In-vivo fast optical measurements of leaves

Abstract

Comparison of the Walz LSA-2050 handheld leaf analyzer to earlier instrument designs. This is preliminary assessment based mainly on specifications, scientific and technical literature and the author’s user experience. It is based on long-term experience with the use of the SPAD-502 chlorophyll meter from Konica-Minolta, the Dualex 3.1 and Dualex 4 dual-excitation instruments from Force-A and recent use of the MPM-100 multiple wavelength pigment meter from Opti-Sciences and the LSA-2050 from Walz. I also show how to import the data logged by the LSA-2050 into R for further analysis.

1 Introduction

All in vivo measurements based on estimating leaf transmittance when expressed as concentrations are on a unit leaf area basis. Given that many variables in the growing environment affect specific leaf area (SLA) and leaf water content, differences in chlorophyll concentration between treatments depend strongly on the base of expression. In practice, the direction of a response to a treatment can be opposite when concentrations are expressed per unit area and per unit dry-mass or per unit fresh-mass.

Leaves have a heterogeneous internal structure with many gas-liquid and liquid-solid interfaces. These interfaces scatter and reflect light while traveling inside a leaf. This lengthens the path of photons within the leaf enhancing absorption by pigments and altering the relationship between absorbance and concentration compared to the same concentration of pigments in a homogeneous solution. Because of this, the methods described below, when used to estimate concentrations depend on a calibration that is affected by the anatomy, water content and even recent illumination of the leaf.

1.1 Leaf transmittance

Chlorophyll concentration can be estimated based on the transmittance of the leaf at specific wavelengths where Chlorophyll is the main absorber of light (Parry et al. 2014). A leaf differs from a solution in that the chlorophyll distribution is not homogeneous. Not only there are cells, cell walls and vacuoles in cells but chlorophyll is located in the chloroplasts, and the location of chloroplasts within mesophyll cells changes in response to illumination. The changes in transmittance can even be used to study chloroplast accumulation movements (justyna?). Thus, the relationship between chlorophyll concentration and light transmittance varies to some extent depending on the irradiance and colour of recent illumination (Hoel and Solhaug 1998), specially blue light sensed by phototropins results in chloroplast accumulation (xxxx?). This has been described as a source of errors in chlorophyll measurements with the SPAD-502 instrument (xxxx?), but as this is an optical phenomenon in the plant it can be expected to affect all the instruments assessing chlorophyll based on leaf transmittance.

The concentration of chlorophyll together with the lengthened path of photons inside a leaf due to scattering, makes leaves very effective absorbers such that transmittance at the wavelength of maximum absorption by chlorophyll is extremely low, making the measurement of high concentrations of chlorophyll in-vivo with the SPAD-502 error-prone. The Opti-Sciences CCM-200 and the Apogee MC-100 chlorophyll meters use similar wavelengths as the SPAD-502.

The Dualex 4 introduced the approach of using a longer wavelength, were transmittance is higher but still mainly dependent on chlorophyll and little affected by other pigments and metabolites. This decreases measurements on healthy leaves with high concentrations of chlorophyll. The MPM-100 uses an approach and wavelengths similar to the Dualex 4 and the MPM-100/S uses an approach similar to the SPAD-502.

The light scattering within leaves decreases the apparent transmittance even in the absence of absorbing pigments. This effect is estimated by measuring the apparent transmittance of far-red or near-infrared light by the leaf, as at these wavelengths absorption by pigments is minimal. Scattering is almost complete in leaves, and in the far-red region, because of the absence of absorption by pigments, approximately half of the incident light exits the leaf through each epidermis. The different instruments measure the attenuation of radiation in this region as a reference.

For estimating chlorophyll concentration, the instruments described in (tab-leaf-tfr?) use two LEDs as light sources and a photodiode as detector. They are all hand-held devices for non destructive chlorophyll measurements on thin leaves. Some of them have additional functions, described in the next section. Transmittance at each wavelength can be estimated as the ratio of the photodiode signal with the leaf between the jaws and the photodiode signal with the leaf absent.

The instruments described in (tab-leaf-tfr?) estimate the concentration of chlorophyll per unit leaf area based on leaf transmittance. The wavelengths used both as target and reference for assessing chlorophyll concentration vary across instruments. A reference is needed to assess the attenuation caused by light scattering independently of chlorophyll concentration. Differences in the wavelength used as reference as long as there is no absorption by pigments have minimal consequences. In contrast difference in the target wavelength are more significant because a measurement using the wavelength of peak absorption (near 660 nm) even if advantageous when assessing low concentrations makes the measurement of high concentrations difficult. As concentrations of chlorophyll in leaves can be high enough to result in very low transmittance, instruments that use longer wavelengths in the shoulder of the peak instead of at its cusp, are in most cases preferable. The LSA-2050, MPM-100, Dualex 4 and MC-100 return approximate concentrations on an area basis while the SPAD-502, MPM-100/S and some other SPAD clones return “SPAD units”. The CCM-200 returns values of “CCI” index. The relationships between values in SPAD units and chlorophyll concentration and between values in CCI units and chlorophyll concentration are both curvilinear, but convex in opposite directions [@]. The in vivo chlorophyll concentration estimates must be in general considered as approximate. For example, comparison of values obtained from leaves with differences in leaf anatomy, require in most cases separate calibrations for accuracy.

Instruments for the non-destructive measurement of chlorophyll concentration based on leaf transmittance. Target and reference wavelengths, full width half maximum (FWHM) given in parentheses when available. Quantities are approximate concentrations or indexes with arbitrary units. The measured area is given in $\mathrm{mm}^2$ and in the case of the LSA-2050 can be reduced with a supplied mask for use on small leaves. Notes: ^*no user experience by author. ⁺The SPAD-502 and SPAD-502 Plus, are identical except for the data logging available in the Plus version. The MPM-100 is available in customized versions based on LEDs emitting at non-standard wavelengths. The Walz LSA-2050 measures in addition to transmittance chlorophyll fluorescence $Fv/Fm$.
Instrument	Target	Reference	Quantity	Area
LSA-2050	715 (25)	770 (30)	conc.	79/28
MPM-100	720	850	conc.	71
MPM-100/S^*	650	940	SPAD index	71
Dualex 4	720	810?	conc.	20
MC-100^*	650	930	conc.	71
CCM-200^*	650	930	CCI index	71
SPAD-502⁺	650	940	SPAD index	6

The readings are expected not to depend on which side of the leaf is illuminated. In fact, the Dualex 4 and the LSA-2050, measure transmittance in opposite directions through leaves. The Dualex has the LEDs in the upper jaw and the sensor in the lower one, while the LSA-2050 locates them the other way around. Like the Dualex, the SPAD-502, MPM-100, MPM-100/S, CMM-200 and MC-100 have the LED in the upper jaw.

The Walz LSA-2050 is aimed at scientific research, with built-in support for recalibration and corrections for different types of leaves. The Dualex 4 aims at being both useful in research and as a tool for crop management. The MPM-100 seems to target the same users as the Dualex 4. The CMM-200, MC-100 and specially the SPAD-502 are marketed as tools for crop management, although they are in practice also frequently used in plant research.

The indexes are given by (Parry et al. 2014) as:

\[\mathrm{SPAD} = k \times \log \frac{T_{940}}{T_{650}} + C\] where $k$ and $C$ are calibration constants.

\[\mathrm{CCI} = \frac{T_{931}}{T_{653}}\]

Even if the reading from the SPAD is a value rather similar to absorbance, it is not linearly related to concentration of chlorophyll in the measured leaves. Markwell et al. (1995) proposed building a calibration curve to convert SPAD units into chlorophyll concentration by fitting a non-linear function of the form:

\[y = 10^{x^k}\] where $x$ is the reading in SPAD units and $y$ the concentration of chlorophyll per unit leaf area. The value of $k$ depends on the individual SPAD instrument and obviously also on the units in which the chlorophyll concentration is expressed. Markwell et al. (1995) described the value of $k$ as more dependent on the instrument than on the plant species or genotype. However, differences in anatomy can be expected to affect the relationship between transmittance and concentration, independently of the instrument used. The advantage of using this function compared to a polynomial or spline, is that it constrains the shape of the curve, which is important when calibrations are based on limited data.

1.2 Epidermal transmittance

Fluorescence is dependent on the number of absorbed photons, this also applies to chlorophyll. Thus, a measurement of fluorescence behind the epidermis, from the chlorophyll in the mesophyll, can be used to estimate the transmittance of the epidermis if it is possible to estimate the fluorescence yield that can be expected in the absence of the epidermis. One approximation is to use a wavelength that is known not to be absorbed by the epidermis to obtain a reference value of fluorescence. This value can be compared against fluorescence excited by other wavelengths where the epidermis can absorb.

This the principle of the Dualex (dual excitation), similar to the earlier approach used by the UVA-PAM, a modified Walz Mini-PAM. The Dualex, however, measures the fluorescence on the opposite side of the leaf from the excitation, while the UVA-PAM and the Walz LSA-2050 measure the fluorescence from the side of the leaf where the excitation light impings. While the Dualex can be used only on thin leaves, the UVA-PAM and the LSA-2050 can be used on thicker objects like fruits. In the case of the LSA-2050 by removing the lower jaw, which contains the light sources for estimation of chlorophyll based on transmittance across the leaf. The UVA_PAM uses a single light guide, and does not measure transmittance. The MPM-100 senses the chlorophyll fluorescence on the same side of the leaf as the excitation impings. However the head has a maximum opening that limits its use to thin objects like leaves.

The instruments described in (tab-epidermis-tfr?) use LEDs as light sources and photodiodes, in most cases filtered, as detectors. They are hand-held devices for non destructive measurements of thin leaves, and in some cases also suitable for thick leaves, fruits and stems. Walz’s Xe-PAM in the exception, it uses a filtered Xenon lamp as light source and a photo-multiplier tube as detector, is not handheld and usually used in a laboratory.

The instruments in (tab-epidermis-tfr?) rely on the excitation of chlorophyll fluorescence with its intensity used as reporter inside the leaf for estimation of the transmittance of the epidermis. All the Dualex instruments sense the fluorescence on the opposite side of the leaf to the excitation, while other instruments sense it on the same side. These instruments differ in the target and reference wavelengths used for assessing epidermal transmittance. The different transmittances depend on the accumulation of different metabolites: UV-B, phenolic acids, UV-A1, flavonoids, blue, non-photosynthetic-related carotenoids, green, anthocyanins. Either blue- or red-excited fluorescence is used as reference assuming that these wavelengths are minimally absorbed in the epidermis.

Instruments for the non-destructive measurement of epidermal transmittance using chlorophyll fluorescence as reporter. Excitation wavelengths, full width half maximum (FWHM) given in parentheses when available. Instruments names followed by ^* indicate that the author has no direct experience in their use. The MPM-100 is also available in customized versions based on LEDs emitting at non-standard wavelengths. The Walz LSA-2050 measures in addition to transmittance chlorophyll fluorescence Fv/Fm.
Instrument	UV-B	UV-A1	Blue	Green	Red^**
LSA-2050	310 (15)	365 (12)	450 (14)	530 (27)	630 (24)
MPM-100		375		525	660
MPM-100/S^*		375		525	660
Dualex 4		375		520	630
Dualex 3 FLAV		375			630
Dualex 3 ANTH^*				520	630
Dualex 3 CA	310				630
UVA-PAM^*		375	470
UVA-PAM/red^*		375			630?
Xe-PAM^*	314 (18)	366 (32)	475 (140)

2 Leaf reflectance

The instruments described in (tab-leaf-rfr?) measure reflectance of leaves, each at a pair of wavelengths. These reflectance values can be combined into indexes that are correlated to specific plant conditions. These indexes are typically used in remote sensing but instruments for their measurement on individual leaves are also available. NDVI and NDGI mainly assess how green vegetation is while PRI assesses differences within the green region of the spectrum.

Instruments for the non-destructive measurement of indexes based on leaf reflectance. Excitation wavelengths, full width half maximum (FWHM) given in parentheses when available. Instruments names followed by ^* indicate that the author has no direct experience in their use.
Instrument	Target	Reference	Quantity
PlantPen NDVI 310^*	660	770	NDVI
PlantPen PRI 210^*	531	570	PRI
PlantPen/NPen N 110^*	560	780	NDGI

A significant part of light “reflected” from a leaf is scattered light that has traveled inside the leaf and been partly absorbed by pigments, rather than reflected at the leaf surface. Thus, similarly to transmittance, reflectance at specific wavelengths, can inform about pigments within leaves. The PlantPen instruments from PSI, as well as remote sensing, RGB and spectral imaging of vegetation rely on this phenomenon.

All of the instruments described in (tab-leaf-rfr?) return values that are indexes rather than concentrations. The indexes as implemented in these instruments are:

\[\mathrm{PRI} = \frac{R_{531} – R_{570}}{R_{531} + R_{570}}\]

\[\mathrm{NDVI} = \frac{R_{660} – R_{770}}{R_{660} + R_{770}}\]

\[\mathrm{NDGI} = \frac{R_{560} – R_{780}}{R_{560} + R_{780}}\] where $R_\lambda$ is reflectance on a band centred at a given wavelength in nanometres. The half maximum full width (HMFW) of the excitation is not given in the specifications. There are variations on the exact wavelengths and HMFW used for these indexes that are in wide use, and this must be taken into account when comparing measurements.

There are indexes like NBI that combine different types of measurements, and these can be obtained only with the more advanced instruments, Dualex 4 and LSA-2050. The Nitrogen Balance Index (NBI) is obtained by dividing the chlorophyll concentration index ($\approx \mu g\, cm^{-2}$) by the epidermal flavonols index, AFLAV ($\approx A_{375}$) (using inconsistent units), to obtain a value sometimes described as “unitless”.

3 Alternative approaches

With a spectrometer it is possible to measure whole leaf spectral reflectance and transmittance, and from them derive absorptance over arbitrary wavelength ranges within the overlapping sensitivity range of the spectrometers and the wavelength range of the illumination. The measurement of epidermal transmittance with a spectrometer is limited to the few species in which the epidermis can be stripped from leaves and measured by itself.

Ocean Optics used to make a portable spectrometer equipped with small integrating spheres in a clip-like arrangement with one sphere in each jaw. In use one should be aware that two spheres collecting light and reflecting it back onto the leaf can result in erroneous readings as some light can travel back and forth across the leaf. This must be taken into consideration in the measuring protocol used, but it is not a design flaw.

I should mention here also the SpectraVue Leaf Spectrometer from CID Bio-Science. This is a badly designed instrument that returns bad data. It should not be used in scientific research. First it suffers from extreme problems of dark noise and stray light, as can be seen in the images in the company’s own advertising. More importantly, because of the light source used and the configuration of the instrument entrance optics the sum of reflectance, transmittance and absorptance reported is very far from the theoretical value of one, and much larger, i.e., the values reported as reflectance, transmittance and absorptance, are not these physical quantities! (The company has been aware of these design problems for several years, as they recognized when I reported them several years ago in a exchange of several e-mails.)

4 Instruments

4.1 Ergonomics and usability

The first impression of the Walz LSA-2050 is that of a refined design and very well tested instrument, making it reliable, ergonomically easy to use and with very well thought out and clear firmware user interface. In some respects it benefits from Walz’s long experience in chlorophyll fluorescence measurements, as well as from the approaches used in the Dualex instruments. The LSA-2050 does the measurements very fast. With the help from its good ergonomics, I was able to measure 15 leaves, each from a different plant, across four growth-chamber shelves in 4 min.

The MPM-100 is not really a clone of the Dualex in its measurement approach but attempts to be a replacement for it. At least the version I have used, has a display with very small text, even if in colour, and some significant rough edges in the firmware user interface. The most important being that when a measurement fails, the screen displays the results from the last good measurement, which is prone to cause mistakes in the manual recording of readings. Another quirk is that when the estimation of one parameter fails the MPM-100 returns no results, which is problematic. In contrast, the Dualex reports those values that could be measured. The MPM-100 is also slower than the Dualex as it averages several consecutive measurements.

The Dualex 4 is also a well designed instrument, a refinement of the earlier Dualex 3 series. It has a couple weak points, though: the hinge is rather weak and with long-term use can develop play or even break. In the first case some misalignment of the two halves of the measurement head can occur. The Dualex has more options for encoding treatments than the Walz, making it more cumbersome, but more flexible in the tagging of logged data.

The SPAD-502 is no longer available in the simple version I have used. The SPAD-502Plus with data logging capabilities is also a well designed and easy to use instrument. I have not used any of the cheaper clones.

4.2 Cost and availability

Instruments for the non-destructive measurement of leaf optical properties related to metabolite concentrations.
Instrument	Handheld	Make	Availability
LSA-2050	y	Walz	2024-
MPM-100	y	Opti-Sciences
MPM-100/S	y	Opti-Sciences
Dualex 4	y	Force-A
Dualex 3 FLAV	y	Force-A
Dualex 3 ANTH	y	Force-A
Dualex 3 CA	y	Force-A
UVA-PAM	y	Walz/
UVA-PAM/red	y	Walz/
Xe-PAM	n	Walz
SPAD-502	y	Konica/Minolta
PlantPen NDVI 310	y	PSI
PlantPen PRI 210	y	PSI
NPen N 110 (NDGI)	y	PSI
MC-100	y	Apogee
CCM-200	y	Opti-Sciences

The SPAD-502Plus sells for 2500-3000 €, with various clones available in the range 600-1500 €. The Dualex 4, now Metos from Pessl after the bankruptcy of Force-A, costs close to 4500 €, the MPM-100 costs close to 3000 € and the Walz LSA-2050 costs close to 6000 €. (These prices are approximate as they vary depending on supplier, import taxes, etc., as well as with accessories ordered and in some cases with the instrument configuration).

5 Instrument tests and intercomparisons

Instruments compared	Measured	Variation from	Reference
SPAD-502, MC-100, Multiplex 3.6, atLEAF+	Chl	Nitrogen supply	Padilla et al. (2018)
SPAD-502, CCM-200, Dualex 4	Chl	Four crops	Dong et al. (2019)
SPAD-502, CCM-200	Chl	22 species	Parry et al. (2014)
SPAD-502, Dualex	Chl + Flav	Multiple	(Meyer2006?)
Dualex 4, LSA-2050, UV-A-PAM	Chl + Flav + Anth	Irradiance	(Bilger2026?)

6 Data logging and data exchange

In this section I only discuss equipment that I have used and are capable of storing data.

6.1 Dualex 4

6.2 MPM-100

The MPM-100 logs measurements to its internal memory. It logs both computed values and raw values. The computed values are calculated based on instructions in a script, with a default script preinstalled. The data are downloaded as a single Excel worksheet. The values in the worksheet are numerical, not equations, simplifying import of the data.

I could not find in the documentation information on whether the excitation is at constant irradiance or if it varies depending on how much is the fluorescence yield. Knowing this would be of interest when when interpreting the raw fluorescence values.

6.3 LSA-2050

The LSA-2050 logs measurements to its internal memory and these data can be downloaded as an Excel workbook using a Windows program provided for free with the instrument and also available for download in its current version. The worksheets contain both hidden and visible values. Many of the values are computed by formulas in the worksheet, something that makes import more complex. The worksheets as returned by the LSA software do seem to contain bare formulas, without cached computed values.

The computed values are necessary for import into foreign software such as R. Thus, the calculated values need to be added to the worksheets. The workbook is compatible with LibreOffice, but in contrast to Excel that automatically calculates the values on file load, in LibreOffice a hard recalculation needs to be manually triggered to have the computed values displayed instead of zeros. After saving the file it can be imported into R. At the moment, Linux and Mac OS X users are out-of-luck as no software for data download is yet available.

There are multiple R packages capable of reading Excel worksheets from workbooks. The first package I tried was the popular ‘readxl’ and it worked on my first attempt. However, the data are in two separate worksheets within a workbook.

When computed values for formulas are not present, the cells are displayed filled with zeros. If precomputed values are lacking reading the file into R also results is the import of zeros instead of the computed values.

The data in the “SAT chart” worksheet are time series with time points along rows, which is not how plotting functions of R expect the data to be organized. Because of this, even if the Excel worksheets can be easily read into R, additional processing is needed. In addition, in many cases only the computed values are needed for statistical analysis and plotting in R, making sub-setting necessary after import.

I have added function read_walz_lsa_xlsx() to package ‘photobiologyInOut’. This function reads the worksheets, checks that the computed values are available, and if not replaces the filler zeros with NA markers with a warning. The data from both worksheets are combined into a single data frame. The SAT time series are stored in one data frame per sample, nested inside the main data frame containing all raw and computed data for each sample in one row. Different subsets of the data in the workbook can also be imported when not all data are needed.

6.4 Example

Code

library(dplyr)


Attaching package: 'dplyr'

The following objects are masked from 'package:stats':

    filter, lag

The following objects are masked from 'package:base':

    intersect, setdiff, setequal, union

Code

library(photobiologyInOut)

Loading required package: photobiology

Loading required package: SunCalcMeeus

Documentation at https://docs.r4photobiology.info/

Code

library(ggplot2)
library(ggpmisc)

Loading required package: ggpp

Registered S3 methods overwritten by 'ggpp':
  method                  from   
  heightDetails.titleGrob ggplot2
  widthDetails.titleGrob  ggplot2


Attaching package: 'ggpp'

The following object is masked from 'package:ggplot2':

    annotate


Attaching package: 'ggpmisc'

The following objects are masked from 'package:photobiology':

    find_peaks, find_spikes, find_valleys

Code

library(patchwork)

Code

faba_light_before.df <-
  read_walz_lsa_xlsx("lza-dualex-mpm-data/faba-light-before.xlsx")

Dropping of geocode dependent data not yet implemented.

Code

faba_light_before.df$condition <- "light.240min"

Code

faba_20min_dark.df <-
  read_walz_lsa_xlsx("lza-dualex-mpm-data/faba-20min-dark.xlsx")

Dropping of geocode dependent data not yet implemented.

Code

faba_20min_dark.df$condition <- "dark.20min"

Code

faba_35min_dark.df <-
  read_walz_lsa_xlsx("lza-dualex-mpm-data/faba-35min-dark.xlsx")

Dropping of geocode dependent data not yet implemented.

Code

faba_35min_dark.df$condition <- "dark.35min"

Code

faba_light_after.df <-
  read_walz_lsa_xlsx("lza-dualex-mpm-data/faba-light-after.xlsx")

Dropping of geocode dependent data not yet implemented.

Code

faba_light_after.df$condition <- "light.10min"

Code

faba_191_LSA.df <-
  bind_rows(faba_light_before.df, 
            faba_20min_dark.df,
            faba_35min_dark.df,
            faba_light_after.df) |>
  mutate(condition = factor(condition),
         condition = reorder(condition, Time))

Code

ggplot(faba_191_LSA.df, aes(condition, nmol.cm2)) +
  geom_boxplot() +
  stat_multcomp(label.type = "letters", size = 2.7) +
  expand_limits(y = 0)

Code

ggplot(faba_191_LSA.df, aes(condition, AFLAV)) +
  geom_boxplot() +
  stat_multcomp(label.type = "letters", size = 2.7) +
  expand_limits(y = 0)

Code

ggplot(faba_191_LSA.df, aes(condition, AANTH)) +
  geom_boxplot() +
  stat_multcomp(label.type = "letters", size = 2.7) +
  expand_limits(y = 0)

Code

ggplot(faba_191_LSA.df, aes(condition, FV.FM)) +
  geom_boxplot() +
  stat_multcomp(label.type = "letters", size = 2.7)

Code

ggplot(faba_191_LSA.df, aes(condition, Fo)) +
  geom_boxplot() +
  stat_multcomp(label.type = "letters", size = 2.7)

Code

ggplot(faba_191_LSA.df, aes(condition, Fm)) +
  geom_boxplot() +
  stat_multcomp(label.type = "letters", size = 2.7)

Code

ggplot(faba_191_LSA.df, aes(condition, Fm - Fo)) +
  geom_boxplot() +
  stat_multcomp(label.type = "letters", size = 2.7)

Code

ggplot(faba_191_LSA.df, aes(nmol.cm2, AANTH)) +
  geom_point() +
  stat_correlation(use_label("r", "r.confint", "P"),
                   label.x = "right") +
  stat_ma_line() +
  stat_ma_eq(use_label("eq"), label.y = "bottom") +
  expand_limits(y = 0)

Code

n <- 6 # "plant"

(ggplot(faba_light_before.df$SAT.F.ls[[n]], aes(t, SAT.F)) +
  geom_point() + 
  geom_line() +
  expand_limits(y = c(0, 1200))) +
(ggplot(faba_20min_dark.df$SAT.F.ls[[n]], aes(t, SAT.F)) +
  geom_point() + 
  geom_line() +
  expand_limits(y = c(0, 1200))) +
(ggplot(faba_35min_dark.df$SAT.F.ls[[n]], aes(t, SAT.F)) +
  geom_point() + 
  geom_line() +
  expand_limits(y = c(0, 1200))) +
(ggplot(faba_light_after.df$SAT.F.ls[[n]], aes(t, SAT.F)) +
  geom_point() + 
  geom_line() +
  expand_limits(y = c(0, 1200)))

6.5 References

Dong, Taifeng, Jiali Shang, Jing M. Chen, et al. 2019. “Assessment of Portable Chlorophyll Meters for Measuring Crop Leaf Chlorophyll Concentration.” Remote Sensing 11 (22): 2706. https://doi.org/10.3390/rs11222706.

Hoel, Bernt Olav, and Knut AsbjØrn Solhaug. 1998. “Effect of Irradiance on Chlorophyll Estimation with the Minolta SPAD-502 Leaf Chlorophyll Meter.” Annals of Botany 82 (3): 389–92. https://doi.org/10.1006/anbo.1998.0683.

Markwell, John, John Osterman, and Jennifer Mitchell. 1995. “Calibration of the Minolta SPAD-502 Leaf Chlorophyll Meter.” Photosynthesis Research 46 (3): 467–72. https://doi.org/10.1007/BF00032301.

Padilla, Francisco M., Romina de Souza, M. Teresa Peña-Fleitas, Marisa Gallardo, Carmen Giménez, and Rodney B. Thompson. 2018. “Different Responses of Various Chlorophyll Meters to Increasing Nitrogen Supply in Sweet Pepper.” Frontiers in Plant Science 9 (November). https://doi.org/10.3389/fpls.2018.01752.

Parry, Christopher, J. Mark Blonquist, and Bruce Bugbee. 2014. “In Situ Measurement of Leaf Chlorophyll Concentration: Analysis of the Optical/Absolute Relationship.” Plant, Cell and Environment 37 (11): 2508–20. https://doi.org/10.1111/pce.12324.

Reuse

CC BY-SA 4.0

--- title: "In-vivo fast optical measurements of leaves" subtitle: "Walz LSA-2050, Force-A Dualex, SPAD, and similar instruments" author: "Pedro J. Aphalo" date: 2026-05-07 date-modified: 2026-05-21 keywords: [Walz LSA-2050] categories: [measurement and control] editor: markdown: wrap: 72 code-fold: true format: html: code-link: true code-tools: true bibliography: lsa-dualex-mpm.bib abstract: | Comparison of the Walz LSA-2050 handheld leaf analyzer to earlier instrument designs. This is preliminary assessment based mainly on specifications, scientific and technical literature and the author's user experience. It is based on long-term experience with the use of the SPAD-502 chlorophyll meter from Konica-Minolta, the Dualex 3.1 and Dualex 4 dual-excitation instruments from Force-A and recent use of the MPM-100 multiple wavelength pigment meter from Opti-Sciences and the LSA-2050 from Walz. I also show how to import the data logged by the LSA-2050 into R for further analysis. draft: true --- ## Introduction All *in vivo* measurements based on estimating leaf transmittance when expressed as concentrations are on a unit leaf area basis. Given that many variables in the growing environment affect specific leaf area (SLA) and leaf water content, differences in chlorophyll concentration between treatments depend strongly on the base of expression. In practice, the direction of a response to a treatment can be opposite when concentrations are expressed per unit area and per unit dry-mass or per unit fresh-mass. Leaves have a heterogeneous internal structure with many gas-liquid and liquid-solid interfaces. These interfaces scatter and reflect light while traveling inside a leaf. This lengthens the path of photons within the leaf enhancing absorption by pigments and altering the relationship between absorbance and concentration compared to the same concentration of pigments in a homogeneous solution. Because of this, the methods described below, when used to estimate concentrations depend on a calibration that is affected by the anatomy, water content and even recent illumination of the leaf. ### Leaf transmittance Chlorophyll concentration can be estimated based on the transmittance of the leaf at specific wavelengths where Chlorophyll is the main absorber of light [@Parry2014]. A leaf differs from a solution in that the chlorophyll distribution is not homogeneous. Not only there are cells, cell walls and vacuoles in cells but chlorophyll is located in the chloroplasts, and the location of chloroplasts within mesophyll cells changes in response to illumination. The changes in transmittance can even be used to study chloroplast accumulation movements [@justyna]. Thus, the relationship between chlorophyll concentration and light transmittance varies to some extent depending on the irradiance and colour of recent illumination [@Hoel1998], specially blue light sensed by phototropins results in chloroplast accumulation [@xxxx]. This has been described as a source of errors in chlorophyll measurements with the SPAD-502 instrument [@xxxx], but as this is an optical phenomenon in the plant it can be expected to affect all the instruments assessing chlorophyll based on leaf transmittance. The concentration of chlorophyll together with the lengthened path of photons inside a leaf due to scattering, makes leaves very effective absorbers such that transmittance at the wavelength of maximum absorption by chlorophyll is extremely low, making the measurement of high concentrations of chlorophyll _in-vivo_ with the SPAD-502 error-prone. The Opti-Sciences CCM-200 and the Apogee MC-100 chlorophyll meters use similar wavelengths as the SPAD-502. The Dualex 4 introduced the approach of using a longer wavelength, were transmittance is higher but still mainly dependent on chlorophyll and little affected by other pigments and metabolites. This decreases measurements on healthy leaves with high concentrations of chlorophyll. The MPM-100 uses an approach and wavelengths similar to the Dualex 4 and the MPM-100/S uses an approach similar to the SPAD-502. The light scattering within leaves decreases the apparent transmittance even in the absence of absorbing pigments. This effect is estimated by measuring the apparent transmittance of far-red or near-infrared light by the leaf, as at these wavelengths absorption by pigments is minimal. Scattering is almost complete in leaves, and in the far-red region, because of the absence of absorption by pigments, approximately half of the incident light exits the leaf through each epidermis. The different instruments measure the attenuation of radiation in this region as a reference. For estimating chlorophyll concentration, the instruments described in @tab-leaf-tfr use two LEDs as light sources and a photodiode as detector. They are all hand-held devices for non destructive chlorophyll measurements on thin leaves. Some of them have additional functions, described in the next section. Transmittance at each wavelength can be estimated as the ratio of the photodiode signal with the leaf between the jaws and the photodiode signal with the leaf absent. The instruments described in @tab-leaf-tfr estimate the concentration of chlorophyll per unit leaf area based on leaf transmittance. The wavelengths used both as target and reference for assessing chlorophyll concentration vary across instruments. A reference is needed to assess the attenuation caused by light scattering independently of chlorophyll concentration. Differences in the wavelength used as reference as long as there is no absorption by pigments have minimal consequences. In contrast difference in the target wavelength are more significant because a measurement using the wavelength of peak absorption (near 660 nm) even if advantageous when assessing low concentrations makes the measurement of high concentrations difficult. As concentrations of chlorophyll in leaves can be high enough to result in very low transmittance, instruments that use longer wavelengths in the shoulder of the peak instead of at its cusp, are in most cases preferable. The LSA-2050, MPM-100, Dualex 4 and MC-100 return approximate concentrations on an area basis while the SPAD-502, MPM-100/S and some other SPAD clones return "SPAD units". The CCM-200 returns values of "CCI" index. The relationships between values in SPAD units and chlorophyll concentration and between values in CCI units and chlorophyll concentration are both curvilinear, but convex in opposite directions [@]. The _in vivo_ chlorophyll concentration estimates must be in general considered as approximate. For example, comparison of values obtained from leaves with differences in leaf anatomy, require in most cases separate calibrations for accuracy. | Instrument | Target | Reference | Quantity | Area | |:-----------|:-----------|:-----------|:-----------|:-----| | LSA-2050 | 715 (25) | 770 (30) | conc. | 79/28 | | MPM-100 | 720 | 850 | conc. | 71 | | MPM-100/S^*^ | 650 | 940 | SPAD index | 71 | | Dualex 4 | 720 | 810? | conc. | 20 | | MC-100^*^ | 650 | 930 | conc. | 71 | | CCM-200^*^ | 650 | 930 | CCI index | 71 | | SPAD-502^+^ | 650 | 940 | SPAD index | 6 | : Instruments for the non-destructive measurement of chlorophyll concentration based on leaf transmittance. Target and reference wavelengths, full width half maximum (FWHM) given in parentheses when available. Quantities are approximate concentrations or indexes with arbitrary units. The measured area is given in $\mathrm{mm}^2$ and in the case of the LSA-2050 can be reduced with a supplied mask for use on small leaves. Notes: ^*^no user experience by author. ^+^The SPAD-502 and SPAD-502 Plus, are identical except for the data logging available in the Plus version. The MPM-100 is available in customized versions based on LEDs emitting at non-standard wavelengths. The Walz LSA-2050 measures in addition to transmittance chlorophyll fluorescence $Fv/Fm$. The readings are expected not to depend on which side of the leaf is illuminated. In fact, the Dualex 4 and the LSA-2050, measure transmittance in opposite directions through leaves. The Dualex has the LEDs in the upper jaw and the sensor in the lower one, while the LSA-2050 locates them the other way around. Like the Dualex, the SPAD-502, MPM-100, MPM-100/S, CMM-200 and MC-100 have the LED in the upper jaw. The Walz LSA-2050 is aimed at scientific research, with built-in support for recalibration and corrections for different types of leaves. The Dualex 4 aims at being both useful in research and as a tool for crop management. The MPM-100 seems to target the same users as the Dualex 4. The CMM-200, MC-100 and specially the SPAD-502 are marketed as tools for crop management, although they are in practice also frequently used in plant research. The indexes are given by [@Parry2014] as: $$\mathrm{SPAD} = k \times \log \frac{T_{940}}{T_{650}} + C$$ where $k$ and $C$ are calibration constants. $$\mathrm{CCI} = \frac{T_{931}}{T_{653}}$$ Even if the reading from the SPAD is a value rather similar to absorbance, it is not linearly related to concentration of chlorophyll in the measured leaves. @Markwell1995 proposed building a calibration curve to convert SPAD units into chlorophyll concentration by fitting a non-linear function of the form: $$y = 10^{x^k}$$ where $x$ is the reading in SPAD units and $y$ the concentration of chlorophyll per unit leaf area. The value of $k$ depends on the individual SPAD instrument and obviously also on the units in which the chlorophyll concentration is expressed. @Markwell1995 described the value of $k$ as more dependent on the instrument than on the plant species or genotype. However, differences in anatomy can be expected to affect the relationship between transmittance and concentration, independently of the instrument used. The advantage of using this function compared to a polynomial or spline, is that it constrains the shape of the curve, which is important when calibrations are based on limited data. ### Epidermal transmittance Fluorescence is dependent on the number of absorbed photons, this also applies to chlorophyll. Thus, a measurement of fluorescence behind the epidermis, from the chlorophyll in the mesophyll, can be used to estimate the transmittance of the epidermis if it is possible to estimate the fluorescence yield that can be expected in the absence of the epidermis. One approximation is to use a wavelength that is known not to be absorbed by the epidermis to obtain a reference value of fluorescence. This value can be compared against fluorescence excited by other wavelengths where the epidermis can absorb. This the principle of the Dualex (dual excitation), similar to the earlier approach used by the UVA-PAM, a modified Walz Mini-PAM. The Dualex, however, measures the fluorescence on the opposite side of the leaf from the excitation, while the UVA-PAM and the Walz LSA-2050 measure the fluorescence from the side of the leaf where the excitation light impings. While the Dualex can be used only on thin leaves, the UVA-PAM and the LSA-2050 can be used on thicker objects like fruits. In the case of the LSA-2050 by removing the lower jaw, which contains the light sources for estimation of chlorophyll based on transmittance across the leaf. The UVA_PAM uses a single light guide, and does not measure transmittance. The MPM-100 senses the chlorophyll fluorescence on the same side of the leaf as the excitation impings. However the head has a maximum opening that limits its use to thin objects like leaves. The instruments described in @tab-epidermis-tfr use LEDs as light sources and photodiodes, in most cases filtered, as detectors. They are hand-held devices for non destructive measurements of thin leaves, and in some cases also suitable for thick leaves, fruits and stems. Walz's Xe-PAM in the exception, it uses a filtered Xenon lamp as light source and a photo-multiplier tube as detector, is not handheld and usually used in a laboratory. The instruments in @tab-epidermis-tfr rely on the excitation of chlorophyll fluorescence with its intensity used as reporter inside the leaf for estimation of the transmittance of the epidermis. All the Dualex instruments sense the fluorescence on the opposite side of the leaf to the excitation, while other instruments sense it on the same side. These instruments differ in the target and reference wavelengths used for assessing epidermal transmittance. The different transmittances depend on the accumulation of different metabolites: UV-B, phenolic acids, UV-A1, flavonoids, blue, non-photosynthetic-related carotenoids, green, anthocyanins. Either blue- or red-excited fluorescence is used as reference assuming that these wavelengths are minimally absorbed in the epidermis. | Instrument | UV-B | UV-A1 | Blue | Green | Red^**^ | |:-----------|:-----------|:-----------|:-----------|:-----------|:-----------| | LSA-2050 | 310 (15) | 365 (12) | 450 (14) | 530 (27) | 630 (24) | | MPM-100 | | 375 | | 525 | 660 | | MPM-100/S^*^ | | 375 | | 525 | 660 | | Dualex 4 | | 375 | | 520 | 630 | | Dualex 3 FLAV | | 375 | | | 630 | | Dualex 3 ANTH^*^ | | | | 520 | 630 | | Dualex 3 CA | 310 | | | | 630 | | UVA-PAM^*^ | | 375 | 470 | | | | UVA-PAM/red^*^ | | 375 | | | 630? | | Xe-PAM^*^ | 314 (18) | 366 (32) | 475 (140) | | | : Instruments for the non-destructive measurement of epidermal transmittance using chlorophyll fluorescence as reporter. Excitation wavelengths, full width half maximum (FWHM) given in parentheses when available. Instruments names followed by ^*^ indicate that the author has no direct experience in their use. The MPM-100 is also available in customized versions based on LEDs emitting at non-standard wavelengths. The Walz LSA-2050 measures in addition to transmittance chlorophyll fluorescence Fv/Fm. ## Leaf reflectance The instruments described in @tab-leaf-rfr measure reflectance of leaves, each at a pair of wavelengths. These reflectance values can be combined into indexes that are correlated to specific plant conditions. These indexes are typically used in remote sensing but instruments for their measurement on individual leaves are also available. NDVI and NDGI mainly assess how green vegetation is while PRI assesses differences within the green region of the spectrum. | Instrument | Target | Reference | Quantity | |:-----------|:-----------|:-----------|:-----------| | PlantPen NDVI 310^*^ | 660 | 770 | NDVI | | PlantPen PRI 210^*^ |531 | 570 | PRI | | PlantPen/NPen N 110^*^ | 560 | 780 | NDGI | : Instruments for the non-destructive measurement of indexes based on leaf reflectance. Excitation wavelengths, full width half maximum (FWHM) given in parentheses when available. Instruments names followed by ^*^ indicate that the author has no direct experience in their use. A significant part of light "reflected" from a leaf is scattered light that has traveled inside the leaf and been partly absorbed by pigments, rather than reflected at the leaf surface. Thus, similarly to transmittance, reflectance at specific wavelengths, can inform about pigments within leaves. The PlantPen instruments from PSI, as well as remote sensing, RGB and spectral imaging of vegetation rely on this phenomenon. All of the instruments described in @tab-leaf-rfr return values that are indexes rather than concentrations. The indexes as implemented in these instruments are: $$\mathrm{PRI} = \frac{R_{531} – R_{570}}{R_{531} + R_{570}}$$ $$\mathrm{NDVI} = \frac{R_{660} – R_{770}}{R_{660} + R_{770}}$$ $$\mathrm{NDGI} = \frac{R_{560} – R_{780}}{R_{560} + R_{780}}$$ where $R_\lambda$ is reflectance on a band centred at a given wavelength in nanometres. The half maximum full width (HMFW) of the excitation is not given in the specifications. There are variations on the exact wavelengths and HMFW used for these indexes that are in wide use, and this must be taken into account when comparing measurements. There are indexes like NBI that combine different types of measurements, and these can be obtained only with the more advanced instruments, Dualex 4 and LSA-2050. The Nitrogen Balance Index (NBI) is obtained by dividing the chlorophyll concentration index ($\approx \mu g\, cm^{-2}$) by the epidermal flavonols index, AFLAV ($\approx A_{375}$) (using inconsistent units), to obtain a value sometimes described as "unitless". ## Alternative approaches With a spectrometer it is possible to measure whole leaf spectral reflectance and transmittance, and from them derive absorptance over arbitrary wavelength ranges within the overlapping sensitivity range of the spectrometers and the wavelength range of the illumination. The measurement of epidermal transmittance with a spectrometer is limited to the few species in which the epidermis can be stripped from leaves and measured by itself. Ocean Optics used to make a portable spectrometer equipped with small integrating spheres in a clip-like arrangement with one sphere in each jaw. In use one should be aware that two spheres collecting light and reflecting it back onto the leaf can result in erroneous readings as some light can travel back and forth across the leaf. This must be taken into consideration in the measuring protocol used, but it is not a design flaw. I should mention here also the *SpectraVue Leaf Spectrometer* from CID Bio-Science. This is a **badly designed** instrument that returns **bad data**. It should not be used in scientific research. First it suffers from extreme problems of dark noise and stray light, as can be seen in the images in the company's own advertising. More importantly, because of the light source used and the configuration of the instrument entrance optics the sum of reflectance, transmittance and absorptance reported is very far from the theoretical value of one, and much larger, i.e., the values reported as reflectance, transmittance and absorptance, are not these physical quantities! (The company has been aware of these design problems for several years, as they recognized when I reported them several years ago in a exchange of several e-mails.) ## Instruments ### Ergonomics and usability The first impression of the *Walz LSA-2050* is that of a refined design and very well tested instrument, making it reliable, ergonomically easy to use and with very well thought out and clear firmware user interface. In some respects it benefits from Walz's long experience in chlorophyll fluorescence measurements, as well as from the approaches used in the Dualex instruments. The *LSA-2050* does the measurements very fast. With the help from its good ergonomics, I was able to measure 15 leaves, each from a different plant, across four growth-chamber shelves in 4 min. The *MPM-100* is not really a clone of the Dualex in its measurement approach but attempts to be a replacement for it. At least the version I have used, has a display with very small text, even if in colour, and some significant rough edges in the firmware user interface. The most important being that when a measurement fails, the screen displays the results from the last good measurement, which is prone to cause mistakes in the manual recording of readings. Another quirk is that when the estimation of one parameter fails the MPM-100 returns no results, which is problematic. In contrast, the Dualex reports those values that could be measured. The MPM-100 is also slower than the Dualex as it averages several consecutive measurements. The *Dualex 4* is also a well designed instrument, a refinement of the earlier *Dualex 3* series. It has a couple weak points, though: the hinge is rather weak and with long-term use can develop play or even break. In the first case some misalignment of the two halves of the measurement head can occur. The Dualex has more options for encoding treatments than the Walz, making it more cumbersome, but more flexible in the tagging of logged data. The *SPAD-502* is no longer available in the simple version I have used. The *SPAD-502Plus* with data logging capabilities is also a well designed and easy to use instrument. I have not used any of the cheaper clones. ### Cost and availability | Instrument | Handheld | Make | Availability | |:-----------|:-----------|:-----------|:-----------| | LSA-2050 | y | Walz | 2024- | | MPM-100 | y | Opti-Sciences | | | MPM-100/S | y | Opti-Sciences | | | Dualex 4 | y | Force-A | | | Dualex 3 FLAV | y | Force-A | | | Dualex 3 ANTH | y | Force-A | | | Dualex 3 CA | y | Force-A | | | UVA-PAM | y | Walz/ | | | UVA-PAM/red | y | Walz/ | | | Xe-PAM | n | Walz | | | SPAD-502 | y | Konica/Minolta | | | PlantPen NDVI 310 | y | PSI | | | PlantPen PRI 210 | y | PSI | | | NPen N 110 (NDGI) | y | PSI | | | MC-100 | y | Apogee | | | CCM-200 | y | Opti-Sciences | | : Instruments for the non-destructive measurement of leaf optical properties related to metabolite concentrations. The *SPAD-502Plus* sells for 2500-3000 €, with various clones available in the range 600-1500 €. The *Dualex 4*, now Metos from Pessl after the bankruptcy of Force-A, costs close to 4500 €, the *MPM-100* costs close to 3000 € and the *Walz LSA-2050* costs close to 6000 €. (These prices are approximate as they vary depending on supplier, import taxes, etc., as well as with accessories ordered and in some cases with the instrument configuration). ## Instrument tests and intercomparisons | Instruments compared | Measured | Variation from | Reference | |:---------------------|:------------|:------------|:----------| | SPAD-502, MC-100, Multiplex 3.6, atLEAF+ | Chl | Nitrogen supply | @Padilla2018 | | SPAD-502, CCM-200, Dualex 4 | Chl | Four crops | @Dong2019 | | SPAD-502, CCM-200 | Chl | 22 species | @Parry2014 | | SPAD-502, Dualex | Chl + Flav | Multiple | @Meyer2006 | | Dualex 4, LSA-2050, UV-A-PAM | Chl + Flav + Anth | Irradiance | @Bilger2026 | ## Data logging and data exchange In this section I only discuss equipment that I have used and are capable of storing data. ### Dualex 4 ### MPM-100 The MPM-100 logs measurements to its internal memory. It logs both computed values and raw values. The computed values are calculated based on instructions in a script, with a default script preinstalled. The data are downloaded as a single Excel worksheet. The values in the worksheet are numerical, not equations, simplifying import of the data. I could not find in the documentation information on whether the excitation is at constant irradiance or if it varies depending on how much is the fluorescence yield. Knowing this would be of interest when when interpreting the raw fluorescence values. ### LSA-2050 The LSA-2050 logs measurements to its internal memory and these data can be downloaded as an Excel workbook using a Windows program provided for free with the instrument and also available for download in its current version. The worksheets contain both hidden and visible values. Many of the values are computed by formulas in the worksheet, something that makes import more complex. The worksheets as returned by the LSA software do seem to contain bare formulas, without cached computed values. The computed values are necessary for import into foreign software such as R. Thus, the calculated values need to be added to the worksheets. The workbook is compatible with LibreOffice, but in contrast to Excel that automatically calculates the values on file load, in LibreOffice a hard recalculation needs to be manually triggered to have the computed values displayed instead of zeros. After saving the file it can be imported into R. At the moment, Linux and Mac OS X users are out-of-luck as no software for data download is yet available. There are multiple R packages capable of reading Excel worksheets from workbooks. The first package I tried was the popular 'readxl' and it worked on my first attempt. However, the data are in two separate worksheets within a workbook. ::: callout-attention When computed values for formulas are not present, the cells are displayed filled with zeros. If precomputed values are lacking reading the file into R also results is the import of zeros instead of the computed values. ::: The data in the "SAT chart" worksheet are time series with time points along rows, which is not how plotting functions of R expect the data to be organized. Because of this, even if the Excel worksheets can be easily read into R, additional processing is needed. In addition, in many cases only the computed values are needed for statistical analysis and plotting in R, making sub-setting necessary after import. I have added function `read_walz_lsa_xlsx()` to package 'photobiologyInOut'. This function reads the worksheets, checks that the computed values are available, and if not replaces the filler zeros with `NA` markers with a warning. The data from both worksheets are combined into a single data frame. The SAT time series are stored in one data frame per sample, nested inside the main data frame containing all raw and computed data for each sample in one row. Different subsets of the data in the workbook can also be imported when not all data are needed. ### Example ```{r} library(dplyr) library(photobiologyInOut) library(ggplot2) library(ggpmisc) library(patchwork) ``` ```{r} faba_light_before.df <- read_walz_lsa_xlsx("lza-dualex-mpm-data/faba-light-before.xlsx") faba_light_before.df$condition <- "light.240min" ``` ```{r} faba_20min_dark.df <- read_walz_lsa_xlsx("lza-dualex-mpm-data/faba-20min-dark.xlsx") faba_20min_dark.df$condition <- "dark.20min" ``` ```{r} faba_35min_dark.df <- read_walz_lsa_xlsx("lza-dualex-mpm-data/faba-35min-dark.xlsx") faba_35min_dark.df$condition <- "dark.35min" ``` ```{r} faba_light_after.df <- read_walz_lsa_xlsx("lza-dualex-mpm-data/faba-light-after.xlsx") faba_light_after.df$condition <- "light.10min" ``` ```{r} faba_191_LSA.df <- bind_rows(faba_light_before.df, faba_20min_dark.df, faba_35min_dark.df, faba_light_after.df) |> mutate(condition = factor(condition), condition = reorder(condition, Time)) ``` ```{r} ggplot(faba_191_LSA.df, aes(condition, nmol.cm2)) + geom_boxplot() + stat_multcomp(label.type = "letters", size = 2.7) + expand_limits(y = 0) ``` ```{r} ggplot(faba_191_LSA.df, aes(condition, AFLAV)) + geom_boxplot() + stat_multcomp(label.type = "letters", size = 2.7) + expand_limits(y = 0) ``` ```{r} ggplot(faba_191_LSA.df, aes(condition, AANTH)) + geom_boxplot() + stat_multcomp(label.type = "letters", size = 2.7) + expand_limits(y = 0) ``` ```{r} ggplot(faba_191_LSA.df, aes(condition, FV.FM)) + geom_boxplot() + stat_multcomp(label.type = "letters", size = 2.7) ``` ```{r} ggplot(faba_191_LSA.df, aes(condition, Fo)) + geom_boxplot() + stat_multcomp(label.type = "letters", size = 2.7) ``` ```{r} ggplot(faba_191_LSA.df, aes(condition, Fm)) + geom_boxplot() + stat_multcomp(label.type = "letters", size = 2.7) ``` ```{r} ggplot(faba_191_LSA.df, aes(condition, Fm - Fo)) + geom_boxplot() + stat_multcomp(label.type = "letters", size = 2.7) ``` ```{r} ggplot(faba_191_LSA.df, aes(nmol.cm2, AANTH)) + geom_point() + stat_correlation(use_label("r", "r.confint", "P"), label.x = "right") + stat_ma_line() + stat_ma_eq(use_label("eq"), label.y = "bottom") + expand_limits(y = 0) ``` ```{r} n <- 6 # "plant" (ggplot(faba_light_before.df$SAT.F.ls[[n]], aes(t, SAT.F)) + geom_point() + geom_line() + expand_limits(y = c(0, 1200))) + (ggplot(faba_20min_dark.df$SAT.F.ls[[n]], aes(t, SAT.F)) + geom_point() + geom_line() + expand_limits(y = c(0, 1200))) + (ggplot(faba_35min_dark.df$SAT.F.ls[[n]], aes(t, SAT.F)) + geom_point() + geom_line() + expand_limits(y = c(0, 1200))) + (ggplot(faba_light_after.df$SAT.F.ls[[n]], aes(t, SAT.F)) + geom_point() + geom_line() + expand_limits(y = c(0, 1200))) ``` ### References ::: {#refs} :::