
Avda. Pompeu Fabra 7-9, local 1
08024 Barcelona
SPAIN
www.campbellsci.es • info@campbellsci.es
Please visit www.campbellsci.com to obtain contact information for your local US or international representative.
Loading...