Campbell Scientific Bowen Ratio Instrumentation User Manual

INSTRUCTION MANUAL

Bowen Ratio Instrumentation

Revision: 9/05

Copyright (c) 1987-2005

Campbell Scientific, Inc.

Warranty and Assistance

The BOWEN RATIO INSTRUMENTATION is warranted by CAMPBELL
SCIENTIFIC, INC. to be free from defects in materials and workmanship
under normal use and service for twelve (12) months from date of shipment
unless specified otherwise. Batteries have no warranty. CAMPBELL
SCIENTIFIC, INC.'s obligation under this warranty is limited to repairing or
replacing (at CAMPBELL SCIENTIFIC, INC.'s option) defective products.
The customer shall assume all costs of removing, reinstalling, and shipping
defective products to CAMPBELL SCIENTIFIC, INC. CAMPBELL
SCIENTIFIC, INC. will return such products by surface carrier prepaid. This
warranty shall not apply to any CAMPBELL SCIENTIFIC, INC. products
which have been subjected to modification, misuse, neglect, accidents of
nature, or shipping damage. This warranty is in lieu of all other warranties,
expressed or implied, including warranties of merchantability or fitness for a
particular purpose. CAMPBELL SCIENTIFIC, INC. is not liable for special,
indirect, incidental, or consequential damages.

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 [phone]. 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

CAMPBELL SCIENTIFIC, INC. does not accept collect calls.

Bowen Ratio Table of Contents

PDF viewers note: These page numbers refer to the printed version of this document. Use
the Adobe Acrobat® bookmarks tab for links to specific sections.

1. System Overview......................................................1-1

1.1 Review of Theory................................................................................. 1-1

1.2 System Description............................................................................... 1-3

1.2.1 Water Vapor Measurement......................................................... 1-3

1.2.2 Air Temperature Measurement................................................... 1-4

1.2.3 Net Radiation and Soil Heat Flux............................................... 1-5

1.2.4 Power Supply.............................................................................. 1-5

2. Station Installation....................................................2-1

2.1 Sensor Height and Separation............................................................... 2-1

2.2 Soil Thermocouples and Heat Flux Plates............................................ 2-2

2.3 Wiring................................................................................................... 2-4

2.4 Battery Connections.............................................................................. 2-7

2.5 System Startup...................................................................................... 2-7

2.6 Routine Maintenance............................................................................ 2-7

2.7 Cleaning the DEW 10........................................................................... 2-8

3. Sample CR23X Program...........................................3-1

4. Calculating Fluxes Using SPLIT..............................4-1

4.1 Data Handling ........................................................................................ 4-1

4.2 Calculating Fluxes................................................................................ 4-1

Appendices

A. References...............................................................A-1

B. 023 Bowen Ratio (Pre July 1993)........................... B-1

Tables

1.2-1. Component Power Requirements..................................................... 1-5

2.3-1. CR23X/Sensor Connections for Example Program......................... 2-4

3-1. Sample CR23X Bowen Ratio Program Flow Chart............................ 3-2

3-2. Output From Example Bowen Ratio Program .................................... 3-5

4.2-1. Input Values for Flux Calculations .................................................. 4-3

i

Bowen Ratio Table of Contents

Figures

1.2-1. Vapor Measurement System............................................................. 1-3

1.2-2. Thermocouple Configuration............................................................1-4

2-1. CSI Bowen Ratio System ....................................................................2-2

2.2-1. Placement of Thermocouples and Heat Flux Plates .........................2-3

2.2-2. TCAV Spatial Averaging Thermocouple Probe..............................2-4

2.3. A block diagram for the connections between the datalogger,

the BR relay driver and components, and the external battery................2-6

2.4-1. Terminal Strip Adapters for Connections to Battery........................2-7

2.7-1. DEW 10 Circuit Board.....................................................................2-9

B-1. 023 Bowen Ratio Vapor Measurement System with Three

Flowmeters............................................................................................. B-1

ii

∂

∂

∂

∂

λρε∂

∂

(

(

Section 1. System Overview

1.1 Review of Theory

By analogy with molecular diffusion, the flux-gradient approach to vertical
transport of an entity from or to a surface assumes steady diffusion of the
entity along its mean vertical concentration gradient.

When working within a few meters of the surface, the water vapor and heat
flux densities, E and H, may be expressed as:

ρ

Ek

=

v

(1)

v

z

HCk

Here ρv is vapor density, ρ is air density, Cp is the specific heat of air, T is
temperature, z is vertical height, and k
vapor and heat, respectively. Air density and the specific heat of air should
account for the presence of water vapor, however, use of standard dry air
values usually causes negligible error. The eddy diffusivities are functions of
height. The vapor and temperature gradients reflect temporal and spatial
averages.

Applying the Universal Gas Law to Eq. (1), and using the latent heat of
vaporization, λ, the latent heat flux density, λ, can be written in terms of vapor
pressure (e).

Here P is atmospheric pressure and ε is the ratio of the molecular weight of
water to the molecular weight of dry air.

In practice, finite gradients are measured and an effective eddy diffusivity
assumed over the vertical gradient:

=ρ

L

=

e

T

pH

and kH are the eddy diffusivities for

v

Pez

(2)

z

k

v

(3)

ee

−

k

λρε

ρ

v

P

zz

()

TT

pH

zz

()

L

In general, k
equal. The ratio of H to L
surface into sensible and latent heat flux. This technique was first proposed by
Bowen (1926). The Bowen ratio, β, is obtained from Eq. (4) and Eq. (5).

and kH are not known but under specific conditions are assumed

v

=

e

HCk

=

is then used to partition the available energy at the

e

)

12

−

12

−

)

12
12

. (5)

−

(4)

1-1

Section 1. System Overview

(

λ

−−−

=

−

where

βλε==

PC

ε is the psychrometric constant.

p

PC

H

L

e

−

TT

p

ee

()

)

12
12

(6)

−

The surface energy budget is given by,

where R

is net radiation for the surface and G is the total soil heat flux. The

n

sign convention used is R

RGHL

ne

positive into the surface and G, H, and Le positive

n

away from the surface. Substituting βL

e

0, (7)

for H in Eq. (7) and solving for Le

yields:

Measurements of R

=

e

, G, and T and e at two heights are then required to

n

. (8)

+1 β

RG

n

L

estimate sensible and latent heat flux.

Atmospheric pressure is also necessary, but seldom varies by more than a few
percent. It may be calculated for the site elevation, assuming a standard
atmosphere, or obtained from a nearby station and corrected for any elevation
difference (Wallace and Hobbes, 1977).

Eq. (6) shows that the sensitivity of β is directly related to the measured
gradients; a 1% error in a measurement results in a 1% error in β.

When the Bowen ratio approaches -1, the calculated fluxes approach infinity.
Fortunately, this situation usually occurs only at night when there is little
available energy, R

0.75), L

and H are assumed to be negligible and are not calculated. Ohmuna

e

- G. In practice, when β is close to -1 (e.g., -1.25 < β < -

n

(1982) describes an objective method for rejecting erroneous Bowen ratio data.

1-2

Section 1. System Overview

FIGURE 1.2-1. Vapor Measurement System

1.2 System Description

1.2.1 Water Vapor Measurement

It is common practice in Bowen ratio measurements to measure wet bulb
depression to develop the water vapor gradient. The position of the two
psychrometers is periodically reversed to cancel systematic sensor errors
(Suomi, 1957; Fuchs and Tanner, 1970).

In the Campbell Scientific system, vapor concentration is measured with a
single cooled mirror dew point hygrometer
multiple level gradient studies (Lemon, 1960). Air samples from two heights
are routed to the cooled mirror after passing through mixing volumes (Figure

1.2-1). The problems associated with wick wetting and water supply in
psychrometers are avoided and systematic sensor errors are eliminated.

Air is drawn from both heights continuously through inverted 25 mm filter
holders fitted with Teflon filters with a 1 µm pore size. The filter prevents dust
contamination in the lines and on the cooled mirror. It also prevents liquid
water from entering the system.

1

, using a technique developed for

A single low power DC pump aspirates the system. Manually adjustable
rotometers are used to adjust and match the flow rates. A flow rate of 0.4
liters/minute with 2 liter mixing chambers gives a 5 minute time constant.

1

Model Dew-10, General Eastern Corp. Watertown, MA

1-3

Section 1. System Overview

A datalogger is used to measure all sensors and control the valve that switches
the air stream through the cooled mirror.

The resolution of the dewpoint temperature measurement is ±0.003°C over a
±35°C range. The limitation is the stability of the Dew-10, approx imately

0.05°C, yielding better than ±0.01 kPa vapor pressure resolution over most of

the environmental range.

Every 2 minutes the air drawn through the cooled mirror is switched from one
height to the other with the valve. Forty seconds is allowed for the mirror to
stabilize on the new dewpoint temperature and 1 minute and 20 seconds worth
of measurements for an individual level are obtained for each 2 minutes cycle.

The dewpoint temperature is measured every second and the vapor pressure is
calculated by the datalogger using the equation described by Lowe (1976).
The average vapor pressure at each height is calculated every 20 minutes.

CR23X

FIGURE 1.2-2. Thermocouple Configuration

1.2.2 Air Temperature Measurement

Air temperature is measured at two heights with chromel–constantan
thermocouples wired as in Figure 1.2-2. The differential voltage is due to the
difference in temperature between T
error. The datalogger resolution is 0.006°C with 0.1 µV rms noise.

The thermocouples are not aspirated. Attempts to aspirate the TCs with the air
from the vapor measurement system were not successful. Testing under 1000

-2

W m

solar radiation, with several radiation shield designs and aspiration rates
of up to 80 cm s
to radiation from the shield/ducting.

Calculations indicate that a 25 µm (0.001 in) diameter TC experiences less
than 0.2°C and 0.1°C heating at 0.1 m s
under 1000 W m

-1

(1 l min-1), showed a significant increase in temperature due

-2

solar radiation (Tanner, 1979). More importantly, error in

and T2 and has no inherent sensor offset

1

-1

and 1 m s-1 wind speeds, respectively,

1-4

the gradient measurement is due only to the difference in the radiative heating
of the two TC junctions and their physical symmetry minimizes this.
Conversely, contamination of only one junction can cause larger errors.

Applying temperature gradients to the TC connectors was found to cause
offsets. The connector mounts were designed with radiation shields and
thermal conductors to minimize gradients.

The prototype systems used two sets of TCs on each system, one 25 µm and
one 76 µm diameter. It was hypothesized that the 25 µm diameter would
suffer less from radiation loading and the 76 µm would be less prone to
breakage. The current design uses a single set of TCs (76 µm standard) with
two parallel junctions at each height as a back up against breakage.

1.2.3 Net Radiation and Soil Heat Flux

Net radiation and soil heat flux are averaged over the same time period as the
vapor pressure and temperature differences.

To measure soil heat flux, heat flux plates are buried in the soil at a fixed depth
of between 5 to 10 cm to reduce errors due to vapor transport of heat.
Typically the plates are buried at a depth of 8 cm. The average temperature of
the soil layer above the plate is measured using 4 parallel thermocouples. The
heat flux at the surface is then calculated by adding the heat flux measured by
the plate to the energy stored in the soil layer. The storage term is calculated
by multiplying the change in soil temperature over the averaging perio d by the
soil heat capacity.

Section 1. System Overview

1.2.4 Power Supply

The current requirements of the components of the Bowen ratio system are
given in Table 1.2-1.

Component Current at 12 VDC

Cooled Mirror 150 - 500 mA
Pump 60 mA
CR23X 5 mA

A 20 watt solar panel (SP20R) and a 70 amp-hour battery are capable of
providing a continuous current of 300 - 350 mA. The solar panel is necessary
if the system is to be used for periods longer than 2-3 days. The datalogger
can control power to the cooled mirror and pump, and can shut down the
system if the battery voltage is low or if measurements are not needed at night.

TABLE 1.2-1. Component Power

Requirements

1-5

Section 1. System Overview

This is a blank page.

1-6

Section 2. Station Installation

Figure 2-1 shows the typical Bowen ratio installation on the CM10 tripod. The 023A
enclosure, mounting arms, and SP20R solar panel all mount to the tripod mast (1 1/4 in.
pipe, inside diameter) with U-bolts. The size of the tripod allows the heights of the arms to
be adjusted from 0.5 to 3 meters. The mounting arms should be oriented due south to
avoid partial shading of the thermocouples.

The net radiometer is mounted on a separate stake (not provided by Campbell Scientific) so
that the tripod is not a significant portion of its field of view. It should be positioned so
that it is never shaded by the tripod or mounting arms and should be mounted so that it
points south.

2.1 Sensor Height and Separation

There are several factors which must be balanced against each other when
determining the height at which to mount the support arms for the temperature
and air intakes. The differences in temperature and moisture increase with
height, so the resolution on the measurements of the temperature and vapor
gradient will improve the farther apart the arms are.

The upper mounting arm must be low enough that it is not sampling air that is
coming from a different environment upwind. The air that the sensors see
must be representative of the soil/vegetation that is being measured. As a rule
of thumb, the surface being measured should extend a distance upwind that is
at least 100 times the height of the sensors. The following references discuss
fetch requirements in detail: Brutsaert (1982); Dyer and Pruitt (1962); Gash
(1986); Schuepp et al. (1990); and Shuttleworth (1992).

The lower mounting arm needs to be higher than the surrounding vegetation so
that the air it is sampling is representative of the bulk crop surface, and not a
smaller scale effect that might be seen in a row crop if the sensors were down
between rows.

2-1