Hello everyone, welcome to the seminar. My name is Mark Blonquist from Apogee
Instruments in Logan, Utah. I'm going to speak about plant canopy water status
estimation from canopy temperature measurements. Before getting that far, I'm
going to spend a fair amount of time talking about actually measuring surface
temperature. This slide provides a brief outline of where the presentations going.
First I'll talk about infrared radiometers, their operation
and calibration. Then surface temperature measurements, specifically
correcting for surface emissivity and field-of-view. I'll also deal with the
challenge of partial canopy cover, the two methods I'll talk about to handle
that are angling radiometers and then a model to account for soil temperature.
Finally we'll get to estimating canopy water status, we'll spend some
time with the simpler method the crop water stress index and then a more
involved method canopy stomatal conductance. The basic
components of an infrared radiometer are the radiation detector itself, an
internal temperature sensor to provide a temperature measurement of the detector,
and then a housing to hold it all together.
The signal generated by the radiation detector is proportional to the
difference between the radiation being absorbed by the detector and also the
radiation being emitted by the detector. All surfaces or objects that have a
temperature above absolute zero emit infrared radiation. The detector in the
radiometer is absorbing the radiation that's emitted towards it, at the same
time the detector is emitting radiation. The signal will be dependent on the
the difference between these two radiation streams. When the absorbed
radiation at the detector is less than the emitted radiation, the signal is
negative. When the absorb radiation at the detector is greater than the emitted
radiation, the signal is positive. The sensor is actually measuring or
detecting radiation but temperature is the quantity that we're
interested in, and of course they're related through the stefan-boltzmann law
where energy is proportional to a constant times the fourth power of
temperature. Another important component, an essential component, of the radiation
detector itself is a filter to block all wavelengths outside of what we call the
atmospheric window. The plot you're looking at here shows wavelengths on the
x-axis and atmospheric transmittance on the y-axis. The wavelengths between 8 and
14 micron or micrometers is defined as the atmospheric window because at these
wavelengths the transmission of the atmosphere is very near 1, meaning the
atmosphere is transparent like a window. The filter on the radiation detector
needs to closely correspond to the atmospheric window to eliminate
influence from the atmosphere, essentially see through the atmosphere
to the surface that we're interested in measuring. Below 8 micron you can have
interference from water vapor and above 14 micron we can have interference from
CO2. The final component that's essential for an infrared radiometer is the
calibration. The photo that you're viewing now is a picture of the
calibration system and ad Apogee Instruments. There's two main components.
The one that I've circled here is the actual cap piece where all the
radiometers are held during calibration procedure. The cap piece is mounted on
top of this piece here, which is the actual blackbody cone or radiation source.
What happens during calibration is the radiometers are mounted in the cap piece
over top of the cone. You can see some insulation around the cone there. Wiith
the two independent pieces, the cap and the cone piece, we can maintain there
temperatures independent of each other and then we can collect a whole data set
across the whole range of temperatures. During the calibration procedure we
also collect the millivolt signals from the radiometers. This produces a large data
set that we can then use to derive custom coefficients for all radiometers
that allow us to convert the signal, which again from the previous slide was
proportional to the radiation difference, we can convert that to a temperature.
Once we have a calibrated radiometer already deployed in the field and start
making measurements. Surface temperature measurements are pretty straightforward
when the surface is a blackbody. Blackbody is defined as any object or
material that emits the theoretical maximum amount of radiation based on its
temperature or we define it as a object or material with an emissivity equal to
1, where emissivity is just the fraction of blackbody emission. Here I
have a plant canopy with emissivity equal to 1. In that scenario the target
temperature returned by the radiometer, would equal the true surface temperature
but in practice natural surfaces are not black bodies. Most plant canopies have
emissivites equal to 0.98 0.99. What that means is, that plant canopy will be
emitting 98 or 99 percent of the theoretical maximum based on temperature,
and there'll be a small fraction of background radiation that's directed
towards the radiometer because it's reflected from the surface. Surface
reflectance is equal to 1 minus the emissivity or in our case that would be
1 minus 0.98. Essentially what this means is if we deploy a radiometer to measure
a plant canopy in an outdoor environment the background is the sky.
The sky is emitting infrared radiation just like the plant canopy and where the
plant canopy is not a perfect black body with an emissivity equal to one a
smaller fraction, one minus the emissivity of the background radiation
in this case coming from the sky, is reflected from the plant canopy and
directed towards the radiometer. Typically the sky temperature and the
surface temperature are much different so we have to account for this reflected
fraction in order to get accurate measurement of the surface temperature.
In order to do so, we need an estimate or measurement of the surface
emissivity, an estimate or measurement of the radiation coming from the sky, in
addition to our measurement of the surface temperature. I''m not going to
go through the derivation of the equation to correct for emissivity but
I'm showing it here. We take our measurement of target temperature and we
input it along with our measurement or estimate of emissivity and our
measurement our estimate of the background temperature, then we can
calculate the actual or true surface temperature. To give you an idea of
the magnitude of effective emissivity, I have a couple simple examples one for a
clear day and one for a cloudy day. Let's assume that our sensor is measuring a
temperature of 25 Celsius, a pretty standard background temperature or sky
temperature for a clear day would be negative 40 Celsius, again our canopy
emissivity is 0.98 and when we plug all those numbers into the equation
above we get a surface temperature almost a full degree warmer than the
temperature we measure with the radiometer. On a cloudy day, the
background temperature is much warmer it's a lot closer to the actual measured
surface temperature and when we plug the numbers into our equation above for a
cloudy day we find that the surface temperature is much closer to the
measured temperature. Here we have a table listing emissivities for several
different surfaces one would encounter in natural settings plant, soils water, and
so forth. You can see that most natural surfaces are near black bodies with
emissivity is in the 0.90 range. we can find some things that are very, very
low emissivity like polished aluminum for example. I also have at the bottom of
the slide some typical values for the background temperature or what would be sky
temperatures in an environmental application. Clear sky is often very cold,
negative 40 to negative 60 C. Overcast sky is often near air temperature. I've
also listed a simple equation that you can use to calculate sky temperature
it's approximated from the air temperature, where we take air
temperature plus 50 times the fraction of clouds minus 60. On a clear day, our
fraction of clouds would be 0 and this equation would simplify to air
temperature minus 60 is equal to the sky temperature. On a completely overcast day,
our fraction of clouds would be 1. This equation would simplify to air
temperature minus 10, would be equal to sky temperature on an overcast day.
This can be used for a simple way to approximate sky temperature to be used with
emissivity correction. Moving on from emissivity another important
consideration is the area of surface that the radiometer actually views. In
other words the field-of-view. I like to draw an analogy whenever I explain
field-of-view of an infrared radiometer and the best analogy that I can think of
is put yourself in front of a flat wall in a dark room with a flashlight. Ff you
hold the flashlight perpendicular to the wall and turn on the
flashlight, you'll see a circle of light. As you move the flashlight closer to the
wall, the circle gets smaller and as you move the flashlight away from the wall
the circle gets larger. If you start to angle that flashlight, so it's no longer
perpendicular to the wall then the circle of light spreads out into an
ellipse. The the same principle holds for a radiometer that you direct towards a
surface. If the radiometer is oriented perpendicular to the surface, then it's
going to view or sense a circle. If the radiometer is close to the surface, it
will be a small circle. If the radiometers moved further from the
surface, it will be a larger circle. If you start angling it away from
perpendicular, it will view an ellipse. The area that being measured or
sensed by the radiometer is dependent on three things: the field-of-view, the
radiometer mounting height, and the radiometer mounting angle. Where the
field-of-view is just defined as the angle or half-angle of the cone, that's formed by
the footprint that the radiometer sees and then the apex of the cone here at
the aperture. Companies should specify what the field-of-view of the radiometer
is. In this example, it's a 22 degree half-angle or a 44 degree full angle
field-of-view. This slide shows multiple different models that are available from Apogee
Instruments and it lists their their fields-of-view. I won't go into detail
about calculating the the field-of-view, but one very helpful tool that runs the
calculations is found online the website here. All you have to plug in this
calculator the three things I mentioned on the previous slide: the field-of-view
specification in terms of the half-angle, the distance of the radiometer from the
target, and then the angle of the radiometer with respect to the target.
The calculator returns the dimensions of the ground area that the
radiometer will be viewing or sensing and also the area of
the footprint. One of the challenges of measuring plant canopy temperature with
an infrared radiometer, is the situation where plant canopy doesn't occupy the
entire field-of-view. Shown here in this example, plant canopy occupies part of
the field-of-view, the radiometer is also seeing some soil. In most situations, the
soil temperature and plant canopy temperature are different so in order to
get an estimate of the canopy temperature we have to account for the
soil temperature somehow or eliminate a soil from the field-of-view. One way to
deal with this is to angle the radiometer, such that we maximize canopy
within the field-of-view. Doing this it is very helpful to rely on the field-of-biew
calculator that I mentioned in the previous slide. It can provide estimates of
the actual area being sensed by the radiometer. Another way to try to
estimate plant canopy temperature when the surface is only partially covered by
the canopy, is to actually measure or estimate the soil, its temperature, and
then approximate the fraction of canopy within the field-of-view of the
radiometer and the fraction of soil within the field-of-view of the
radiometer. Here again I won't go through the derivation of this equation, but we
can use the simple relationship with the measurement of the soil temperature,
estimate of the fraction of surface that's soil, a fraction of surface that's canopy,
and then the actual surface temperature that we measure with the radiometer,
and calculate the canopy temperature.
Once we have a measurement of actual canopy temperature we're ready to
progress to use that canopy temperature to estimate canopy water status, but
before we go through the two methods that I mentioned in the introduction I
want to talk a little bit about the theory behind using canopy temperature
as an indicator of water status. Plant leaves are actually covered with small
microscopic pores called stomata. In order to photosynthesize, plants have to
take up CO2. Stomata open in the presence of light in order to allow CO2 to enter
for photosynthesis. An inevitable trade-off when stomata open is water
loss, plant leaves are saturated with water and the atmosphere is often dry
sometimes very dry, so water evaporates out of this stomata when they're open in
order for CO2 uptake to occur. Plants uptake soil water to replace the water
being lost through the stomata during photosynthetic uptake of CO2 and as
plants draw down water in the soil and soil water becomes limiting, plant water
uptake obviously starts to decline. This tends to close stomata. Stomatal
closure reduces the stomatal aperture, or the degree of opening, and this in turn
reduces transpiration ,or the evaporation of water from the stomata. Evaporation is
a cooling process so as stomata closed in response to a drawdown of soil water
plants aren't as cool as they otherwise would be so the canopy temperature
thereby increases. One of the most important factors we have to remember,
however, is that soil water status is not the only control on canopy temperature.
Canopy temperature by itself is actually a poor indicator of plant water
status because there's multiple factors that control canopy temperature. In
addition to transpirational cooling, which is partially controlled by the soil water
status, air temperature, humidity, radiation, and wind, all the environmental
conditions the plants are subject to will influence the canopy temperature.
In order to use canopy temperature as a means of estimating water status, we
have to account for all of the variables that can influence fluency it. An early
method and a rather simple method for using canopy temperature to quantify
water status is called the crop water stress index abbreviated CWSI. This
method was developed by Idso and colleagues in 1981. At the end of the
presentation, I'll provide the full citation for the paper that Idso and
colleagues published. Essentially they found through their field data that
the difference between canopy and air temperature for a well watered canopy
declined as vapor pressure deficit increased. They also found that this
canopy to air temperature difference was relatively constant for a canopy
experiencing significant water stress. They use these two boundaries, we're
defining as a non-water stressed baseline and a water stress baseline,
sort of reference points to compare actual measurements of canopy and air
temperature. Here's the equation that they developed. The crop water stress
index is the measurement of canopy minus air temperature minus the non-water stress
baseline and divided by the difference between the water stressed and non-water
stressed baseline. The required measurements in order to
apply this empirically based crop water stress index are canopy temperature, air
temperature, and relative humidity. The air temperature and humidity are
required to calculate the vapor pressure deficit. Just to illustrate how the
crop water stress index works, let's say our measured canopy minus air temperature
value was 0 for a vapor pressure deficit of 3. We would then calculate the crop
water stress index by taking the difference between this measured value
and the non-water stress baseline, which we define as "a" and then dividing that by
the difference between the water stress baseline and the non-water stress
baseline which we define as "b". Well crop water stress index is just "a" over "b", or
in our example we take our measured value of canopy minus air temperature of
zero, subtract off the non water stress baseline, then divide the numerator by
the difference between the water stress baseline and non-water stress baseline.
That gives us a crop water stress index of 0.46 or approximately halfway between
the non-water stress and water stress baseline. The crop water stress index is
actually designed to output a value of 0 when the measured value of canopy to air
temperature falls on top of the non-water stress baseline, and it outputs a
value of 1 when the measured value of canopy to air temperature falls on top
of the water stress baseline. The major advantages of using the empirical crop
water stress index are its simplicity, it only requires three measurements,
and the measurement errors are calibrated out. Typically the baselines
are derived from field measurements, which allows for field calibration. The
disadvantages are that empirical data is required to determine the non-water
stress and water stress baselines so we have to have some field collection data
before we start. The alternative to this would be finding crop specific
values of the non0-water stress and water stress baseline in the literature.
Another major disadvantage is the environmental conditions must be similar
from one hour to the next if we're going to compare hourly values of crop water
stress index or from one day to the next if we're going to compare daily values
of the crop water stress index. The reason being we said in the previous
slide that multiple factors beyond the amount of soil water available for
plants to uptake control canopy temperature, wind speed, radiation, air
temperature all of these variables will impact the canopy to air temperature
difference. If we have significantly different wind speed from one day to the
next or a significantly different radiation environment from one day to
the next, this won't be accounted for in our baselines, it makes the empirical
crop water stress index error-prone. To show how well the crop water stress
index works, I have here 11 days of data collected over a cornfield
near North Platte, Nebraska in the middle of July. In the upper graph, we're showing
the canopy to air temperature difference and in the lower graph we're showing the
crop water stress index. The green line is the non-water stress baseline, the red
line is the water stress baseline, and the black line is the measured value of
canopy to air temperature. You see in the first couple days the
crop water stress index is quite low but it tends to increase over the course of
a few days before dropping back down again. Here the response going from a
value near 0.5 or 0.6 back down to a value near zero was caused by 20
millimeters of rainfall. The crop water stress index behaves how we would
expect it increases as soil water is drawn down, when rain falls it responds
by decreasing to near zero. You can see near the end of this data set it's
starting to increase again indicating water stress due to soil water drawdown.
One more interesting point to make that there is within day variability. This may
represent actual increase in the crop water stress index over the course of
the day or it might indicate variable environmental conditions. In addition to
the empirical crop water stress index, Jackson and colleagues in 1981 also
developed a theoretical or energy balance based version of the crop water
stress index. Rather than relying on the empirically derived baselines, they
took all of the equations describing the energy balance for the plant canopy,
shown here in the diagram, and they combined them and rearranged them to solve for
the canopy to air temperature difference. I won't combine the equations and show
the the Tc minus Ta equation, but what that gives you is a means of calculating
the non-water stress and water stress baseline in real time so that conditions
don't have to be the same from hour to hour or day to day as required with the
empirical crop waters index. The drawback to this more
theoretical or energy balance based version is that requires a lot more data
in order to actually get values of the crop water stress index. In addition
to canopy temperature, air temperature, and relative humidity, we also have to
have a measurement or estimate of net radiation, a measurement of wind speed,
and then measurements or estimates of canopy height and leaf area index. I
won't say anything more in this presentation about the theoretical
version of the crop water stress index, but again at the end of the presentation
will provide the reference to the paper in which it was derived. A method similar
to the energy balance based version of the crop water stress index is a direct
calculation of the plant canopy stomatal conductance. We can take the
exact same plant canopy energy balance equations that were used to derive
the theoretical crop water stress index, and we can combine them and rearrange
them to solve for this term, the plant canopy stomatal conductance. Stomatal
conductance is the actual quantification of the degree of stomatal opening or
stomatal closure. This calculation of plant canopy stomatal conductance
has the advantage of being a physiological variable that's directly
related to stomatal aperture. It actually counts for all variables
influencing canopy temperatures so unlike the empirical crop water stress
index, it should work well under all environmental conditions. To
demonstrate the calculation of canopy stomatal conductance, I have the same data
set for corn near North Platte, Nebraska. In the graph, on the top the black line
is the actual canopy stomatal conductance calculated from the equation
on the previous slide and the green line is a value are calling potential
canopy stomatal conductance it's derived from a leaf level model scaled up to
the canopy. The details are found in the Blonquist et. al. paper referenced on
the previous slide. I won't go into the nuts and bolts of the leaf level model, but I
will provide the reference for the paper at the end of this slide show. A
comparison of the actual to potential canopy stomatal conductance gives an
index of water status. You can see here the ratio starts out near one and then
declines as the canopy draws down water from the soil. Just like we
saw with the crop water stress index, it dropped to near zero following a
rainfall event. The ratio of actual to potential canopy conductance increases
after the rainfall event and then after a couple days being near one it starts
to decline again as water stress sets back in. Also like the crop water stress
index, there is some within day variability of the ratio of actual to
potential canopy conductance. This slide just provides an indication of how well
the calculation of actual canopy conductance works. Here we're showing the
values from the equation, compared to potential values derived
from a scaled up leaf level model for all days immediately following rainfall
for the entire summer from corn crop near North Platte, Nebraska. The reason
we're only showing data for days immediately following rain falls, we
would expect the actual conductance to closely match the potential conductance
under well water conditions and indeed that's what they find the data match a
one-to-one line relatively well. There are some outliers where the actual
conductance is significantly higher than the potential conductance and it's
possible these are times when the canopy is wet, unfortunately we didn't have a
wetness sensor to determine when the canopy was wet and when the canopy was
dry. Just to provide some some summary in conclusion, canopies temperatures can
be used as a means to determine plant canopy water status, and tere's multiple
ways to do so. I've demonstrated a simple and a more complex method for using
canopy temperatures to estimate water status. We found that both methods were
sensitive to water stress and rainfall. Maybe the most important conclusion
that I can make is that in order for the methods to work canopy temperature
measurements must be accurate. The correction for surface emissivity is
significant and should be done, and field-of-view must be considered especially
for conditions of partial canopy cover, want to make sure that canopy is being
measured rather than some mixture of canopy and soil. Here are the references
for the papers that I mentioned during the talk, much more detail can be found
regarding the three different methods in each paper. I hope this presentation was
helpful for everyone. I really appreciate the opportunity to be a part of Decagon
seminar series and thank the audience for their participation.
For more infomation >> Remove DFplayer noise, murmur sound. Caption on Please! mini MP3 player for arduino. - Duration: 2:24. 
For more infomation >> Commission calls for emergency declaration on opioid crisis - Duration: 1:27. 
For more infomation >> Nearly 200 job vacancies for Monterey County teaching positions - Duration: 2:49. 
For more infomation >> Bookings for Carnival Cruises Strong for 2018 - Duration: 1:37. 
For more infomation >> Boston police, officials come out for a night of fun - Duration: 1:41. 
For more infomation >> Question 9 - Brett Hudson to the Minister for Economic Development - Duration: 2:21. 
For more infomation >> Traffic stop for tinted windows leads to heroin-related arrest - Duration: 0:14. 
For more infomation >> If You Eat One Banana Daily, Here Is What Will Happen To Your Body - Duration: 3:07. 
For more infomation >> レクサスISがマイナーチェンジ！ 新型は走りとデザインをより深化 - Duration: 5:12. 
For more infomation >> Forecast Focus for July 31 - Duration: 3:31. 
For more infomation >> E.Coli risk ends for Westchester area - Duration: 2:09. 
For more infomation >> Running Back, Linebacker Suddenly An Issue For The Broncos - Duration: 4:18. 
For more infomation >> Worry is the enemy not work | daily sprout 305 - Duration: 0:47. 
