May 25, 2026

Duckweed gas exchange protocol

DOI
https://dx.doi.org/10.17504/protocols.io.81wgbo111lpk/v1
  • Alison Gill1,2,
  • Robert Rintoul1,2,3,
  • Lorna McAusland1,3,
  • Erik Murchie1,3,
  • Matthew Gilliham1,2,
  • Jenny Mortimer1,2
  • 1Australian Research Council Centre of Excellence in Plants for Space;
  • 2School of Agriculture, Food and Wine and the Waite Research Institute, University of Adelaide, Urrbrae, South Australia, Australia;
  • 3Division of Plant and Crop Sciences, School of Biosciences, University of Nottingham, Sutton Bonington, Leicestershire, United Kingdom
  • Alison Gill: Joint first authors;
  • Robert Rintoul: Joint first authors;
Alison Gill
Icon indicating open access to content
QR code linking to this content
DOI: https://dx.doi.org/10.17504/protocols.io.81wgbo111lpk/v1
Protocol Citation: Alison Gill, Robert Rintoul, Lorna McAusland, Erik Murchie, Matthew Gilliham, Jenny Mortimer 2026. Duckweed gas exchange protocol. protocols.io https://dx.doi.org/10.17504/protocols.io.81wgbo111lpk/v1
Manuscript citation:
Rintoul R, Gill A, McAusland L, Murchie E, Gilliham M, Mortimer J. 2026. Developing carbon assimilation methods in duckweed for new insights into photosynthesis and growth mechanisms. Plant Physiology.
License: This is an open access  protocol  distributed under the terms of the  Creative Commons Attribution License,  which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: December 22, 2025
Last Modified: May 25, 2026
Protocol  Integer ID: 235609
Keywords: co2 responses in duckweed, duckweed gas exchange protocol infrared gas analysis, co2 uptake in plant, instantaneous measurements of duckweed sample, measuring co2 response, quantifying co2 uptake, duckweed sample, co2, application to aquatic species, photosynthetic mechanism, duckweed, floating angiosperm, plant, aquatic chamber, aquatic species, customised aquatic chamber
Funders Acknowledgements:
Australian Research Council Centre of Excellence in Plants for Space
Grant ID: CE230100015
Adelaide-Nottingham University joint scholarship
Grant ID: RCR
Abstract
Infrared gas analysis (IRGA) is the primary technique for quantifying CO2 uptake in plants, but technical obstacles have limited its application to aquatic species, inhibiting exploration of their photosynthetic mechanisms. Using an IRGA with a customised aquatic chamber, we developed methodology for measuring CO2 responses in duckweed, a fast-growing, floating angiosperm. This protocol can be used to generate light, CO2 and temperature response curves as well as take instantaneous measurements of duckweed samples. See doi.org/10.1093/plphys/kiag311 for the associated paper.
Instrument and peripherals
This methodology uses a LI-6800 portable photosynthesis system equipped with aquatic chamber (6800-18) to measure aquatic gas exchange in duckweed. Usually, the aquatic chamber volume is used to contain an algal suspension, where air is ‘bubbled’ uniformly throughout, allowing gas exchange between the air and suspension, before flowing through to the sample infrared gas analyser (IRGA) for measurement. This design intended for algal suspensions or similar (Figure 1).



Figure 1. (Top) LI-6800 with aquatic chamber, set up for measuring gas exchange in an algal suspension. Used with permission from LICOR, available at: https://www.licor.com/products/photosynthesis/LI-6800/aquatic. (Bottom) Schematic of the aquatic chamber flow path. The LI-6800 head has the flow partition valve, sample and reference infra-red gas analyzers (IRGAs), and the mixing volume for sample air. The water vapor equilibrator brings water vapor in the sample and reference air streams into equilibrium. The cuvette holds the aquatic sample. Sample air is routed in a subsample loop through the aquatic sample, water vapor equilibrator, and mixing fan volume before being routed to the sample IRGA, where CO2 and H2O are measured. Used with permission from LICOR, available at: https://www.licor.com/support/LI-6800/topics/chamber-aquatic-overview.html.

We used sample adaptor kit (LI-COR part no. 9968-338) to replace the sample window. This plug contains a well which can hold 1 mL of media with duckweed samples. The LI-COR 6800 sensor head and chamber can be oriented ‘handle up’ so that duckweeds are positioned parallel to the air flow, similar to wind blowing across the surface of a pond, with the duckweed covering the media surface (Fig. 1). This allowed us to measure duckweed gas exchange closer to in situ, rather than the duckweed being bubbled aggressively like algal suspensions.



Figure 2. (Top) LI-6800 with sample adaptor kit and alternative orientation. (Bottom) Schematic of the 6800-18 orientation for duckweed with sample adaptor well and airflow diagram. For more detail, see doi.org/10.1093/plphys/kiag311.
To increase the accuracy of temperature control, a recirculating water bath should be attached to ports on the aquatic chamber. Texchange (which heats air within the mixing fan volume) can be controlled and the re-circulating water bath minimises any potential cooling effect from the sample chamber block and ambient air temperature. However, the sample temperature cannot be controlled or measured directly.

For all measurements of duckweed gas exchange to follow, set conditions to the following: flow 500 µmol s-1 with pump speed at auto, humidity (air) 70%, CO2 (reference; Ca) 400 µmol mol-1, temperature (Texchange) 25˚C, and mixing fan speed 10,000 rpm.
Preparing the aquatic chamber for samples
The aquatic chamber must be prepared in three steps for aquatic samples: (1) Warm up tests, (2) hydrating the membrane, and (3) loading the duckweed samples.
The warmup procedure follows that of other LI-6800 chambers:
  • Replenish instrument chemicals (soda lime, sorbeads, and water) and attach a new CO2 canister.
  • Ensure chamber plugs have been replaced, and that the aquatic pump is OFF (if left ON this will result in failed Sample Cell Calibration tests, suggesting a leak to the chamber) with no media in the sample adaptor well.
  • Run the warm up tests and address any potential issues as per LI-COR’s advice.
The semi-permeable membrane must be hydrated before any measurements take place, during which the IRGAs warm up, improving the overall stability of the system:
  • Turn the aquatic pump ON (the aquatic pump should remain ON whilst there is media within the chamber).
  • Load the sample adaptor well with 1 mL of media or water (this methodology utilises N-Medium).
  • Set the environmental constants as per the subsequent methodology to be run.
  • Allow the LI-6800 to run for approximately 1 hour for the IRGAs to sufficiently warm up and reach stability.
Finally, duckweed samples can be loaded into the sample adaptor well:
  • Remove and replace the 1 mL of media from the sample adaptor well with fresh media
  • Using an inoculating loop or similar, transfer duckweed fronds into the sample adaptor well to cover most of the surface
  • Allow the duckweed and media to reach steady state gas exchange before beginning measurements. In general, this ranges from 2 to 15 minutes depending on chamber conditions and how different they are to growing conditions. For conditions similar to stock growth conditions, steady state can be reached rapidly, within 2 minutes. For conditions with much greater CO2 or light intensity than ambient, steady state can take up to 15 minutes.
Correcting for incident light on sample plane
Care needs to be taken when setting light intensity, as the incident light on the sample (Qsample) is just over 50% of the controlled Qin parameter. This is due to light loss between the internal surfaces of the chamber, which limits the maximum value for Qsample to around 1611 µmol m-2 s-1, and is a key point of difference to measuring leaf gas exchange in the regular 6800 chambers. The fluorometer output (Qf), is again likely higher due to the light lost as it passes through the first sample window; however, this is already accounted for in the Qin value. The 6800-18 user manual details the method for determining Qsample, which can be calculated from Equations 1-3, with components defined in Table 1 and set points for subsequent light curves defined in Table 2.

  • Equation 1.

  • Equation 2.

  • Equation 3.

Table 1. Components used in calculating Qsample.

Table 2. Qin set points for respective Qsample light intensities, 90% red light and 10% blue light at Qin.


Correcting headspace CO2
The aquatic chamber is a two-compartment system, where the duckweed is in one compartment and the sample IRGA is in the second compartment. Connecting these is the sub-sample loop; this circulates air flow from the mixing fan volume, through the sample chamber, then to either the sample IRGA or back to the sample chamber. As the measured air is only a fraction of the air sampled from the headspace, rather than the full volume of headspace air, best practice is to correct for differences in CO2 that arise between the two compartments. To determine the proportion of air that flows to the sample IRGA, users must first measure the subsample flow rate (outlined in the 6800-18 user manual (LI-COR, 2022b), page 100). To correct for the potential difference in CO2, the actual ΔCO2 between the sample and reference IRGA can calculated using Equation 4, which in turn is used to calculate the headspace atmospheric CO2 mole fraction (Ca) using Equation 5. With components described in Table 3.

  • Equation 4.

  • Equation 5.

Table 3. Components for calculating actual headspace ΔCO2 and headspace atmospheric CO2 mole fraction (Ca).

For subsequent CO2 response curves, N-Medium alone (no duckweed) was used to determine CO2 setpoints for each response curve. When modelling response curves the headspace CO2 was calculated retrospectively for each replicate.
Light response curves
  • Light response curves are run close to ambient CO2 at 400 µmol mol-1.
  • Equilibrate the duckweed in the chamber at high light intensity (2979.52 µmol m-2 s-1) for sufficient time to allow for a steady state. This can take some time, often 10-15 min.
  • Once the CO2 flux is at steady state, run the AUTO MATCH function from the Measurement tab – this calculates a measurement correction factor that accounts for differences between the reference and sample IRGA readings.
  • When CO2 flux is again at steady state post-match, a light response program can be run. In these examples the following 13 light intensity (µmol m-2 s-1) setpoints were used as Qsample: 1600, 1400, 1200, 1000, 800, 600, 400, 200, 100, 50, 0. The setpoints entered into the LI-6800 as Qin are given in Table 2.
  • The minimum and maximum wait times are set at 120 and 180 seconds, respectively. This may need adjusting on a species basis, however, was appropriate to allow steady state assimilation at each light set point for the clones investigated. The total maximum runtime for these curves was 39 minutes. As reference CO2 concentration did not change, matching is not vital but can still be set to auto match between measurements.
  • For noisy samples, it is possible to take average values over a certain time (e.g., 20-45 s) using the setting in the “Log setup” tab. This should be considered for all measurements.
CO2 response curves
As it is not possible to measure Ci in the aquatic chamber and therefore present routine A/Ci curves, CO2 flux as a function of increasing reference CO2 concentration (Ca) is used for duckweed instead.
  • The light intensity is chosen based on the previously measured light response curve. We determined that 1000 µmol m-2 s-1 at the sample (Qin of 1862.20 µmol m-2 s-1) was saturating for all three species, using the photosynthesis package in R (Stinziano et al., 2023).
  • Equilibrate the duckweed in the chamber for sufficient time to allow for a steady state. This can take some time, often 10-15 min.
  • Ensure that CO2 flux is at steady state, then run the AUTO MATCH function.
  • When CO2 flux is again at steady state post-match, a CO2 response program can be run. In these examples the following 14 CO2 reference concentration (µmol mol-1) setpoints were used: 400, 300, 200, 100, 50, 0, 400, 500, 600, 700, 800, 1000, 1200, 1400.
  • The minimum and maximum wait times are set at 600 and 660 seconds, respectively. This may need adjusting on a species basis, however, was appropriate to allow steady state assimilation at each light set point for the clones investigated as well as ensure equilibrium between the air and media CO2 concentration. It is important that the wait time is not too short to allow the sample to adjust to the new CO2 concentration, but also not too long, as the biochemistry of the sample will be altered over time. The total maximum runtime for these curves was 2 hours 34 minutes.
  • Matching is vital at every CO2 concentration to ensure that the reference and sample analyser are close to equal.
Temperature response curves
  • The light intensity and CO2 concentration setpoints are chosen using the previously determined saturating light intensity (i.e., 1000 µmol m-2 s-1 at the sample (Qin of 1862.20 µmol m-2 s-1)) and desired CO2 concentration (usually close to ambient, i.e., 400 µmol mol-1 CO2).
  • A recirculating water bath is used to control the temperature of Taquatic, starting with the lowest temperature setpoint (17.5°C) and Texchange set to the same setpoint.
  • Once duckweed is added to the chamber adaptor well, allow the CO2 flux to reach steady state. This takes around 10-15 minutes. Run the AUTO MATCH function.
  • Wait for CO2 flux to return to steady state post-match and equilibrate, then take a log.
  • The water bath temperature is then increased to the next temperature (°C), experimental range as follows: 17.5, 20, 22.5, 25, 27.5, 30, 32.5, 35, 37.5. Our water bath took approximately 15 minutes to reach the next temperature.
  • Once Taquatic stabilised to the same temperature as the water bath, samples are held at the respective temperature for 300 seconds before taking the next log to allow for full temperature acclimation. In order to reduce variability, an average of the CO2 flux over the previous 45 s was logged, instead of a single timepoint, using the “Logging options” settings in the “Log setup” tab. The total maximum runtime for these curves was about 3 hours. AUTO MATCH was run manually a few minutes prior to the log being taken.
Instantaneous measurements
Instantaneous (or survey) measurements measure the rates of photosynthetic gas exchange under the environmental conditions that the plant is experiencing. In this case, wait times should be reduced to a minimum to log data as soon as conditions in the chamber have equilibrated but before the duckweed adjusts to the new conditions. Usually, Qsample, CO2, temperature and relative humidity should be set to match the duckweed growth conditions. In this instance, we set environmental conditions to 1000 µmol m-2 s-1 at the sample (Qin of 1862.2 µmol m-2 s-1), 25˚C and RH 70%, with measurements taken every 5 minutes from just before lights on at 7:00 to lights off at 23:00. The CO2 canister and N-Medium in the sample adaptor well were both replaced at the midpoint.
Sample normalisation
  • Area normalisation: Image the duckweed from an aerial view, with a ruler on the same plane as the fronds, and use ImageJ to calculate area (Schneider et al., 2012). Set the image scale using the ruler, and then determinethe 2D surface area is determined using the colour threshold tool, with parameters corresponding to the green area of duckweed fronds.
  • Fresh weight normalisation: Measure sample weight after gas exchange measurements using a fine balance which is capable of accurately measuring milligrams.
  • Dry weight normalisation: Flash freeze duckweeds in liquid nitrogen, then freeze dry for 16 hours.
  • Chlorophyll normalisation: Using the same freeze-dried samples, total chlorophyll a + b content was determined as described in Lichtenthaler and Babani (2022). For duckweed, see doi.org/10.1093/plphys/kiag311.
Data analysis and curve fitting
Curves can be fitted using the photosynthesis package (v. 2.1.5, Stinziano et al., 2021, 2023) in RStudio.