Leaf cooling curves: measuring leaf temperature in sunlight

 

Protocol

 

Author

Andrea Leigh

OVERVIEW

This protocol outlines a method for measuring leaf temperature in sunlight by generating a cooling curve for an initially sunlit leaf that is suddenly shaded. The time constant for cooling is calculated and extrapolated backwards in time to determine the initial temperature of the sunlit leaf. This procedure enables a determination of the extent to which reflected radiation affects the accuracy of thermography readings under field conditions.

BACKGROUND

The advent of infrared technology has greatly increased the ability of plant ecologists to accurately measure the temperature of leaves in the natural environment. However, applying thermography potentially is hindered by reflected solar radiation leading to over-estimation of leaf temperature. The amount of radiation reflected into a camera sensor can vary, with the greatest effect occurring when the angle between the camera and the target is the same as that between the target and the sun (Vollmer et al. 2004). Most long wave infrared cameras are bolometric and thus measure total energy falling on the detector without distinguishing between reflected radiation and radiation in thermal equilibrium with the leaf.

Here, we outline a simple method for measuring the temperature of sunlit leaves using well-established portable infrared technology. We generate a cooling curve for an initially sunlit leaf that is suddenly shaded, calculate the time constant for cooling and extrapolate backwards in time to determine the initial temperature of the sunlit leaf. This procedure enables us to ascertain a) the extent to which reflected radiation affects the accuracy of readings and b) whether a rapidly shaded leaf can provide us with a good estimate of the leaf’s original sunlit temperature.

For further background information, see Leigh et al. 2006.

The following protocol is for determining the temperature of leaves on a calm, sunny day, where forced convection is minimal.

MATERIALS/EQUIPMENT

  • ThermaCAM SC2000 infrared camera with an uncooled microbolometer detector and built-in 24° lens (Flir Systems AB, USA)
  • Heat lamp (Ceramic 150 W Pandorel bulb and 30 cm diameter aluminium reflector; Vaucluse and ASP, Adelaide, Australia)
  • Cardboard to provide the shading effect
  • ThermaCAM Researcher 2000 software on a PC computer

UNITS, TERMS, DEFINITIONS

Ecophysiology

Infrared thermography

Bolometric

Specular reflection

PROCEDURE

Camera

To obtain infrared images of chosen leaves (while still attached to plant) under both field and laboratory conditions, use a ThermaCAM SC2000 infrared camera with an uncooled microbolometer detector and built-in 24° lens (Flir Systems AB, USA) mounted on a tripod.

Field conditions

  1. When taking images, aim to avoid wind speeds exceeding 3 km/hr, and ensure all leaves are fully sunlit (although the actual angle of the sun relative to the leaf surface and the camera lens may vary, depending on the time of day and the angle of the leaf on the plant).
  2. In the camera controls, set leaf emissivity to 0.95 (Jones 1999; Jones et al. 2002), and enter ambient temperature, relative humidity, and the distance between the leaf and the lens according to ambient conditions prior to each series of measurements.
  3. Set the camera lens perpendicular to the main plane of the leaf’s surface at a distance of 0.5 – 1 m from the leaf when using the standard 24° built-in lens and 0.1 – 0.2 m when using the close-up lens.
  4. Take an image series of 60 to 90 frames taken approximately one second apart. In each time series, image a single leaf, with the first three frames taken of the leaf in full sunlight. Immediately after the third frame is taken, shade the leaf. In the field, this can be done manually using a sheet of cardboard and all subsequent images are made as the leaf cools.
  5. Download images to a PC computer with ThermaCAM Researcher 2000 software installed.
  6. Using the software, draw an area of standard size on the same position on the leaf of every image.
  7. Calculate the average temperature of the pixels within this area.

Laboratory conditions

The objective of the laboratory experiment is to determine the degree to which reflected radiation can pose a problem during thermography.

  1. Air movement in lab should be reduced to a minimum, i.e. turn off fans etc that might create convective currents.
  2. Heat leaves with a heat lamp (Ceramic 150 W Pandorel bulb and 30 cm diameter aluminium reflector; Vaucluse and ASP, Adelaide, Australia) directed at an angle similar to that of the camera lens to raise the leaf temperature to approximately 10∘C above the ambient temperature.
  3. In the camera controls, set leaf emissivity to 0.95 (Jones 1999; Jones et al. 2002), and enter ambient temperature, relative humidity, and the distance between the leaf and the lens according to ambient conditions prior to each series of measurements.
  4. Set the camera lens perpendicular to the main plane of the leaf’s surface at a distance of 0.5 – 1 m from the leaf when using the standard 24° built-in lens and 0.1 – 0.2 m when using the close-up lens.
  5. Take an image series of 60 to 90 frames taken approximately one second apart. In each time series, image a single leaf, with the first three frames taken of the leaf in full sunlight. Immediately after the third frame is taken, shade the leaf. In the laboratory, instead of shading the leaf, remove the heat lamp immediately after the third image is taken. Because the points subsequent to the initially lamp-lit ones are of images made when the lamp is turned off (the -shaded’ leaf), the shape of the cooling curve is unaffected by the quality of the light to which the leaf was subject during the heating phase.
  6. Download images to a PC computer with ThermaCAM Researcher 2000 software installed.
  7. Using the software, draw an area of standard size on the same position on the leaf of every image.
  8. Calculate the average temperature of the pixels within this area.

Calculations

  1. Assuming the cooling of a shaded leaf is accurately described by a single time constant, , the change in temperature of the leaf, dT, over a time interval, dt, is:
    dT = – (T – Ta) dt/ (Eqn 2.1)
  2. Here, T is the leaf temperature at time t, Ta is the temperature at which leaf cooling asymptotes. Integrating the equation above and applying the boundary condition, T(t = 0) = T0, where T0 is the initial sunlit temperature of the leaf, gives:
    T (t) = (T0 – Ta) e-t/ + Ta (Eqn 2.2)
  3. Taking logarithms of both sides of equation 1 and rearranging the terms gives:
    ln ((T – Ta) / (T0 – Ta)) = y = -t/ (Eqn 2.3)
  4. The negative inverse of the slope of the line in a plot of the left hand side of equation 1 (y) versus time provides . Note that the error in y can be determined by taking differentials of the left hand side of equation 3:
    dy = dT / ( T – Ta) (Eqn 2.4)
  5. For a fixed error in the temperature measurement, the error in y diverges as T approaches Ta. In our analysis, we fit equation 4 to data recorded within the first two cooling time constants ( x 2) or roughly the first 30 seconds for the leaves we have studied. Here, the error in y is low and an accurate value of the cooling time constant can be determined.
  6. The original sunlit temperature can now be determined by extrapolating backwards in time. If the temperature of the first shaded image is T1 measured at time t1 (the time interval between shading the leaf and the first after shading) equation 2 can be extrapolated to T0, the temperature at t = 0:
    T0 = Ta + (T1 – Ta)e t1/ (Eqn 2.5)

See Figure 2.1 (from Leigh et al. (2006) Functional Plant Biology 33, 515-519.) for an example of cooling curves of Proteaceae leaves.

NOTES AND TROUBLESHOOTING TIPS

  • The absolute accuracy of a typical infrared camera can be as low as 2∘C (Flir Systems AB, USA), although if adjusted before every image is more likely to be around 0.5∘C.
  • In Proteaceae, Leigh et al. (2006) found:
    • The time constant for leaf cooling, averaged 17 seconds
    • Reflectance can produce a measurement error of > 2.0∘C
    • imaging a leaf within one second of shading provides a comparatively good estimate of the sunlit leaf temperature
    • A higher level of accuracy of absolute temperature measurement could only be obtained with a more accurate camera.

FLIR infrared cameras, thermal imaging and software

LITERATURE REFERENCES

NOTE: this protocol is reproduced/adapted from Leigh A, Close JD, Ball MC, Siebke K, Nicotra AB (2006) Leaf cooling curves: measuring leaf temperature in sunlight. Functional Plant Biology 33, 515-519. DOI: 10.1071/FP05300

Other references:

Jones HG (1999) Use of thermography for quantitative studies of spatial and temporal variation of stomatal conductance over leaf surfaces. Plant Cell and Environment 22, 1043-1055.

Jones HG, Stoll M, Santos T, de Sousa C, Chaves MM, Grant OM (2002) Use of infrared thermography for monitoring stomatal closure in the field: application to grapevine. Journal of Experimental Botany 53, 2249-2260.

Vollmer M, Henke S, Karstädt D, Möllmann K-P, Pinno F (2004) Identification and suppression of thermal reflections in infrared thermal imaging. In ‘InfraMation 2004’. (Infrared Training Centre)

SEARCH TERMS AND CLASSIFICATION

Species on which this protocol has been used: Family – Australian Proteaceae spp (various) incl. Eucalyptus pauciflora Sieber ex Spreng.

Setting used: Lab and field (hot sunny conditions, i.e. average temperature 35 ∘C ∓ 5 ∘C SD and ~2000 μmol m-2 s-1 PAR)

Leave a Reply