Decagon Devices SRS User manual

SRS
Spectral Reflectance Sensor
Operator’s Manual
Decagon Devices, Inc.
Version: January 15, 2014 — 12:16:45

SRS Sensors
Decagon Devices, Inc.
2365 NE Hopkins Court
Pullman WA 99163
Phone: [phone]
Fax: [phone]
Website: www.decagon.com
Email: supp[email protected] or [email protected]
Trademarks
c
2007-2013 Decagon Devices, Inc.
All Rights Reserved
ii

SRS Sensors CONTENTS
Contents
1 Introduction 1
1.1 Customer Support . . . . . . . . . . . . . . . . . . . . 1
1.2 About This Manual . . . . . . . . . . . . . . . . . . . 2
1.3 Warranty ......................... 2
1.4 Seller’s Liability . . . . . . . . . . . . . . . . . . . . . . 2
2 About SRS 3
2.1 Overview ......................... 3
2.2 Specifications ....................... 4
3 Theory 6
3.1 Normalized Difference Vegetation Index (NDVI) . . . 6
3.2 Fractional Interception of Photosynthetically Active
Radiation ......................... 8
3.3 Canopy Phenology . . . . . . . . . . . . . . . . . . . . 10
3.4 Photochemical Reflectance Index (PRI) . . . . . . . . 11
3.5 Sun-Sensor-Surface Geometry Considerations . . . . . 12
3.6 Calculating Percent Reflectance from Paired Up and
Down Looking Sensors . . . . . . . . . . . . . . . . . . 14
4 Connecting the SRS 19
4.1 Connecting to Decagon Data Logger . . . . . . . . . . 19
4.2 3.5 mm Stereo Plug Wiring . . . . . . . . . . . . . . . 20
4.3 Connecting to a Non-Decagon Logger . . . . . . . . . 20
4.4 Pigtail End Wiring . . . . . . . . . . . . . . . . . . . . 21
5 Communication 23
5.1 SDI-12 Communication . . . . . . . . . . . . . . . . . 23
6 Understanding Data Outputs 25
6.1 Using Decagon’s Em50 series data loggers . . . . . . . 25
6.1.1 Up Looking Sensor Outputs . . . . . . . . . . . 25
6.1.2 Down Looking Sensor Outputs . . . . . . . . . 25
6.2 Using other data loggers . . . . . . . . . . . . . . . . . 26
7 Installing the SRS 27
7.1 Attaching and Leveling . . . . . . . . . . . . . . . . . 27
7.2 Cleaning and Maintenance . . . . . . . . . . . . . . . . 27
iii

CONTENTS SRS Sensors
8 Troubleshooting 28
8.1 DataLogger........................ 28
8.2 Sensors .......................... 28
8.3 Calibration ........................ 28
9 Declaration of Conformity 29
iv

SRS Sensors 1 INTRODUCTION
1 Introduction
Thank you for choosing Decagon’s Spectral Reflectance Sensor (SRS).
We designed the SRS for continuous monitoring of Normalized Differ-
ence Vegetation Index (NDVI) and/or the Photochemical Reflectance
Index (PRI) of plant canopies. We intend it to be low cost, easily
and quickly deployable, and capable of reliable operation over years.
Deploy the sensors over plant canopies to record first appearance of
green canopy, canopy closure, canopy senescence, light use efficiency,
and other variables. Customers can use these measurements to de-
termine light capture, water use, phenology and biomass production.
This manual will help you understand the sensor features and how
to use this device successfully.
1.1 Customer Support
If you ever need assistance with your sensor, have any questions or
feedback, there are several ways to contact us. Decagon has Cus-
tomer Service Representatives available to speak with you Monday
through Friday, between 7am and 5pm Pacific time.
Note: If you purchased your sensor through a distributor, please con-
tact them for assistance.
Email:
supp[email protected] or [email protected]
Phone:
509-332-5600
Fax:
509-332-5158
If contacting us by email or fax, please include as part of your mes-
sage your instrument serial number, your name, address, phone, fax
number, and a description of your problem or question.
1

1 INTRODUCTION SRS Sensors
1.2 About This Manual
Please read these instructions before operating your sensor to ensure
that it performs to its full potential.
1.3 Warranty
The sensor has a 30-day satisfaction guarantee and a one-year war-
ranty on parts and labor. Your warranty is automatically validated
upon receipt of the instrument.
1.4 Seller’s Liability
Seller warrants new equipment of its own manufacture against de-
fective workmanship and materials for a period of one year from the
date of receipt of equipment.
Note: We do not consider the results of ordinary wear and tear,
neglect, misuse, or accident as defects.
The Seller’s liability for defective parts shall in no event exceed the
furnishing of replacement parts “freight on board” the factory where
originally manufactured. Material and equipment covered hereby
which is not manufactured by Seller shall be covered only by the
warranty of its manufacturer. Seller shall not be liable to Buyer for
loss, damage or injuries to persons (including death), or to property
or things of whatsoever kind (including, but not without limitation,
loss of anticipated profits), occasioned by or arising out of the instal-
lation, operation, use, misuse, nonuse, repair, or replacement of said
material and equipment, or out of the use of any method or process
for which the same may be employed. The use of this equipment con-
stitutes Buyer’s acceptance of the terms set forth in this warranty.
There are no understandings, representations, or warranties of any
kind, express, implied, statutory or otherwise (including, but with-
out limitation, the implied warranties of merchantability and fitness
for a particular purpose), not expressly set forth herein.
2

SRS Sensors 2 ABOUT SRS
2 About SRS
2.1 Overview
The SRS are two-band radiometers that measure either incident or
reflected radiation in wavelengths appropriate for calculating the
Normalized Difference Vegetation Index (NDVI) or the Photochemi-
cal Reflectance Index (PRI). Sensors are manufactured in four differ-
ent versions: NDVI-hemispherical, NDVI-field stop, PRI-hemispherical
and PRI-field stop. The hemispherical versions (Figure 1) are built
with Teflon diffusers for making cosine-corrected measurements, and
are primarily designed for up looking measurements of incident radi-
ation. The field stop versions (Figure 2) have a field of view restricted
to 20◦and are designed for pointing downward to measure canopy
reflected radiation in NDVI and PRI wavelengths.
The reflected radiation from a vegetated surface is highly variable,
depending on the amount and type of vegetation cover. This vari-
ability requires a relatively large number of sensors to properly char-
acterize this surface. The field stop and hemispherical versions can
both be used to quantify canopy reflected radiation. The correct
choice of sensor will depend on the objectives of the study. The
hemispherical sensor will do a better job of averaging reflected radia-
tion over a broad area, but if it is not installed normal to the canopy
surface it will also average sky, leading to measurement error. The
field stop sensor can be aimed at a particular spot or have a particu-
lar orientation giving the user more control over what portion of the
canopy is being measured. When using the field stop sensor in an
off-nadir orientation the user should be careful that the sensor is not
pointed above the horizon.
Calculating NDVI or PRI requires knowing both the incoming and
reflected radiation. Unlike the reflected radiation, the incoming radi-
ation is spatially uniform above the canopy. So, you only need one up
facing radiometer to compute the vegetation indices for many down
facing radiometers. The up looking radiometer should be leveled and
have a hemispherical field of view.
3

2 ABOUT SRS SRS Sensors
The SRS is a digital sensor. Its outputs follow the SDI-12 stan-
dard. The SRS is best suited for use with Decagon’s Em50, Em50R,
and Em50G data loggers. However, customers can use the SRS with
other loggers, such as those from Campbell Scientific.
Figure 1: Hemispherical Version Figure 2: Field Stop Version
2.2 Specifications
Accuracy: 10% or better for spectral irradiance and radiance values
Measurement Time: <300 ms
NDVI Wavebands: 630 and 800 nm central wavelengths, with 50
and 40 nm full width half maximum band widths
PRI Wavebands: 531 and 571 nm central wavelengths, with 10 nm
full width half maximum band widths
Dimensions: 43 x 40 x 27 mm
Power Requirements: 3.6 to 15 V DC, 4 mA (reading, 300 ms) 30
µA (quiescent)
Operating Temperature: −40 to 50 ◦C
Connector Types: 3.5 mm (stereo) plug or stripped & tinned lead
wires (Pigtail)
Cable Length: 5 m standard; custom cable length available upon
request.
4

SRS Sensors 2 ABOUT SRS
Other Features:
•SDI-12 digital sensor, compatible with Decagon’s EM50
family and CSI loggers
•In-sensor storage of calibration values
•Four versions
NDVI-hemispherical
NDVI-field stop
PRI-hemispherical
PRI-field stop
•NDVI or PRI sensors with Teflon cosine correcting heads
•NDVI or PRI sensors with 20 degree field stops sealed
with clear acrylic
•NIST traceable calibration to known spectral radiance or
irradiance values
•Sensors can be mounted facing up or down, singly or in
tandem, leveled or aimed
•Fully sealed from the elements and UV resistant to mini-
mize drift over time
5

3 THEORY SRS Sensors
3 Theory
Decagon designed the SRS instruments to measure the NDVI and
PRI vegetation indices from plant canopies. We caution users that
NDVI and PRI are measurements of electromagnetic radiation re-
flected from canopy surfaces, and therefore provide indirect or cor-
relative associations with several canopy variables of interest and
should not be treated as direct measurements of these variables.
NDVI has a well-established and long history of use in remote sens-
ing research and ecological applications related to canopy structure.
PRI, while showing great promise for quantifying canopy physiolog-
ical function, is far more experimental with new uses and caveats
continually being uncovered. While NDVI and PRI can be powerful
tools for inferring structure and function of plant canopies, you must
take into account their limitations when interpreting the data. Sec-
tion 3 provides an overview of the theory and discusses some of the
uses and limitations of each vegetation index.
3.1 Normalized Difference Vegetation Index (NDVI)
A number of nondestructive methods exist for remotely monitoring
and quantifying certain canopy characteristics. Some of those char-
acteristics are: foliar biochemistry and pigment content, leaf area
index (LAI, Nguy-Robinson et al., 2012), phenology, and canopy
photosynthesis (Ryu et al., 2010). One of the most common nonde-
structive techniques involves measuring the NDVI. The NDVI is one
of a large number of vegetation indices. The principle derives from
a well known concept that vegetation reflects light differently in the
visible spectrum (400 to 700 nm) compared to the near infrared (>
700 nm).
Green leaves absorb light most strongly in the visible spectrum,
but are highly reflective in the near infrared region (Figure 3). Be-
cause bare soil, detritus, stems, trunks, branches, and other non-
photosynthetic elements show relatively little difference in reflectance
between the visible and near infrared, measuring the difference be-
tween reflectance in these two bands can be related to the amount
6

SRS Sensors 3 THEORY
of photosynthetic vegetation in the field of view of a radiometer.
See Royo and Dolors (2011) for an extensive introduction to using
spectral indices for plant canopy measurements.
Figure 3: Reflectance spectra for bare soil (Soil) and a healthy
wheat crop at various stages of development: heading (H), anthesis
(A), milk-grain stage (M), and post maturity (PM). Consider two
things about this figure: First, the considerable difference between
reflectance spectra from the soil and all stages of plant
development. Second, the changes in the visible spectra as the
canopy matures and senesces. We reproduced this figure with
permission from Royo and Dolors (2011).
Calculate NDVI with equation 1.
NDV I =ρNIR −ρred
ρNIR +ρred
(1)
Where, ρred and ρNIR are percent reflectances in the red and near
infrared (NIR). We assume percent reflectance to be the ratio of re-
flected to incident radiation in the specified waveband. A detailed
description of how to calculate reflectances from measured radiation
values is provided in equation number 4. NDVI has been shown to
7

3 THEORY SRS Sensors
correlate well with green LAI, although the relationship is specific
for each crop or natural canopy. For example, Aparicio et al. (2002)
studied NDVI versus LAI in more than twenty different durum wheat
genotypes in seven experiments over two years and found the rela-
tionship shown in Figure 4. Nguy-Robinson (2012) also studied the
behavior of NDVI versus LAI in maize and soybean. Their data sug-
gest a similar relationship between the two crops, but not identical.
These relationships have been developed for a wide range of crop
and natural canopies and we encourage our customers to seek out
the best relationship for their application.
Figure 4: Relationship between leaf area index and NDVI for 20-25
durum wheat genotypes studied over two years in seven different
experiments by Aparicio et al. (2002). Values shown were taken at
anthesis and milk-grain stage. Used with permission from author.
3.2 Fractional Interception of Photosynthetically Ac-
tive Radiation
The use of NDVI for determination of leaf area index has limita-
tions. Like many nondestructive techniques (e.g., fisheye and cep-
tometer techniques), the measurement of NDVI becomes less and less
sensitive as LAI increases above a certain point (Figure 3). Nguy-
8

SRS Sensors 3 THEORY
Robinson et al. (2012) suggest changes in LAI are difficult to detect
when LAI is much greater than 3 m2m−2. This should not be surpris-
ing considering the spectral measurement being made. NDVI mea-
surements rely on reflected light from leaf surfaces. As the canopy
fills and upper leaves begin to cover lower leaves, the leaf area will
continue to increase without making a further contribution to re-
flected radiation. Furthermore, foliar chlorophyll is a very efficient
absorber of radiation in red wavelengths so that reflectance from
leaves is typically very low in the red region (Figure 3). Therefore,
increasing LAI, and thus canopy chlorophyll content does not sub-
stantially change red reflectance beyond a certain point. Thus, NDVI
has limited predictive ability in canopies with high LAI. For some
applications NDVI saturation at high LAI may not be as important
as it would appear.
Although the technique may give poor estimates of LAI at high LAI,
shaded leaves tend to have much less impact on resource capture
compared to sunlit leaves, and therefore contribute proportionally
less to canopy productivity. As a general modeling parameter, an
estimate of sunlit leaves may be adequate for estimating photosyn-
thesis and biomass accumulation (i.e., carbon uptake) for some ap-
plications.Monteith (1977) proposed the now well-known relationship
between biomass accumulation and radiation capture seen in equa-
tion 2.
An, canopy =fsSt(2)
In equation 2, An,canopy is the biomass accumulation or carbon assim-
ilation and is a conversion efficiency often referred to as light use ef-
ficiency (LUE). The LUE depends on a variety of factors such as pho-
tosynthetic acclimation, physiological stress level, and plant species.
fsis the fractional interception of radiation by the canopy, and St
is the total incident radiation. The relationship between NDVI and
LAI in Nguy-Robinson et al. (2012) and the relation between frac-
tional interception and LAI (Campbell and Norman, 1998) show that
NDVI and fractional interception are approximately related linearly
(Figure 5). NDVI can provide a good estimate of the fractional in-
terception by green leaves in a canopy; a value that is critical for
carbon assimilation models.
9

3 THEORY SRS Sensors
Figure 5: Relationship between fractional canopy interception and
NDVI, where NDVI is converted to LAI using Nguy-Robinson et al.
(2012). Campbell and Norman (1998) give the relationship between
LAI and fractional interception.
3.3 Canopy Phenology
Like all spectral measurements, NDVI is an indirect measurement.
Over the years, researchers have correlated parameters of interest,
like LAI and fs, to measurements made at 630 nm and 800 nm.
Researchers have estimated other variables using NDVI besides these
relationships. Two of these variables are the focus of Ryu et al.
(2010), who used a simple two-band LED-based sensor, similar to the
SRS-NDVI, to measure canopy phenology and associated changes in
photosynthesis in an annual grassland over a four year period. Ryu
et al (2010). show an exponential relationship between NDVI and
canopy photosynthesis, but found that the LAI of grassland never
increases above 2.5 m2m−2. Ecosystem phenology can also be tracked
in the time series data from their NDVI sensor with errors on the
order of a few days. It should be noted that they filtered their data
by limiting NDVI measurements to a particular sun elevation angle
(e.g., sampling under identical sun zenith and azimuth angles from
day to day).
10

SRS Sensors 3 THEORY
3.4 Photochemical Reflectance Index (PRI)
As described above, the NDVI is primarily useful as a proxy for
canopy structural variables. Although structural properties are crit-
ical, sometimes it is useful to have information about canopy func-
tional properties. For example, estimating the gross primary produc-
tivity (GPP) of ecosystems is critical for modeling the global carbon
balance. The simple model presented in Equation 2 can be used to
predict GPP from three variables: incident light (St), intercepted
light (fs), and light use efficiency (). Stcan generally be estimated
depending on geographic location and time of day or measured with
a PAR sensor or pyranometer. Considering the near linear relation-
ship between NDVI and fractional interception noted above, a simple
two-band spectral reflectance sensor like the SRS-NDVI can provide
an estimate of fs. The light use efficiency term () remains to be
quantified in order to make accurate predictions of GPP.
Gamon et al. (1990, 1992) proposed a dual band vegetation index
(similar to the NDVI) to predict . The foundation of the measure-
ment is based on the absorbance of xanthophyll pigments at 531 nm
that correlates with LUE in many plant species (Gamon et al., 1997).
This ratio is called the Photochemical Reflectance Index (PRI) and
is calculated with Equation 3.
P RI =ρ531 −ρ570
ρ531 +ρ570
(3)
Where, ρ531 and ρ570 are percent reflectances at 531 and 570 nm,
respectively. When combined, you can use NDVI and PRI to predict
biomass accumulation or GPP of an ecosystem without the expense
and work of some of the other approaches (Gamon et al., 2001). Be-
cause of the low cost, light weight, small footprint and low power use
of the sensors, they can be deployed very quickly, over long periods
of time, or in a spatially distributed network to quantify spatiotem-
poral variations in canopy productivity (Garrity et al., 2010).
In addition to LUE, the PRI has also been shown to correlate with
numerous other physiological variables associated with plant photo-
synthetic performance from the leaf to the ecosystem level (Gamon et
al., 1992, 1997, 2001). Xanthophylls absorb radiation at 531 nm and
11

3 THEORY SRS Sensors
will absorb more radiation as a consequence of saturation of chloro-
phyll centers. A normalized difference of reflectance at 570 nm that
remains unchanged despite changes in light saturation to 531 nm
will indicate the level of xanthophyll absorbance and the efficiency
of plant light use. Because increased xanthophyll absorbance is not
solely correlated with LUE, researchers have investigated many other
relationships too.
Numerous studies correlate PRI to various ecophysiological variables
including the epoxidation state of xanthophyll, maximum photo-
chemical efficiency of photosystem II, effective quantum yield, maxi-
mum photosynthesis rate, electron transport under saturating light,
non-photochemical quenching, and chlorophyll to carotenoid con-
tent ratio (Sims & Gamon, 2002; Garrity et al., 2011; Garbulsky
et al., 2011; Porcar-Castell et al., 2012). Garbulsky et al. (2011)
and Porcar-Castell et al. (2012) provide excellent overviews of what
has been done with PRI including analyses of PRI correlations with
several of these variables at the leaf, canopy, and ecosystem levels.
We encourage our customers to use these references as a starting
resource.
3.5 Sun-Sensor-Surface Geometry Considerations
Spectral reflectance measurements are inherently variable due to ra-
diation source, reflecting surface, and sun-sensor-surface geometry.
Hence, it is not uncommon for a time series of NDVI or PRI to con-
tain high amounts of variability. Consider changes in daily NDVI
measured at four-day intervals in a subalpine meadow. (Figure 6)
Figure 6 shows the control treatment green-up under drought con-
ditions. The well-watered treatment in Figure 7 includes the same
time period but has already undergone initial green-up.
There are three things to notice about these data. First, canopy
green-up is clearly visible in the time series of the control (upper
graph) treatment. Second, the track of daily NDVI is generally con-
cave, which indicates that sampling across a consistent sun angle (e.g.
60 ◦elevation angle (Ryu et al., 2012)) or around solar noon is advis-
able when summarizing an entire day to a single value. If comparing
12

SRS Sensors 3 THEORY
measurements acquired under different sun-sensor-surface configu-
rations, it is necessary to first calculate a bidirectional reflectance
distribution function (BRDF). Once you have empirically derived
a BRDF model from the measurements and canopy-specific param-
eters, you can use it to reduce variations that arise from changes
in sun-sensor-surface geometry across diurnal time series. For addi-
tional details on BRDF normalization of vegetation index time series,
see Hilker et al. (2008).
Sometimes NDVI values will exhibit erratic behavior due to envi-
ronmental conditions (see Day 178 on both graphs). Data filtering
(e.g., visual inspection for short time series or automated despiking
algorithms for longer time series) may be required to remove spurious
data points.
Figure 6: Daily variation in NDVI measurements 1 m above a
sub-alpine meadow
13

3 THEORY SRS Sensors
Figure 7: Another daily variation in NDVI measurements 1 m
above a sub-alpine meadow
3.6 Calculating Percent Reflectance from Paired Up
and Down Looking Sensors
Equation 1 shows that NDVI is the ratio of the difference to the
sum of NIR and red reflectances. Each reflectance is the ratio of up-
welling (down looking sensor) to incident (up looking sensor) radiant
flux in each of the wave bands. Calculating this ratio is only possi-
ble when measurements of downwelling and upwelling radiation are
collected simultaneously under the same ambient conditions. Com-
bining measurements made with sensors located long distances apart
is typically not recommended because atmospheric conditions (e.g.,
cloud cover, aerosols) can be highly variable in space. Reasonable
distances between up looking and down looking sensors will depend
on the typical radiation environment of a given location.
It is also important to arrange paired up looking and down look-
ing sensors to collect data at the same time, which will account for
14

SRS Sensors 3 THEORY
temporal variability in radiation conditions. In cases where multiple
down looking sensors have been deployed within close proximity to
each other, it is only necessary to have one up looking sensor. The
measurements from the single up looking sensor can be combined
with the measurements from each of the down looking sensors to cal-
culate reflectances
In the event that up looking measurements are not available, re-
arrangement of the vegetation index equations allows for a rough
approximation of the measurements. The following derivation is for
NDVI, but similar equations apply to the PRI. If Rnis the reflected
NIR radiation from the canopy, Rris the reflected red radiation, In
is the incident NIR, and Iris the incident red, then
NDV I =Rn/In−Rr/Ir
Rn/In+Rr/Ir
=(Ir/In)Rn−Rr
(Ir/In)Rn+Rr
=αRn−Rr
αRn+Rr
(4)
Where α=Ir/In, equation 4 allows the computation of NDVI from
just the down facing measurements if you know the ratio of red to
NIR spectral irradiance, α. Although not extensively tested, we have
found that this ratio (α= 1.86 for NDVI bands) can be used as a
rough approximation during midday under relatively clear sky condi-
tions. However, we caution that direct measurements of downwelling
radiation is more accurate by accounting for any fluctuations in α
that occur with changes in atmospheric conditions or across large
variations in sun elevation angle.
In the event that you do not want to use the default αvalue or
measurements from an up facing sensor are not available, it is is pos-
sible to use a spectralon panel or similar reflectance standard with a
field stop SRS to measure incident irradiance. To measure incident
irradiance with a down facing sensor, place a reflectance standard
within the field of view of the field stop sensor, making sure that the
reflectance panel is level, uniformly illuminated and that the field of
view of the sensor is fully within the area of the reflectance panel.
Measurements obtained from field stop sensors pointed at the re-
flectance panel must be multiplied by πto convert radiance values
to irradiance values. Irradiance values can then be used in Equation
4 or to calculate αdirectly.
15

3 THEORY SRS Sensors
References
Aparicio, N., Villegas, D., Casadesus, J., Araus, J.L., and Royo,
C., (2000). Spectral vegetation indices as nondestructive tools for
determining durum wheat yield. Agronomy Journal, 92: 83-91.
Aparicio, N.; Villegas, D.; Araus, J.L.; Casadess, J.; Royo, C.,
(2002). Relationship between growth traits and spectral reflectance
indices in durum wheat. Crop Science, 42: 1547-1555.
Campbell, G.S. and Norman, J.M., (1998). An Introduction to En-
vironmental Biophysics. Springer-Verlag. New York.
Gamon, J.A., Field, C.B., Bilger, W., Bjorkman, O., Fredeen, A.L.,
Penuelas, J., (1990). Remote sensing of the xanthophylls cycle and
chlorophyll fluorescence in sunflower leaves and canopies. Oecologia,
85: 1-7.
Gamon, J.A., Peuelas, J., Field, C.B., (1992). A narrow-waveband
spectral index that tracks diurnal changes in photosynthetic effi-
ciency. Remote Sensing of Environment, 41: 35-44.
Gamon, J. A., Serrano, L., Surfus, J. S., (1997). The photochemical
reflectance index: an optical indicator of photosynthetic radiation
use efficiency across species, functional types, and nutrient levels.
Oecologia, 112: 492-501.
Gamon, J. A., Field, C. B., Fredeen, A. L., Thayer, S., (2001). As-
sessing photosynthetic downregulation in sunflower stands with an
optically based model. Photosynthesis Research, 67: 113-125.
Garbulsky, M.F., Peuelas, J., Gamon, J., Inoue, Y., Filella, Y. (2011).
The photochemical reflectance index (PRI) and the remote sensing
of leaf, canopy and ecosystem radiation use efficiencies: A review and
meta-analysis. Remote Sensing of the Environment, 115: 281-297.
Garrity, S.R., Vierling, L.A., Bickford, K., (2010). A simple filtered
photodiode instrument for continuous measurement of narrowband
16