The present invention relates generally to a sensor and particularly to a meter that measures chlorophyll-a concentration (Chi) with a much greater accuracy than existing field instruments.

Background

Importance of Phytoplankton

Although invisible to the human eye, microscopic algae free-floating in the ocean (phytoplankton) provide indispensable ecosystem services to the human population. Phytoplankton sustain the oceanic food web and ultimately the marine proteins upon which we rely [6]. Phytoplankton are marine plants that photosynthesise and in doing so produce half of the oxygen we breath [6]. Phytoplankton are also the first crucial component of the ocean biological carbon pump, which transfers carbon from the surface to the ocean interior and ultimately lowers atmospheric C0 ₂ concentration by approximately 50% [ 1 3]. To understand how these critical ecosystem services may be affected by environmental and climatic variability, we need to monitor the current distribution and variability of microscopic algae. Phytoplankton biomass has traditionally been quantified as the concentration of chlorophyll-a (Chi), which is the only photosynthetic pigment ubiquitous in all marine algae. Chi is also used to estimate net primary production (the rate at which solar energy is converted into chemical energy and made available to the ecosystem, [6]) and the eutrophication status of both marine and inland waters.

I Given its importance, phytoplankton biomass has been selected as an Essential Climate Variable by the steering committee for Global Climate Observing System. It is therefore crucial to sustain accurate observations of phytoplankton Chi. However, accurate in-situ observations are laborious and expensive and thus relatively limited in space and time.

Limitations of existing methods and instruments

The traditional method used to quantify Chi relies on collecting discrete water samples, concentrating phytoplankton cells onto filters and extracting the pigments in these cells using a solvent (e.g. acetone). Pigment concentrations are finally quantified using various methods among which the "gold standard" is High Performance Liquid Chromatography (HPLC, [ 1 6]). This traditional method is laborious and relatively expensive. As a consequence, relatively few HPLC-based Chi measurements are available.

The only existing operational method to automatically estimate Chi in the field is fluorometry. Instruments exploiting this method excite chlorophyll-a molecules by illuminating them with blue light and record the red light that these emit as fluorescence [6]. This red fluoresced light (F) is to first order related to Chi [6]. However, F depends strongly on phytoplankton photo-physiological processes, among which the most important is called "non-photochemical quenching" [6]. This process causes F to decrease as the intensity of sunlight increases. As a consequence, for a constant value of Chi, fluorescence can vary by up to a factor of 6 in response to changing solar irradiance (Figure 2; [ 1 8]). In addition, F depends also on the composition of phytoplankton that are measured (e.g. diatoms vs. picoplankton) and the Chl:F ratio can vary by nearly six fold across different phytoplankton cultures [ 14]. In summary, although easy to collect, in-vivo fluorescence data produce highly inaccurate estimates of Chi.

As a consequence, in-situ measurements of Chi remain either laborious, expensive, and scarce (i.e., HPLC) or error-prone (i.e., fluorometry). This data scarcity makes it difficult to assess the health and productivity of marine and freshwater ecosystems (of importance in estimating good environmental status for statutory monitoring) and their role in the global carbon cycle, to manage marine living resources and understand their responses to climate change.

The present invention seeks to address the above limitations and provide an instrument for accurately and automatically measuring Chi in situ both in the field and in the laboratory.

The present invention is based on analyses of data collected under diverse open-ocean regimes using a commercial instrument (ACs) [2, 3, 4, 1 5, 1 7, I ]. Results from the analysis of one of these datasets are presented in Figure I to demonstrate that the current algorithm used to estimate chlorophyll concentration in the field (ALGO I ) is highly accurate (top left plot). This is because ALGO I is based on highly-accurate particulate absorption measurements, that are obtained by subtracting from bulk absorption measurements a baseline signal determined using filtered seawater. This baseline correction greatly improves the stability of the ACs output. The results also demonstrate that the "packaging effect" [ 1 0] does not degrade the accuracy of ALGO I over the wide range of Chi investigated. In addition, Figure I demonstrates that this algorithm can be considerably simplified while at the same time maintaining its accuracy by using bulk absorption measurements (ALG02, top right plot), provided that the drift in the ACs baseline (bottom plot) is accounted for.

Physical principle

The physical principle at the basis of the meter of the present invention is the absorption of light by chlorophyll-a, which is, to first order, directly proportional to Chi. Phytoplankton cells have a broad range of pigments that strongly absorb light in the blue region of the spectrum (i.e. between 400 and 500 nm, Figure 3). However, chlorophyll-a is unique, because it has a distinct peak of absorption in the red spectral region at about 676 nm (Figure 3). By measuring the height of this red absorption peak, the concentration of chlorophyll-a can be accurately determined [ I , 2, 3, 4, 17]. This method was first proposed about twenty years ago [5], but was not further developed, tested and applied, probably because the oceanographic community was not familiar with absorption measurements.

During the last nine years the PI has been deploying a commercially-available in-situ spectrophotometer (WETLabs ACs absorption and attenuation meter) in different oceanic regions and has exploited the above principle to estimate Chi with high accuracy and with unprecedented spatio-temporal resolution (Chi estimates are available at I Hz and typically binned every minute) [2, 17, 15, 3, 4, I ]. Remarkably, results from multiple oceanographic cruises spanning a wide range of environments demonstrated that, by simply multiplying the height of the red peak of particulate absorption spectra by a constant scaling factor, it is possible to estimate Chi with high accuracy. In all these cruises, the Chi estimated from the red peak was on average within ± 10% of the Chi estimated from independent collocated discrete samples analysed using the HPLC method without applying any specific calibration [2, 17, 15, 3, 4, I ] (Fig. 4).

Potential application of the technology to environmental sciences

Meters formed in accordance with the present invention have a wide variety of applications, with a major focus on increasing greatly the accuracy and resolution of Chi measurements.

For example the meter could be deployed in a continuous flow-through mode on all vessels monitoring freshwater and marine ecosystems. Other areas include monitoring Chi in pools, wastewaters, aquaculture, aquaria, and laboratory environments.

Meters could be installed on autonomous platforms including gliders and Biogeochemical-Argo floats. This could be used to greatly expand the database of accurate Chi measurements.

In some embodiments, the system is flexible enough to allow, with only minor modifications (e.g. by replacing LEDs and interference filters), detection of additional absorbing compounds such the phycobilin pigments. Phycobilins are markers for potentially toxic cyanobacterial blooms which are impacting an increasing number of lakes worldwide. Modifications of the meter could thus have additional applications for the management of environmental resources.

Improving the existing method

The method currently used to accurately estimate Chi in the field is rather complex, because it employs measurements of absorption by suspended particles at three wavelengths to estimate the height of the red-absorption peak (Fig. 3). This particulate absorption has been measured using a system that automatically supplies (for 1 0 minutes every hour) 0.2-um filtered seawater to the ACs instrument (e.g., [2, 1 5]). This filtered seawater is used as a baseline that is then subtracted from the bulk absorption signal to obtain, after correcting for the scattering error [2, 1 5], a highly- accurate estimate of the particulate absorption coefficient (a _p ). These data are then used to compute the height of the red absorption peak (Fig. 3). Thus, at present to obtain a highly-accurate Chi estimate from ACs data it is necessary to:

^• have an automated filtration system to compute the baseline correction;

^• have collocated optical scattering measurements to derive the scattering correction;

^• employ three wavelengths (650, 676, 714 nm) to compute the height of the a _p peak in the red spectral region.

Results from analysis of data collected during several Atlantic Meridional Transects and other oceanographic expeditions covering a wide range of Chi, however, suggested that the meter of the present invention could be made significantly simpler by: ( I ) using only two wavelengths (676 and 690 nm); (2) avoiding the scattering correction; and (3) avoiding the baseline correction. The baseline correction is currently needed to improve the stability of the WETLabs ACs instrument. This sensor is notoriously unstable for well-known reasons: it uses a tungsten-halogen light source that drifts significantly over time (Fig. I , bottom plot); it employs moving mechanical parts (i.e., rotating filter wheel) which are sensitive to movements of the overall instrument (e.g., due to ship's motion); it is affected by a strong temperature dependence, which is tentatively compensated using laboratory determined correction coefficients, that can vary over time. Technical approach

In some embodiments the present invention provides an instrument based on light- emitting diodes (LEDs) and electronics and without moving mechanical parts to provide stability. This stability allows for simplification of the accurate determination of Chi in situ by avoiding the need for the baseline correction derived by measuring filtered seawater.

A two-wavelength system is employed. Some of the criteria for meters formed in accordance with the present invention include: required LED power; detector sensitivity; optimal spectral channels; and signal stability.

In one aspect the present invention provides a chlorophyll absorption meter for measuring chlorophyll-a concentration in situ, the meter comprising: an LED for generating a first wavelength in the range of about 670nm to 680nm; an LED for generating a second wavelength in the range of about 685nm to 695nm; a sample chamber for receiving a liquid sample; a sample detector for detecting light at the first and second wavelengths after it is transmitted through a sample; and means for determining light transmittance at the first and second wavelengths.

In one embodiment the optical set up is based on two separate LEDs. The light may be collimated and filtered to select wavelengths at 676 ± 5 or ± I and 690 nm ± 5 or ± I . For example the light may be collimated and filtered to select wavelengths at approximately 676 and 690 nm (Fig. 5) or at exactly 676 nm and 690 nm. A reference detector may be used monitor the input light to the sampling chamber.

Collimated light may interact with a sample (such as phytoplankton cells) in a reflective sample chamber and the transmitted signal may be efficiently captured by the sample detector by means of a light diffuser.

Both detectors and LEDs may have dedicated temperature sensors to monitor and correct for potential temperature-related drifts.

Optical windows may be machined to allow the reflective tube and sleeves to connect with the main meter body. An electronic board may be provided to switch on the LEDs in an alternating mode, to allow the sample detector to separately measure the light transmitted through the sample in each of the two wavelength ranges.

The detectors will also record a signal when the LEDs are off, which will be used to correct for residual ambient stray light and dark currents.

A custom-made software may be provided to easily control and operate the system.

A reference material (e.g., stable coloured glass), with a given transmittance difference between 676 and 690 nm may be provided, which can be used to monitor the stability of the output signal. Alternatively or additionally, the signal from Ch/-free water will be used as a reference.

In some embodiments the differences between the recorded transmittance at 676 nm and that at 690 nm may be used as a proxy to Chi, as shown for ALG02 in Fig. I .

In some embodiments the second wavelength is used as a reference wavelength.

In some embodiments a first LED is used to generate the first wavelength and a second, different LED is used to generate the second wavelength. The first and second LEDs wavelength emissions may be generally centred at the said wavelengths. In some embodiments the present invention uses not only LEDs centred at specific wavelengths, but also with added interference filters between the LEDs and the sample to restrict the band width of each emission. An interference filter may be provided for each LED emitter to restrict the bandwidth.

In some embodiments a miniaturised, water-tight and depth-rated version of the instrument may be provided. A further aspect provides a marine vessel provided with one or more meters as described herein.

A further aspect provides a method for measuring chlorophyll-a concentration in situ, the method comprising: generating a first wavelength in the range of about 670nm to 680nm; generating a second wavelength in the range of about 685nm to 695nm; providing a liquid sample; detecting light at the first and second wavelengths after it is transmitted through the sample; determining light transmittance at the first and second wavelengths; and determining the difference between the light transmittance values at the first and second wavelengths.

Different aspects and embodiments of the invention may be used separately or together.

Further particular and preferred aspects of the present invention are set out in the accompanying independent and dependent claims. Features of the dependent claims may be combined with the features of the independent claims as appropriate, and in combinations other than those explicitly set out in the claims.

The present invention is more particularly shown and described, by way of example, in the accompanying drawings.

The example embodiments are described in sufficient detail to enable those of ordinary skill in the art to embody and implement the systems and processes herein described. It is important to understand that embodiments can be provided in many alternative forms and should not be construed as limited to the examples set forth herein.

Accordingly, while embodiments can be modified in various ways and take on various alternative forms, specific embodiments thereof are shown in the drawings and described in detail below as examples. There is no intent to limit to the particular forms disclosed. On the contrary, all modifications, equivalents, and alternatives falling within the scope of the appended claims should be included. Elements of the example embodiments are consistently denoted by the same reference numerals throughout the drawings and detailed description where appropriate.

Unless otherwise defined, all terms (including technical and scientific terms) used herein are to be interpreted as is customary in the art. It will be further understood that terms in common usage should also be interpreted as is customary in the relevant art and not in an idealised or overly formal sense unless expressly so defined herein.

I I In the description, all orientational terms, such as upper, lower, radially and axial ly, are used in relation to the drawings and should not be interpreted as limiting on the invention. Figure I : Relationships between estimates of Chi from HPLC and in-situ absorption- based algorithms for a dataset collected during an Atlantic Meridional Transect cruise. Top left: standard ACs algorithm based on measurements of the particulate absorption coefficients at three wavelengths (ALGO I ); Top right: algorithm to be applied to the measurements of bulk absorption measurements at two wavelengths (ALG02). Data used for computing ALG02 have been corrected for the significant drift in the baseline of ACs instrument (bottom plot). This baseline was quantified as the difference between the absorption coefficients at 676 and 690 nm determined when 0.2-um filtered water flows through the ACs instrument (<Jf ,[2, 1 5]). Figure 2: Effect of non-photochemical quenching on Chi estimated from in-vivo fluorescence [ 1 8]. Green circles indicate the ratio of Chi to Chi determined from F, while small blue dots rep- resent photosynthetically active radiation.

Figure 3: Typical particulate absorption spectrum (green) measured with and ACs instrument. The peak due to chlorophyll-a at about 676 nm is clearly visible. Dashed lines provide a graphical explanation for the current Chi algorithm that is applied to ACs absorption spectra (ALGO I ).

Figure 4: Relationships between estimates of Chi based on ACs absorption (red circles) and fluorescence (blue squares) data versus HPLC-based total Chi from discrete samples collected during an AMT cruise (modified from [4]). While absorption was transformed into Chi by applying a constant conversion factor taken from the literature, fluorescence data were "calibrated" by means of discrete measurements during the cruise. Thus, even if F is calibrated it is an inaccurate measure of Chi. Thus have additional applications for the management of environmental resources.

Figure 5: Example optical design of a meter, including two different LEDs. The two LEDs provide fixed light emission wavelength sources at specific wavelengths which are restricted to a relatively narrow range.

In this embodiment LEDs centred at specific wavelengths are used and in addition interference filters are used to restrict the band width of each emission.

By restricting the inherently broad spectral emissions of LEDs an accurate estimate of the chlorophyll concentration can be obtained.

Although illustrative embodiments of the invention have been disclosed in detail herein, with reference to the accompanying drawings, it is understood that the invention is not limited to the precise embodiments shown and that various changes and modifications can be effected therein by one skilled in the art without departing from the scope of the invention as defined by the appended claims and their equivalents. REFERENCES

[ I ] Brewin et al., Opt Express (2012)

[2] Dall'Olmo et al., Biogeosciences (2009)

[3] Dall'Olmo et al., Biogeosciences (201 I )

[4] Dall'Olmo et al., Opt Express (2012)

[5] Davis et al., J Geophys Res-oceans ( 1997)

[6] Falkowski & Raven, Aquatic photosynthesis: ( 1997)

[7] Graham & Nimmo Smith, Limnol Oceanogr Methods (2010)

[8] Graham et al., Journal of Geophysical Research-Oceans (2012)

[9] htt|¾;/ wwmse¾uossscLcon^

[ 10] Morel & Bricaud, Deep-sea Res ( 1981 )

[ I I ] Nimmo Smith, Limnol Oceanogr Methods (2008)

[ 12] Nimmo Smith et al., Experiments In Fluids (2002)

[ 1 3] Parekh et al., Geophys Res Lett (2006)

[ 14] Roesler & Barnard, Methods in Oceanography (201 3)

[ 15] Slade et al., J Atmos Ocean Tech (2010)

[ 16] Van Heukelem & Thomas, J Chromatogr A (2001 )

[ 17] Westberry et al., Opt Express (2010)

[ 18] Wollschlager et al., J Exp Mar Biol Ecol (2014)