Dec 17, 2024
Measuring coral reflectance and calculating NDVI as a proxy for chlorophyll a with the DIVING-PAM-II
- Kay Watty1,
- Verena Schoepf1,
- Rene van der Zande1
- 1Department of Freshwater and Marine Ecology, Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, Amsterdam, The Netherlands
Protocol Citation: Kay Watty, Verena Schoepf, Rene van der Zande 2024. Measuring coral reflectance and calculating NDVI as a proxy for chlorophyll a with the DIVING-PAM-II. protocols.io https://dx.doi.org/10.17504/protocols.io.rm7vzjqd5lx1/v1
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: August 17, 2024
Last Modified: December 17, 2024
Protocol Integer ID: 105874
Abstract
Reflectance measurements have been widely used in remote sensing and visual ecology work to quantify the color of surfaces and organisms objectively. The Normalized Difference Vegetation Index (NDVI) is an established method to estimate chlorophyll a (Chl a) content in various organisms and ecosystems. It is based on the spectral properties of Chl a and the reflectance it causes in the red and near-infrared spectrum. The DIVING-PAM-II from Walz GmbH comes with a spectrometer (MINI-SPEC). It allows for spectral measurements in the range of 400 – 850 nm. This range is suitable for quantifying coral reflectance and subsequently calculating the NDVI. Estimating Chl a content non-invasively allows for continuous high-resolution monitoring of signs of bleaching and pigment loss during laboratory experiments or in the field. This protocol explains how to use the DIVING-PAM-II spectrometer to obtain coral reflectance measurements and calculate the NDVI as a proxy for Chl a.
Materials
- DIVING-PAM-II
- Spectrometer (MINI-SPEC) + accessories
- WinControl
Background
Background
Conventional methods for measuring chlorophyll a (Chl a) content in scleractinian corals typically require sacrificing the coral to access the Chl a. The main limitation of this method is the destructive sampling, which only provides data for limited time points. Coral bleaching is associated with the expulsion of symbionts and a consequent loss of Chl pigments, causing the coral to pale (Brown, 1997; Douglas, 2003). Despite standardized methods of measuring the progression of bleaching with coral health charts (Siebeck et al., 2006), determining color changes is often subjective and lacks high resolution to allow for the detection of small optical changes. Measuring color with spectrometers offers an objective quantification method that provides detailed information about the spectral properties of a sample. When measuring color, we are in fact measuring the variation in optical properties, one of which is reflectance (Johnsen, 2016).
Spectral reflectance is a unitless ratio where light reflected from an object is compared across a wavelength range to light reflected from a standard. The standard is usually a flat, diffuse white surface which reflects light equally in all directions. Measured reflectance is dependent on the angle of incident light to the detector. Therefore, established geometries are often used for complex biological surfaces in order to allow comparing new data with previous work (Johnsen, 2016).
Scleractinian corals can be broadly categorized into blue and brown corals with the brown category being more common (Hochberg et al., 2004). All corals depict a relatively low reflectance between 400 and 500 nm and a higher, peaked pattern between 550 and 650 nm. The presence of Chl a creates a unique spectral shape with a prominent and narrow absorption feature near 675 nm (Hochberg et al., 2004). These properties have been used in remote sensing to estimate Chl a by calculating the Normalized Difference Vegetation Index (NDVI). It is calculated using the formula NDVI = (near-infrared - red)/(near-infrared + red) and is based on the reflectance caused by Chl a (Leal et al., 2015). The index is commonly used for terrestrial plants but has also been successfully applied to aquatic organisms. However, the application of NDVI to corals in ecological research has been scarce (Denis et al., 2024; Leal et al., 2015; Naugle et al., 2024; Rocha et al., 2013; Wijgerde et al., 2014). NDVI can be useful for monitoring Chl a content over time and detecting signs of bleaching, which alters the spectral shape due to the loss of pigments. Bleached corals will display overall higher reflectance and pigment loss will alter the spectral shape towards reflectance more similar to carbonate sand (Hochberg et al., 2003; Yamano et al., 2003). Low Chl a content will yield a low NDVI. Hitherto, only two papers have provided a calibration for Chl a vs NDVI in scleractinian corals (Denis et al., 2024; Naugle et al., 2024, see supplementary materials). Therefore, it remains unclear whether this relationship works equally well for all coral species, and users of this protocol are advised to keep this in mind and ideally perform their own calibrations.
The DIVING-PAM-II from Walz GmbH comes with a spectrometer (MINI-SPEC) which allows to easily measure spectral reflectance in laboratory experiments and in the field (Walz, 2023). It obtains spectral measurements in the range of 400 - 850 nm and is suitable for calculating the NDVI. The reflectance head uses a common geometry (45°) to obtain the measurement (Walz, pers. communication), which is favorable for comparisons with previous work (Johnsen, 2016). Furthermore, the spectrometer is submersible, simplifying its use in aquatic environments. In summary, the application of NDVI on corals with the aid of the DIVING-PAM-II has the potential to non-invasively monitor coral health and obtain high-resolution data over time without the disadvantages of conventional methods.
Calibration of the spectrometer
Calibration of the spectrometer
Configure the spectrometer for reflectance measurements by attaching the reflectance head to the spectrometer body (see Walz manual, Seventh Edition, April 2023, Fig. 9, page 21).
Be cautious when attaching the screws as the plastic may weaken over time, potentially causing the head to snap off and leaving the thread inside the spectrometer body.
Attach the white standard (i.e. highly reflective reference material) to the reflectance head and tighten the screw.
Do not touch the white surface, keep away dirt, and store the white standard in a protective casing to maintain its optical properties.
Connect the PAM to the computer, open WinControl, and navigate to the Spectrum tab.
Select Refl. 100% to establish the 100% reflection signal. The calibration result is shown in the graph (The dark current is measured in the factory and stored in the flash memory. It should therefore not need additional calibration. Dark current refers to the electrical noise that is inherent to the spectrometer's detector when no light is present. Its calibration is crucial to account for this baseline noise. If necessary, the dark current can be newly established by fully darkening the probe and pressing Calibr. Dark).
Measure while the white standard is still attached. The result should display a line of 100% reflectance across the entire wavelength range.
The PAM stores the calibrated reference signal and uses it for all subsequent measurements (Walz, personal communication).
We encountered a problem in our DIVING-PAM-II where the reference signal gets lost after turning the device off. The consequence is that measured spectra are not saved upon downloading the data even though the PAM screen displays accurate reflectance. As a workaround, leave the PAM turned on until you finish measuring and calibrate the white standard newly before measuring again. Turning the PAM off after performing the measurements keeps the measured spectra.
Remove the white standard and disconnect the PAM from the laptop to proceed with subsequent measurements.
Measuring coral reflectance
Measuring coral reflectance
Change the operation mode on the PAM to Reflectance. PAM screen path: Main Menu -> Sensor -> Spectrometer -> Operation Mode.
Navigate down from the main screen to reach the Spectrometer window.
Darken the room to reduce interference with other light sources (in the field, interference can be limited by covering the entire measurement area with the probe).
Submerge the spectrometer in the water and verify that no air bubble blocks the detector window or LED light guides. Bubbles may appear on the protective foam.
Position the reflectance head onto the coral so the measurement window covers the measurement area.
Light is only turned on for demonstrational purposes here.
Press SPEC to measure. The spectrometer should flash and the screen display the measured spectrum. Repeat the measurement for two more points of interest on the coral fragment. Data of triplicates will be pooled to represent one fragment.
Note down the respective coral identifier for subsequent data processing.
Processing the data
Processing the data
Connect the PAM to WinControl.
Navigate to the Memory tab.
Download the file that contains your measurements.
In the presence of spectral data, the Spectrum window is active. Navigate to the Spectrum window. The right panel contains a list of your measurements. Selecting a measurement displays the respective spectrum in the graph.
Download the spectral data by right-clicking into the graph and selecting Export all. The downloaded file contains all spectra listed underneath a row containing the wavelengths (nm). The first two entries represent our calibration and white standard measurements.
The following steps should be adjusted as needed for individual requirements.
Remove the time annotations and enter the coral identifiers to the respective entries.
Transpose the data to have the wavelength and coral identifiers as column headers (cal = calibration, ws= white standard).
Save the sheet as a CSV file. The data is now ready for processing in R.
Calculating the NDVI
Calculating the NDVI
Use the following formula to calculate the NDVI for the respective spectra where R670 and R750 represent the average diffuse reflectance in the red and near-infrared wavelength regions:
The wavelength that best represents your data might slightly differ from this example. Most literature refers to using 675 nm for the red region (Leal et al., 2015; Rocha et al., 2013; Wijgerde et al., 2014). However, to get the best fit for your data, choose the Chl a absorption peak (i.e. trough in reflectance) from your spectra, usually between 660 - 680 nm (Galvao et al., 2000).
Protocol references
Brown, B. (1997). Coral bleaching: causes and consequences. Coral Reefs, 16:S129–S138.
Denis, H., Bay, L. K., Mocellin, V. J., Naugle, M. S., Lecellier, G., Purcell, S. W., Berteaux- Lecellier, V., and Howells, E. J. (2024). Thermal tolerance traits of individual corals are widely distributed across the Great Barrier Reef. Proceedings B, 291(2030):20240587.
Douglas, A. (2003). Coral bleaching—how and why? Marine Pollution Bulletin, 46(4):385–392.
Galvao, L., Vitorello, I., and Pizarro, M. (2000). An adequate band positioning to enhance NDVI contrasts among green vegetation, senescent biomass, and tropical soils. International Journal of Remote Sensing, 21(9):1953–1960.
Hochberg, E. J., Atkinson, M. J., and Andr ́efou ̈et, S. (2003). Spectral reflectance of coral reef bottom-types worldwide and implications for coral reef remote sensing. Remote Sensing of Environment, 85(2):159–173.
Hochberg, E. J., Atkinson, M. J., Apprill, A., and Andrefouet, S. (2004). Spectral reflectance of coral. Coral Reefs, 23:84–95.
Johnsen, S. (2016). How to measure color using spectrometers and calibrated photographs. Journal of Experimental Biology, 219(6):772–778.
Leal, M. C., Jesus, B., Ezequiel, J., Calado, R., Rocha, R. J. M., Cartaxana, P., and Serˆodio, J. (2015). Concurrent imaging of chlorophyll fluorescence, chlorophyll a content and green fluorescent proteins-like proteins of symbiotic cnidarians. Marine Ecology, 36(3):572–584.
Naugle, M. S., Denis, H., Mocellin, V. J., Laffy, P. W., Popovic, I., Bay, L. K., and Howells, E. J. (2024). Heat tolerance varies considerably within a reef-building coral species on the Great Barrier Reef. Communications Earth & Environment, 5(1):525.
Rocha, R. J., Calado, R., Cartaxana, P., Furtado, J., and Serˆodio, J. (2013). Photobiology and growth of leather coral Sarcophyton cf. glaucum fragments stocked under low light in a recirculated system. Aquaculture, 414:235–242.
Siebeck, U., Marshall, N., Klu ̈ter, A., and Hoegh-Guldberg, O. (2006). Monitoring coral bleaching using a colour reference card. Coral Reefs, 25:453–460.
Walz (2023). DIVING-PAM-II. Underwater Chlorophyll Fluorometer. Manual. Heinz Walz GmbH, Germany.
Wijgerde, T., van Melis, A., Silva, C. I., Leal, M. C., Vogels, L., Mutter, C., and Osinga, R. (2014). Red light represses the photophysiology of the scleractinian coral Stylophora pistillata. PLoS one, 9(3):e92781.
Yamano, H., Tamura, M., Kunii, Y., and Hidaka, M. (2003). Spectral reflectance as a potential tool for detecting stressed corals. Journal of the Japanese Coral Reef Society, 2003(5):1–10.