The following data analysis tool will fit Rapid Light Curve (RLC) (1) data using the equation developed by Webb et. al (2) in 1974, further adapted by Silsbe and Kromkampfor (3) in 2012 for use with PSII quantum yield values. The following equation is used:
ΦPSII (E) = alpha * Ek * (1-exp(-E/Ek)) * E-1
ΦPSII is Photosystem II quantum yield; E is the light intensity;
Rapid Light Curves are performed using various Pulse Amplitude Modulated (PAM) fluorometers, the principle of the method being a stepwise increase in light intensity followed by a measurement of PSII quantum yield at the end of each light step. These stepwise increases can be of various lengths 10 seconds to 1 minute, depending on the design. An additional design feature of the RLC that is up to the experimenter to decide is weather to dark adapt or not the sample before the RLC.
To use the provided analysis tool, you will need to know the light intensities used in your RLC, and the measured effective PSII quantum yield at each light intensity. Depending on the used instrument the effective PSII quantum yield may be shown as Y(II) (Walz instruments), QYLss (PSI instruments), or (Φ) PSII; not to be confused with the Maximum quantum yield of PSII (Fv/Fm) which is always measured on dark-adapted samples.
!Note there are no restrictions to the number of samples or light steps that you can add to the Raw data spreadsheet
Data Entry: Begin by entering the light intensities used during your RLC in the first row (#1). This row, highlighted in blue and bold, is reserved solely for light intensities. Follow up by entering corresponding Y(II) values for each RLC in subsequent rows.
Analysis: Once data is populated, click on "Analyze data".
!important note #1 - the data fitting process will attempt to provide the best data fit however, noisy data may result in unusual output values, such as alpha values of "1", which are not biologically plausible.
!important note #2 - Data quality is crucial for robust RLC analysis. To illustrate the range of data quality and its effects on the results, the "Raw data" is populated with eight pre-loaded examples. These samples demonstrate varying data quality, from high-quality ("Raw data" row #2) to poor-quality examples ("Raw data" row #9).
Datasets #1 to #3 exemplify high-quality data that yield dependable results from the model.
Dataset #4, while of lower quality, may still produce usable outcomes.
Datasets #5 to #9 contain poor-quality, noisy data that is generally unsuitable for analysis. Specifically, dataset #9 represents an extreme case of low-quality data; the model can still fit these values but tends to generate biologically implausible results, such as an alpha value of 1. It is recommended to avoid using such datasets and their corresponding model outputs for any serious analysis.
The results sheet will provide calculated parameters that describe your sample:
alpha (α) - represents the rate of photosynthesis under light-limiting conditions; higher values relative to control would indicate a better capacity to perform photosynthesis in low light - often seen in low-light adapted organisms.
Ek - represents the minimum light intensity at which photosynthesis is saturated; higher values relative to a control would indicate better capacity to perform photosynthesis in high light - often seen in high-light adapted organisms.
rETRmax - represents the relative maximum Electron Transfer Rate in PSII; higher values relative to a control would indicate better overall capacity to perform photosynthesis in environments with elevated light.
R-squared - represents a measure of how closely the data fits the model's predictions; Values leaning towards 1 signify an excellent fit, while those near 0 suggest the model might not be the most apt for your data..
The core of this tool is powered by JavaScript, utilizing the robust numeric.js library for efficient matrix operations and optimization routines. We employ an unconstrained fitting procedure, ensuring flexible and precise model fitting to diverse datasets.
Occasionally, data might not fit very well into the model (evident from a low R-squared). In such instances, consider aggregating data from multiple technical or biological replicates and re-run the analysis.
If nothing happens when you press "Analyse", or there is no value for R-squared make sure:
you don't have entire columns empty; right click (in Windows OS) on the column header allows the deletion of the entire column if needed
no missing data in the Raw data spreadsheet,
that there are no "0" values in the first row of the "Raw data" sheet.
Ralph, P. J., & Gademann, R. (2005). Rapid light curves: a powerful tool to assess photosynthetic activity. Aquatic botany, 82(3), 222-237. https://doi.org/10.1016/j.aquabot.2005.02.006
Webb, W. L., Newton, M., & Starr, D. (1974). Carbon dioxide exchange of Alnus rubra: a mathematical model. Oecologia, 17, 281-291. https://doi.org/10.1007/BF00345747
Silsbe, G. M., & Kromkamp, J. C. (2012). Modeling the irradiance dependency of the quantum efficiency of photosynthesis. Limnology and Oceanography: Methods, 10(9), 645-652. https://doi.org/10.4319/lom.2012.10.645