Devices, systems, and methods for monitoring algal populations

Abstract

Provided herein are devices that can provide continuous, quantitative real-time data on the biomass concentration and the algal health, as indicated by light absorbance and color. Unlike the standard dry biomass or optical density measurements, these biomass sensors can provide continuous measurements available in real-time. The real-time biomass 5 concentration and health data can be used to enable real-time control to optimize productivity of algal ponds. Also provided herein are systems and methods for monitoring algal cultivation.

Description

[0001] Attorney Docket No. 10975-081W01

[0002] Devices, Systems, and Methods for Monitoring Algal Populations

[0003] STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0004] This invention was made with government support under grant number DEEE0009672, awarded by the Department of Energy. The government has certain rights in the invention.

[0005] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims benefit of priority of U. S. Provisional Application No.

[0006] 63 / 722,240, filed November 19, 2024, which is hereby incorporated by reference in its entirety.

[0007] BACKGROUND

[0008] Microalgae have great potential as a feedstock for renewable fuels and bioproducts; however, cun’ent productivity levels are insufficient due to fluctuating environments of light and inorganic carbon, inadequate monitoring of concentration and cellular health, and the inability to control or influence various parameters. Industrial-scale algal productivity is limited by manual sequencing batch operation, with much of each growth phase at high cell density and low growth rate, owing to shading. Whereas the health status is typically monitored subjectively by visual assessment, operators are hampered by the need to make manual, off-line measurements of algal biomass and nutrients, leading to sub-optimal harvest timing and slow responses to pests. Moreover, prediction of large-scale cultivation performance is poor, reducing the rate at which innovations can be screened and implemented.

[0009] Bioprocess optimization needs sensors to enable automated monitoring and control. While sensors for temperature, pH, and light intensity are common, there are no sensors currently available for algal biomass concentration and nutrient sensing. The development of algal biomass sensors offers the opportunity to maximize the productivity of algal cultivations. Attorney Docket No. 10975-081W01

[0010] SUMMARY

[0011] Described herein are devices for monitoring an algal population in water. These devices can comprise (a) a housing; (b) an optical measurement system coupled to the housing and comprising (i) a light source configured to emit light at one or more wavelengths; (ii) optical windows defining a measurement path from the light source and passing through a sample of the water comprising the algal population when the device is deployed in the water; and (iii) a sensor configured to detect light transmitted from the light source along the measurement path; and (c) a controller operatively connected to the light source and the sensor, the controller configured to control light emission from the light source and acquire measurement data from the sensor.

[0012] In some embodiments, the housing comprises a buoyant housing configured to float on a surface of the water. In some embodiments, the housing comprises a body portion with a first measurement leg and a second measurement leg extending downward from the housing. In certain embodiments, the body portion comprises a cylindrical or tapered cylindrical shape.

[0013] In some embodiments, the first measurement leg and the second measurement leg house the optical measurement system. In some embodiments, the first measurement leg and the second measurement leg are adjustable to accommodate deployment depths of the optical measurement system of from 1 inch to 24 inches below the surface of the water, such as from 1 inch to 12 inches below the surface of the water or from 1 inch to 6 inches below the surface of the water.

[0014] In some embodiments, the optical windows comprise a material that provides for optical transmission across ultraviolet, visible, and near-infrared wavelengths. In certain embodiments, the optical windows comprise quartz.

[0015] In some embodiments, the measurement path has a length of from 0.5 cm to 5 cm, such as approximately 1.0 cm.

[0016] In some embodiments, the controller is disposed within the body portion. In some embodiments, the controller is further configured to calculate an absorbance value at one or more wavelengths from the measurement data from the sensor. In some embodiments, the controller is configured to calculate sum-normalized spectral data by dividing each wavelength measurement by total visible spectrum absorbance to enable species identification independent of algal biomass concentration. Attorney Docket No. 10975-081W01

[0017] In some embodiments, the device further comprises a communication module operatively connected to the controller and configured to transmit measureme t data from the sensor, measurement data processed by the controller, or a combination thereof to a data collection platform. In certain embodiments, the communication module comprises a wireless transceiver configured for long-range data transmission.

[0018] In some embodiments, the one or more wavelengths comprise one or more near-infrared wavelengths one or more visible wavelengths, or a combination thereof. In certain embodiments, the one or more wavelengths comprise one or more near-infrared wavelengths and one or more visible wavelengths.

[0019] In some embodiments, the light source comprises one or more light emitting diodes (LEDs). In certain embodiments, the light source comprises a white LED configured to emit broadband illumination across the visible spectrum and an infrared LED configured to emit light at a wavelength of from 800 nm to 950 nm.

[0020] In some embodiments, the sensor comprises a multi-channel spectral sensor. In certain embodiments, the multi-channel spectral sensor is configured to detect light at two or more (e.g., three or more, four or more, five or more, six or more, seven or more, or eight or more) wavelengths of from 400 nm to 700 nm and one or more wavelengths of from 850 nm to 950 nm.

[0021] In some embodiments, the controller is configured to perform a three-phase measurement protocol comprising a background measurement phase performed with the light source disabled, a visible spectrum measurement phase with the white LED activated, and a near-infrared measurement with the infrared LED activated. In certain embodiments, each measurement phase includes a thermal stabilization delay and collection of a plurality of absorbance readings at one or more wavelengths.

[0022] In some embodiments, the device further comprises a battery configured to power the device for at least 12 hours, such as at least 24 hours, at least 48 hours, at least 72 hours, or at least one week. In certain embodiments when the device comprises a battery, the controller can be configured to track battery status and provide low-power warnings. In some embodiments, the device further comprises a photovoltaic cell providing power to one or more components of the device.

[0023] Also described herein are methods for monitoring an algal population in water. These methods can comprise deploying a device described herein in the water (e.g., such that the housing is floating on a surface of the water); emitting light at one or more wavelengths from Attorney Docket No. 10975-081W01

[0024] the light source; detecting light transmitted from the light source along the measurement path passing through sample of the water comprising the algal population using the sensor; and analyzing the detected transmitted light to determine information regarding the algal population in the water.

[0025] In some embodiments, analyzing the detected transmitted light comprises determining absorbance values at one or more wavelengths from the detected transmitted light. In some embodiments, the determining absorbance values comprises performing background subtraction to remove ambient light and sensor dark current effects. In certain embodiments, the performing background subtraction comprises collecting measurements with the light source disabled and subtracting averaged background values from sample measurements.

[0026] In some embodiments, the one or more wavelengths comprise one or more near-infrared wavelengths; and analyzing the detected transmitted light comprises determining absorbance values at one or more near-infrared wavelengths and determining an algal biomass concentration in the water from the absorbance values at one or more near-infrared wavelengths.

[0027] In some embodiments, the one or more wavelengths comprise one or more visible wavelengths; and analyzing the detected transmitted light comprises determining absorbance values at one or more visible wavelengths and determining spectral characteristics from the absorbance values at one or more visible wavelengths. In some embodiments, the method further comprises detecting contamination events by comparing the spectral characteristics to reference spectral patterns for healthy cultures.

[0028] In some embodiments, the method further comprises performing sum normalization by dividing the absorbance value at each of the one or more visible wavelengths by total visible spectrum absorbance to allow for species identification independent of algal biomass concentration. In some embodiments, the method further comprises identifying algal species in the water based on sum-normalized spectral patterns.

[0029] Also described herein are systems for monitoring an algal population in a body of water. These systems can comprise one or more of the devices described herein; a data collection platform configured to receive transmissions from the one or more devices described herein; and a processing system operatively connected to the data collection platform and configured to analyze received measurement data from the one or more sensors, measurement data processed by the one or more controllers, or a combination thereof. Attorney Docket No. 10975-081W01

[0030] In some embodiments, the one or more devices are deployed in an open raceway pond (e.g., to monitor and / or autonomously control commercial algal production). In certain embodiments, the devices are deployed in a body of water (e.g., an open raceway pond) having a surface area of at least 0.25 acres (e.g., at least 0.5 acres, at least 0.75 acres, at least 1 acre, at least 1.5 acres, or more). In other embodiments, the one or more devices are deployed in an natural body of water (e.g., a pond, lake, river, estuary, or ocean to monitor potentially harmful algal blooms).

[0031] In some embodiments, the processing system is configured to analyze received measurement data from the one or more sensors, measurement data processed by the one or more controllers, or a combination thereof to determine algal biomass concentration in the body of water, assess algal health in the body of water, identify an algal species in the body of water, detecting contamination in the body of water, or any combination thereof.

[0032] In some embodiments, the system further comprises an automated control system operatively connected to the processing system and configured to adjust cultivation parameters in the body of water based on the determined algal biomass concentration, the assessed algal health status, the identified algal species, the detected contamination, or a combination there. In certain embodiments, the automated control system is configured to control pond pump operation, harvest timing, or a combination thereof based on the determined algal biomass concentration.

[0033] In some embodiments, the system comprises two or more devices deployed at different locations in the body of water that each transmit measurement data from the one or more sensors, measurement data processed by the one or more controllers, or a combination thereof to the data collection platform, to each other, or a combination thereof. In some of these embodiments, the processing system can be configured to analyze received measurement data and correlate the received measurement data with a location of each of the one or more sensors to obtain information regarding the algal population at different locations in the body of water. In some of these embodiments, the processing system is configured to analyze received measurement data and analyze the received measurement data to obtain a collective understanding of the algal population in the body of water.

[0034] DESCRIPTION OF DRAWINGS

[0035] Figure 1. Schematic of example device described herein.

[0036] Figure 2. Flow chart of sensor measurement cycle and timing. Attorney Docket No. 10975-081W01

[0037] Figure 3. Prototype device housing system showing assembled unit with adjustable measurement legs and underwater optical components with quartz windows for 1.0cm path length measurements during pond deployment trials.

[0038] Figure 4A. Calibration curves for Nannochloropsis oceanica showing the effect of infrared LI ID brightness (PWM duty cycle) on linearity and sensitivity. Duty cycle of 15% provides optimal performance withf?2= 0.99, slope = 0.203 ± 0.003, avoiding saturation while maintaining high signal-to-noise ratio.

[0039] Figure 4B. Calibration curve for Monoraphidium minutum using 890 nm near-infrared (NIR) measurement. Strong linearity (i?2= 0.998) with slope = 0.293 demonstrates consistent sensor performance across diverse species while revealing species-specific optical characteristics.

[0040] Figure 40. Calibration results for Phaeodactylum tricornutum showing linear correlation (R2= 0.953) despite a limited concentration range. Temporal instability after 15 minutes likely reflects species-specific settling behavior characteristic of diatom morphology.

[0041] Figure 4D. Tetraselmis suecica calibration using a conventional spectrophotometric measurement protocol. Limited reliability of conventional OD measurements for this species required alternative validation approaches while maintaining linear correlation (?2= 0.918).

[0042] Figure 4E. Improved Tetraselmis suecica calibration using dilution -based OD estimation method. Enhanced correlation (R2= 0.961) demonstrates sensor reliability when reference measurement challenges are addressed through alternative approaches.

[0043] Figure 4F. Comparison of conventional OD calibration curve (solid line, circles) and improved dilution-based estimation (dashed line, triangles). Lower OD points from original readings demonstrate initial measurement accuracy before temporal degradation effects.

[0044] Figure 5. Universal calibration curve combining all tested algae species using 890 nm NIR measurement. Consistent slopes (0.29-0.35) across diverse species support universal applicability while species-specific calibration provides optimal accuracy.

[0045] Figure 6A, Individual channel correlations with optical density for Nannochloropsis oceanica. All visible wavelength channels demonstrate strong linear relationships (R2> 0.99) with optical density, supporting the summation approach for alternative biomass estimation.

[0046] Figure 6B. Optical density calibration using visible spectrum summation method across multiple LED brightness levels for Nannochloropsis oceanica. Independence from brightness variations (duty cycles 100-200) demonstrates measurement robustness compared to individual channel approaches Attorney Docket No. 10975-081W01

[0047] Figure 6C. Cross-species comparison of visible spectrum summation method showing maintained linearity, but increased slope variability compared to NIR approach, reflecting pigment composition differences between species.

[0048] Figure 7A. Raw absorbance spectra for Namochloropsis oceanica showing discrete 8channei measurements across varying biomass concentrations. Consistent spectral signatures with concentration-dependent scaling support both quantitative analysis and species identification capabilities.

[0049] Figure 7B. Comparative raw absorbance spectra across six algae species showing distinct spectral signatures. Species-specific patterns enable identification capabilities while common features (chlorophyll peaks at 445 nm and 680 nm) confirm measurement validity across diverse organisms.

[0050] Figure 7C. Overlaid raw absorbance spectra for all measured species highlighting species specific variations in pigment composition. Universal chlorophyll features validate measurement consistency while accessory pigment differences enable species discrimination and physiological monitoring.

[0051] Figure 8A. Comparison of raw absorbance (left) and sum-normalized absorbance (right) for Namochloropsis oceanica. Normalization eliminates concentration dependence while preserving species-specific spectral patterns, enabling comparison across different biomass levels.

[0052] Figure 8B. Sum-normalized absorbance spectra for all measured species showed enhanced pigment composition pattern discrimination. Normalization reveals subtle interspecies differences in accessory pigment ratios while maintaining species-specific identification signatures.

[0053] Figure 8C. Overlaid sum-normalized spectra for ail species demonstrating species¬ specific fingerprints for identification and monitoring applications. Clear discrimination between algae types supports automated classification, while standard features validate measurement consistency.

[0054] Figure 9A, Gaussian-distributed continuous spectra for multiple species generated from 8-channel discrete measurements. Smooth spectral curves enable comparison with traditional spectrophotometry while preserving measurement accuracy and species¬ specific features. Attorney Docket No. 10975-081W01

[0055] Figure 9B. Overlaid Gaussian -reconstructed spectra for all species demonstrating maintained species-specific fingerprints. Reconstruction preserves essential identification features while enabling comparison with traditional spectrophotometric databases.

[0056] Figure 9C. Sum-normalized Gaussian-distributed spectra for all measured species. Traditional spectral presentation format facilitates comparison with existing databases while maintaining species discrimination capability and concentration independence Figure 9D. Comparison of sensor-generated Gaussian spectra with literature reference data. Good agreement in major spectral features validates the measurement approach while highlighting resolution limitations compared to high-resolution spectrophotometry.

[0057] Figure 10A. Complete spectral evolution during bleaching experiment showing progressive loss of chlorophyll absorption features. Temporal progression demonstrates quantitative assessment of pigment degradation under controlled bleaching conditions.

[0058] Figure 10B. Sum-normalized spectral analysis of bleaching progression revealing early-stage changes in pigment ratios before visible degradation. Enhanced sensitivity supports early detection of culture stress and contamination events.

[0059] Figure 10C. Temporal analysis of bleaching experiment with blue indicating healthy culture states and red representing advanced degradation. Progressive transition reveals sensor sensitivity to physiological changes with detection capability at 0.02% concentration levels.

[0060] Figure 10D. NIR channel response during bleaching experiments showing approximately 15% signal reduction despite complete photosynthetic system destruction, demonstrating continued biomass sensitivity independent of pigment content.

[0061] Figure 11A. Principal component analysis of sum-normalized spectral data showing species clustering and discrimination. Clear separation between major algae groups demonstrates automated classification capability for operational monitoring applications.

[0062] Figure 11B. Principal component analysis using spectral channel ratios showing enhanced species discrimination and reduced environmental variability. Ratio-based approach provides superior clustering performance for field deployment applications.

[0063] Figure 11C. Enhanced PCA visualization showing distinct groupings for experimental categories. Clear separation between bleach experiments, calibration Attorney Docket No. 10975-081W01

[0064] samples, and strain variations demonstrates automated classification capability for culture monitoring applicati on s.

[0065] Figure 12A. Complete dataset from five sequential NarinocMoropsis oceanica cultivation runs showing consistent sensor performance across independent experiments. Clear growth curves validate measurement reliability under dynamic cultivation conditions.

[0066] Figure 12B. First 100 hours of cultivation showing detailed growth dynamics with color-coded light intensity overlay. Strong correlation between growth patterns and illumination cycles validates sensor sensitivity to cultivation conditions and biological responses.

[0067] Figure 12C. Comparison of sensor OD measurements (NIR and visible summation) with benchtop spectrophotometer references throughout cultivation period. Excellent correlation validates sensor accuracy under dynamic cultivation conditions.

[0068] Figure 12F). Overall correlation between sensor and reference OD measurements from pond deployments. Strong linear relationship (i?2= 0.987, slope = 1.042) confirms sensor accuracy across operational range with minimal systematic bias

[0069] Figure 12E. Residual analysis showing normally distributed measurement errors with mean = 0. Absence of systematic trends validates linear calibration model and confirms measurement consistency across operational range.

[0070] Figure 12F. Correlation between NIR absorbance and Total Organic Carbon measurements providing independent biomass validation. Strong correlation (R2= 0.954) confirms optical measurement accuracy for biomass assessment independent of reference method limitations.

[0071] Figure 13A. Raw absorbance spectra measured at varying optical densities during cultivation across five pond runs. Consistent spectral characteristics validate measurement repeatability and biological relevance under dynamic cultivation conditions.

[0072] Figure 13B. Sum-normalized spectral evolution for single pond cultivation showing stable spectral patterns throughout cultivation indicating healthy culture conditions and absence of contamination or stress events.

[0073] Figure 13C. Sum-normalized spectral comparison across all pond runs at Sequivalent optical densities. Good consistency between independent experiments validates normalization approach and demonstrates biological measurement repeatability Attorney Docket No. 10975-081W01

[0074] Figure 14. Ambient light interference effects on individual spectral channels showing wavelength-dependent sensitivity. Blue and violet channels exhibit higher interference susceptibility while longer wavelengths demonstrate enhanced stability for outdoor deployment.

[0075] Figure 15A. Multi-day field deployment showing systematic diurnal paterns in all spectral channels correlating with solar intensity cycles. Predictable interference patterns enable algorithmic correction for enhanced outdoor measurement accuracy.

[0076] Figure 15B. Correlation between ambient light intensity and NIR signal variations showing greater influence from voltage fluctuations than environmental light interference, informing correction algorithm development priorities.

[0077] Figure 15C. Voltage patterns during 24-hour deployment cycles showing systematic variations likely reflecting wall outlet fluctuations. Predictable patterns enable development of voltage-based correction algorithms for enhanced measurement stability.

[0078] Figure 151). Relationship between LED brightness levels and measurement accuracy showing optimal performance at moderate brightness settings. Results validate current brightness selection while guiding power management strategies for maximizing accuracy and battery life.

[0079] Figure 16. Schematic illustrating components of an example algal cultivation device design.

[0080] Figure 17A. Photograph showing a side view of an example device described herein. The sensor features a self-floating design that encloses the battery and electronics in a buoyant chamber, while keeping optical windows for measurement. The housing is fully sealed but still allows NIR and visible light to pass through for absorbance measurements via transmittance through quartz windows. An adjustable anchoring system accommodates different water levels in outdoor ponds. The current battery lasts more than 10 days of autonomous operation, and the wireless range is approximately 80-100 meters in typical field conditions.

[0081] Figure 17B. Photograph showing an internal field of the housing. The custom PCB features modular, removable ports for all sensor components and the ESP32-S2 microcontroller, facilitating rapid battery swaps and component replacement during field deployments. The design incorporates expansion capability for secondary battery charging or solar panel integration. This modular architecture enables system upgrades and repairs without complete unit replacement, reducing long-term operational costs. Attorney Docket No. 10975-081W01

[0082] Figure 17C. Photograph showing the LED board and AS7341 sensor. The illumination system includes a custom LED board (left) featuring a sunlike white LED for broad spectral coverage and an NIR LED for biomass measurement. The AS7341 11-channel color sensor (right) captures absorbance data across 8 visible wavelength bands plus near-infrared. This sensor-illumination pairing facilitates simultaneous measurement of biomass concentration and spectral signatures for health assessment. 3D printed mounts allow for flexibility with future board designs.

[0083] Figure 17D. Photograph showing example electronics for an example algal sensor. Figure 17E. Photographs showing a top view (left) and a bottom view (right) of the casing and electronics for an example device described herein.

[0084] Figure 18A. Plot showing a linear trend between total sensor signal (red, green, and blue channels, x-axis) and absorbance (optical density, OD) measurement (y-axis) at 750 nm values in a benchtop spectrophotometer.

[0085] Figure 18B. Plot showing a linear trend between sensor IR channel signal and absorbance (optical density, OD) measurement at 750 nm (y-axis) from a benchtop spectrophotometer.

[0086] Figure 19. Plot showing a linear trend between individual sensor signal(s) (green, x-axis) and spectrophotometer absorbance (optical density, OD) measurement at 750 nm (y-axis) in the light and dark.

[0087] Figure 20. Plot showing the results from algal biomass concentration data from a 5-day cultivation of Nannochloropsis oceanica microalgae in a 100-L open raceway pond.

[0088] Figure 21. Plot showing the results from a channel comparison, concentration range in ambient light.

[0089] Figure 22. Plot showing normalized results for colors, concentration range in ambient light.

[0090] Figure 23. Plot showing results for concentration and light range in ambient light. Figure 24. Plot showing the detection of different algal species.

[0091] Figure 25. Plot showing the detection of algal cultivation contamination.

[0092] Figure 26. Illustration of an example automated harvest system including the devices described herein.

[0093] Figure 27. Flow diagram illustrating elements and workflow used for automated harvest and control Attorney Docket No. 10975-081W01

[0094] Figure 28. Plot of measured biomass over time during the example automated harvest. The set points were 80, 60, 115, and 110 mg / L in chronological order. The last two set points, 110 and 115 mg / L, indicate the current capabilities after minor adjustments to the algorithm, which increase consistency and accuracy.

[0095] Figure 29. Plot illustrating biomass tracking through NIR absorbance measurement. Figure 30. Plot illustrating biomass tracking through absorbance at 555 nm.

[0096] Figure 31. Plot overlaying absorbance at 555 nm with total organic carbon (ppm). Figure 32. Plot demonstrating this use of a sensor to monitor total organic carbon in a progression through three ponds over 12 days.

[0097] Figure 33A. Schematic illustration of aspects of a device and system described herein.

[0098] Figures 33B. Illustration of aspects of a device and system described herein.

[0099] DETAILED DESCRIPTION

[0100] Definitions

[0101] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Methods and materials are described herein for use in the present invention; other, suitable methods and materials known in the art can also be used. The materials, methods, and examples are illustrative only and not intended to be limiting. All publications, patent applications, patents, sequences, database entries, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control.

[0102] As used herein, “algae” refers to photosynthetic or phototrophic microorganisms capable of converting light energy into chemical energy, regardless of taxonomic classification. The term encompasses eukaryotic algae (including but not limited to green algae, red algae, diatoms, and other microalgal lineages) as well as prokaryotic organisms that perform oxygenic or anoxygenic photosynthesis. This includes cyanobacteria and other photosynthetic bacteria such as purple non-sulfur bacteria, green sulfur bacteria, and green non-sulfur bacteria. The term may further include naturally occurring, genetically modified, or synthetically engineered organisms that exhibit phototrophic growth or utilize light-driven energy capture pathways. Examples of phototrophic microorganisms can include, but are not limited to, Spirulina maximum, Spirulina platensis, Dunaliella salina, Botrycoccus braunii, Attorney Docket No. 10975-081W01

[0103] Chlorella vulgaris, Chlorella pyrenoidosa, Sereiwstrum capricomutum, Scenedesmus auadricauda, Porphyridium cruentum, Scenedesmus acutus, Dunaliella sp., Scenedesmus obliquus, Anabaenopsis sp., Aulosira sp., Cylindrospermum sp., Synechococcus sp., Synechocystis sp., Tolypothrix sp., or a combination thereof

[0104] As used herein, "light" generally refers to sunlight but can be solar or from artificial sources including incandescent lights, LEDs, fiber optics directing / transmitting light from a natural or artificial light source, or metal halide, neon, halogen and fluorescent lights. The light can include light of a wavelength that is photosynthetically active in the phototrophic microorganism, meaning light that can be utilized by the microorganism to grow and / or produce chemical products of interest.

[0105] Ranges of values defined herein include all values within the range as well as all subranges within the range. For example, if the range is defined as an integer from 0 to 10, the range encompasses all integers within the range and any and all subranges within the range, e.g., 1-10, 1-6, 2-8, 3-7, 3-9, etc.

[0106] Devices

[0107] The following description sets forth exemplary aspects of the present disclosure. It should be recognized, however, that such description is not intended as a limitation on the scope of the present disclosure. Rather, the description also encompasses combinations and modifications to those exemplary aspects described herein.

[0108] Described herein are field-deployable optical devices configured for monitoring algal populations, including algal populations in cultivation systems and / or algal populations in natural bodies of water (e.g., lakes, ponds, rivers, estuaries, oceans, etc.). The devices may provide continuous, quantitative real-time data regarding algal biomass density and algal health status through measurement of light absorbance and spectral characteristics across multiple wavelengths. The devices may operate autonomously in outdoor cultivation environments while maintaining measurement accuracy comparable to laboratory instrumentation.

[0109] The devices may utilize multi-wavelength optical measurements to assess both algal biomass concentration and cellular health parameters. Near-infrared wavelength measurements may provide algal biomass quantification that remains substantially independent of pigment composition variations between different algal species. Visible Attorney Docket No. 10975-081W01

[0110] spectrum measurements across discrete wavelength channels may enable spectral analysis for species identification, contamination detection, and physiological monitoring applications.

[0111] Referring now to Figure 33A, described herein are devices (100) for monitoring an algal population in water (102). These devices can comprise (a) a housing (104); (b) an optical measurement system (106) coupled to the housing and comprising (i) a light source (108) configured to emit light at one or more wavelengths; (ii) optical windows (110) defining a measurement path (112) from the light source and passing through a sample of the water comprising the algal population when the device is deployed in the water; and (iii) a sensor (114) configured to detect light transmitted from the light source along the measurement path; and (c) a controller (116) operatively connected to the light source and the sensor, the controller configured to control light emission from the light source and acquire measurement data from the sensor.

[0112] The device may incorporate any suitable housing design that allows deployment directly within a body of water (e.g., an algal cultivation pond or tank). The housing may maintain the optical measurement system at a consistent depth while accommodating water level variations during cultivation cycles. The housing may provide waterproof enclosure for electronic components while allowing optical access to the algal culture through designated measurement zones (e.g., optical windows operatively positioned with respect to the light source and sensor).

[0113] Referring to Figure 33B, in some embodiments, the housing (104) can comprise a body portion (120) with a first measurement leg (122) and a second measurement leg (124) extending downward from the housing. The housing can have any suitable shape. In certain embodiments, the body portion comprises a cylindrical or tapered cylindrical shape. In certain embodiments, the body portion comprises a substantially bowl-shaped body. These shapes can provide structural stability and uniform weight distribution when the device is deployed in water. In particular, these shapes may offer enhanced stability compared to alternative geometric configurations while minimizing drag forces during water circulation in cultivation systems. These shapes may also distribute forces evenly around the perimeter, reducing rotational movement.

[0114] In some embodiments, the device includes an attachment point (e.g., affixed to the housing) that allows the device to be tethered or anchored within a body of water. In some embodiments, the housing can be submersible in water. Attorney Docket No. 10975-081W01

[0115] In some embodiments, the housing can be non-buoyant. In other embodiments, the housing can comprise a buoyant housing. The buoyant housing may include airtight, adjustable casing that enables the device to float with buoyancy control. The airtight casing may prevent water ingress while maintaining internal air volume for buoyancy adjustment. The adjustable casing may allow modification of the internal air volume to achieve desired floating characteristics or accommodate different sensor configurations. Buoyancy control may enable fine-tuning of the floating position to maintain optimal measurement depth under varying water density conditions.

[0116] The measurement legs may provide structural support for the optical components while creating a defined measurement zone within the algal culture. In some embodiments, the first measurement leg and the second measurement leg house the optical measurement system.

[0117] In some embodiments, the first measurement leg and the second measurement leg are adjustable to accommodate deployment depths of the optical measurement system of from 1 inch to 24 inches below the surface of the water, such as from 1 inch to 12 inches below the surface of the water or from 1 inch to 6 inches below the surface of the water. The adjustable depth capability may facilitate optimization of measurement positioning for different cultivation systems, algal species, or operational requirements. Shallow deployment depths may provide measurements representative of surface conditions, while deeper positioning may sample algal culture characteristics at greater depths where light penetration and mixing conditions may differ.

[0118] The optical measurement system may employ controlled illumination sources and multi-channel spectral detection to acquire absorbance measurements across selected wavelength ranges. Background subtraction techniques may compensate for ambient light interference and sensor drift effects. Data processing algorithms may convert raw spectral measurements into biomass concentration values and spectral characteristics suitable for automated analysis and decision-making systems.

[0119] The device includes optical operatively positioned with respect to the light source and sensor so as to provide a path for light to pass from the light source through a sample of the water comprising the algal population to the sensor. In some embodiments, the optical windows comprise a material that provides for optical transmission across ultraviolet, visible, and near-infrared wavelengths. In certain embodiments, the optical windows comprise quartz. In some embodiments, the optical windows can be different for each wavelength Attorney Docket No. 10975-081W01

[0120] region (e.g., they can include more than one material that has high optical transmission for each wavelength region).

[0121] The measurement path extends between the light source and the sensor, creating a defined optical path length for consistent absorbance calculations In some embodiments, the measurement path has a length of from 0.5 cm to 5 cm, such as approximately 1.0 cm.

[0122] In some embodiments, the controller is disposed within the body portion of the housing. The controller may be operatively connected to the light source and the multi-channel spectral sensor to provide centralized control and data processing capabilities for the optical measurement system. The controller may serve as the primary processing unit that coordinates all measurement operations, timing sequences, and data analysis functions within the device. The controller may be configured to control light emission parameters, acquire spectral measurements from the multi-channel spectral sensor, and calculate absorbance values from the spectral measurements through integrated computational algorithms.

[0123] In some embodiments, the controller may be configured to control light emission by managing the timing, intensity, and duration of illumination from both a white LED and an infrared LED component. The controller may implement pulse width modulation control to regulate LED brightness levels and maintain consistent illumination conditions across measurement cycles. The controller may coordinate sequential activation of different light sources to enable background measurements, visible spectrum measurements, and near-infrared measurements as part of a comprehensive measurement protocol.

[0124] The controller may acquire spectral measurements by interfacing with the sensor through digital communication protocols such as I2C or SPI interfaces. The controller may configure sensor parameters including integration times, gain settings, and channel selection to optimize measurement conditions for different algal culture densities and environmental conditions. The controller may collect raw spectral data from each wavelength channel and store the measurements in memory for subsequent processing and analysis.

[0125] The controller may calculate absorbance values from the spectral measurements by implementing Beer-Lambert law calculations that convert raw transmission measurements into absorbance data. The controller may perform background subtraction to remove ambient light interference and sensor dark current effects from the raw measurements. The controller may apply calibration coefficients and correction factors to compensate for sensor response variations and environmental influences on measurement accuracy. Attorney Docket No. 10975-081W01

[0126] The controller may perform a three-phase measurement protocol comprising background measurement with light sources disabled, visible spectrum measurement with the white LED activated, and near-infrared measurement with the infrared LED activated. The three-phase protocol may enable comprehensive spectral characterization while providing background correction capabilities for enhanced measurement accuracy. Each measurement phase may include thermal stabilization delays and multiple readings for statistical analysis to improve measurement precision and reliability.

[0127] The controller may comprise, for example an ESP32-S2 microcontroller that provides integrated wireless communication capabilities and enhanced processing power for complex spectral analysis algorithms. The ESP32-S2 microcontroller may offer dual-core processing architecture that enables simultaneous sensor control and data processing operations. The ESP32-S2 microcontroller may include integrated Wi-Fi capabilities that support wireless data transmission without requiring separate communication modules.

[0128] Alternative controller configurations may utilize an Arduino microcontroller as an alternative processing platform that provides simplified programming interfaces and extensive community support for sensor applications. Arduino microcontrollers may offer various form factors and processing capabilities suitable for different deployment requirements and complexity levels. Arduino-compatible boards may provide standardized interfaces for sensor integration and modular expansion capabilities for enhanced functionality.

[0129] The controller may operate with an ESP32-S2 Feather board with integrated battery management systems for improved power management during autonomous operation. The ESP32-S2 Feather board may provide integrated lithium battery charging circuits, voltage regulation, and power monitoring capabilities that enhance operational reliability. The integrated battery management system may enable automatic power optimization and low- power sleep modes that extend deployment duration while maintaining measurement capabilities.

[0130] The controller may be configured to calculate sum-normalized spectral data by dividing each wavelength measurement by total visible spectrum absorbance to enable species identification independent of biomass concentration. Sum normalization may eliminate concentration-dependent scaling effects while preserving species-specific spectral patterns that enable taxonomic discrimination. The controller may implement normalization algorithms that process raw spectral data in real-time to generate normalized spectral Attorney Docket No. 10975-081W01

[0131] signatures suitable for automated species identification and contamination detection applications

[0132] The controller may implement calibration corrections that compensate for sensor response variations and optical path characteristics to ensure accurate absorbance calculations across all wavelength channels. Calibration algorithms may apply predetermined correction factors that account for manufacturing variations, wavelength-dependent sensitivity, and optical path geometry to generate standardized absorbance measurements. The calibration process may operate automatically during each measurement cycle to maintain measurement accuracy without requiring manual intervention.

[0133] Biomass concentration determination algorithms may process the near-infrared absorbance measurements through correlation calculations that convert optical measurements into quantitative biomass concentration values. The biomass determination process may utilize predetermined correlation coefficients that relate near-infrared light attenuation to algal cell density while accounting for optical path length and sensor calibration factors. The biomass calculation algorithms may generate real-time concentration estimates that enable immediate assessment of cultivation status and growth trends.

[0134] Spectral analysis algorithms may process the visible wavelength absorbance measurements through normalization calculations that generate species-specific spectral signatures suitable for automated identification and health assessment. The spectral processing may implement sum normalization algorithms that divide each wavelength measurement by total visible spectrum absorbance to eliminate concentration-dependent scaling effects while preserving species-specific absorption patterns. The normalized spectral data may enable species identification and contamination detection independent of biomass concentration variations.

[0135] Pattern recognition algorithms may analyze the processed spectral data through comparison with reference databases or trained classification models to enable automated species identification and contamination detection. The pattern recognition process may utilize statistical analysis or machine learning algorithms to classify spectral measurements and identify deviations from normal culture conditions. The automated classification capability may provide immediate identification of species changes or contamination events without requiring human interpretation.

[0136] The controller may coordinate data packaging and transmission preparation by organizing the processed biomass concentration values, spectral characteristics, and sensor Attorney Docket No. 10975-081W01

[0137] status information into standardized data formats suitable for transmission. The data packaging process may include measurement timestamps, quality indicators, and sensor diagnostic information that enable comprehensive remote monitoring and performance assessment. The data organization may optimize transmission efficiency while maintaining the information content required for remote analysis applications.

[0138] In some embodiments, the device further comprises a communication module (130) operatively connected to the controller and configured to transmit measurement data from the sensor, measurement data processed by the controller, or a combination thereof to a data collection platform (132). Wireless communication capabilities may enable remote data transmission and monitoring without requiring physical access to the device during operation. The wireless communication module may be configured to transmit measurement data to remote monitoring stations, data logging systems, or automated control platforms without requiring physical access to the device. The wireless communication module may support various communication protocols suitable for outdoor deployment environments where reliable data transmission over extended distances may be required. In certain embodiments, the communication module comprises a wireless transceiver configured for long-range data transmission.

[0139] In some embodiments, the communication module may comprise a LoRa transceiver configured for long-range data transmission over distances exceeding 100 meters in typical field conditions. LoRa transceivers may provide low-power, long-range communication capabilities that enable data transmission across large cultivation facilities without requiring intermediate relay stations. The LoRa protocol may offer enhanced penetration through obstacles and interference sources commonly encountered in outdoor cultivation environments while maintaining reliable data transmission.

[0140] Alternative wireless communication configurations may utilize an HC-05 Bluetooth module or other Bluetooth modules for shorter-range communication applications where monitoring stations may be positioned in proximity to the device deployment locations. Bluetooth modules may provide simplified pairing and connection procedures while offering adequate transmission range for smaller cultivation systems or laboratory applications.

[0141] Bluetooth communication may enable direct connection to mobile devices or portable monitoring equipment for field data collection and analysis.

[0142] The wireless communication system may incorporate a radio system for data transmission that allows the device to operate at commercial distances from monitoring Attorney Docket No. 10975-081W01

[0143] stations, other devices, or a data collection platform. This can allow a deployed device to communicate with an external data collection platform. In cases where a plurality of devices are deployed, the devices can also communicate with one another, if desired for system operation Radio communication systems may provide enhanced transmission range and reliability compared to standard wireless protocols while maintaining compatibility with existing communication infrastructure. Radio systems may enable deployment across large commercial cultivation facilities where device locations may be separated by significant distances from central monitoring stations.

[0144] In some embodiments, the one or more wavelengths comprise one or more near-infrared wavelengths one or more visible wavelengths, or a combination thereof. In certain embodiments, the one or more wavelengths comprise one or more near-infrared wavelengths and one or more visible wavelengths.

[0145] In some embodiments, the light source comprises one or more LEDs. In certain embodiments, the light source comprises a white LED configured to emit broadband illumination across the visible spectrum and an infrared LED configured to emit light at a wavelength of from 800 nm to 950 nm. Alternative illumination configurations may utilize multiple LEDs emitting light within different regions of the electromagnetic spectrum, such as violet, blue, cyan, green, yellow, orange, and red LEDs for targeted spectral illumination. The discrete wavelength LEDs may provide enhanced spectral selectivity compared to broadband white LED illumination. Violet LEDs may provide illumination at approximately 405 nm, blue LEDs at approximately 450 nm, cyan LEDs at approximately 490 nm, green LEDs at approximately 525 nm, yellow LEDs at approximately 590 nm, orange LEDs at approximately 610 nm, and red LEDs at approximately 660 nm. The discrete wavelength approach may enable optimized illumination for specific spectral channels while reducing spectral overlap between adjacent measurement channels.

[0146] In some embodiments, the sensor comprises a multi-channel spectral sensor. In certain embodiments, the multi-channel spectral sensor is configured to detect light at two or more (e.g., three or more, four or more, five or more, six or more, seven or more, or eight or more) wavelengths of from 400 nm to 700 nm and one or more wavelengths of from 850 nm to 950 nm.

[0147] The multi-channel spectral sensor may comprise an AS7341 11 -channel spectral color sensor that provides eight visible wavelength channels plus additional near-infrared and wide spectrum channels. The AS7341 sensor may offer wavelength channels at approximately 415 Attorney Docket No. 10975-081W01

[0148] nm (FT), 445 nm (F2), 480 nm (F3), 515 nm (F4), 555 nm (F5), 590 nm (F6), 630 nm (F7), 680 nm (F8), and 890 nm (NIR). The 11 -channel configuration may provide comprehensive spectral coverage while maintaining compact sensor dimensions suitable for portable deployment applications.

[0149] Enhanced spectral sensor configurations may utilize an AS7343 sensor providing 12 channels plus UV capability as an enhanced spectral sensor option. The AS7343 sensor may extend spectral coverage into the ultraviolet region while maintaining the visible and near-infrared channels of the AS7341 configuration. The additional UV channel may enable detection of UV-absorbing compounds or cellular components that may provide additional information regarding algal physiology or contamination status.

[0150] Higher resolution spectral analysis may be achieved through AS7262 / AS7263 spectral triad sensors providing 18 channels total coverage. The AS7262 sensor may provide six visible wavelength channels spanning 430 nm to 670 nm, while the AS7263 sensor may provide six near-infrared channels spanning 680 nm to 940 nm. The combined 18-channel configuration may offer enhanced spectral resolution compared to single-sensor approaches while maintaining discrete channel measurement simplicity.

[0151] Alternative high-resolution configurations may incorporate mini-spectrometers such as a Hamamatsu C12880MA providing 340 nm to 850 nm coverage with 288 pixels for higher resolution spectral analysis. The mini-spectrometer approach may provide continuous spectral coverage with enhanced resolution compared to discrete channel sensors. The 288- pixel array may enable detailed spectral characterization while maintaining compact sensor dimensions suitable for field deployment applications.

[0152] The multi-channel spectral sensor may operate with integration times calculated as (ATIME + 1) x (ASTEP + 1) x 2.78μs with ATIME of 50 units and ASTEP of 999 steps. The integration time calculation may determine the total exposure duration for each spectral measurement cycle. The ATIME parameter may control the number of integration cycles, while the ASTEP parameter may determine the duration of each integration step. The calculated integration time may provide optimal signal accumulation while maintaining measurement speed suitable for continuous monitoring applications.

[0153] In some embodiments, the controller is configured to perform a three-phase measurement protocol comprising a background measurement phase performed with the light source disabled, a visible spectrum measurement phase with the white LED activated, and a near-infrared measurement with the infrared LED activated. In certain embodiments, each Attorney Docket No. 10975-081W01

[0154] measurement phase includes a thermal stabilization delay and collection of a plurality of absorbance readings at one or more discrete wavelengths.

[0155] In some embodiments, the device further comprises a battery (128) configured to power the device for at least 12 hours, such as at least 24 hours, at least 48 hours, at least 72 hours, or at least one week. In certain embodiments when the device comprises a battery, the controller can be configured to track battery status and provide low-power warnings. In some embodiments, the device further comprises a photovoltaic cell providing power to one or more components of the device (e.g., to directly power the controller, the light source, and / or the sensor, and / or to recharge the battery ).

[0156] The battery power system may comprise lithium-ion battery packs that provide high energy density and stable voltage output throughout the discharge cycle. The battery power system may be sized to support continuous device operation, wireless data transmission, and processing requirements while maintaining compact dimensions suitable for the housing design.

[0157] The battery power system may incorporate power management circuits that optimize energy consumption through intelligent power scheduling and low-power sleep modes during inactive periods. Power management may include automatic shutdown of non-essential components during measurement intervals and dynamic adjustment of processing speeds based on operational requirements. The power management system may extend operational duration while maintaining measurement accuracy and data transmission capabilities.

[0158] A voltage monitoring system (and / or the controller) may be configured to track battery status and provide low-power warnings to enable proactive maintenance scheduling before complete power depletion occurs. The voltage monitoring system may continuously measure battery voltage levels and calculate remaining operational time based on current power consumption patterns. The voltage monitoring system may transmit battery status information along with measurement data to enable remote assessment of device operational status and maintenance requirements.

[0159] The voltage monitoring system may implement programmable voltage thresholds that trigger low-power warnings at predetermined battery levels to ensure adequate time for maintenance or battery replacement. The voltage monitoring system may provide multiple warning levels including early warning notifications and critical power alerts that enable appropriate response timing. The voltage monitoring system may automatically reduce power Attorney Docket No. 10975-081W01

[0160] consumption or modify operational parameters when low battery conditions are detected to extend operational duration until maintenance can be performed

[0161] Methods

[0162] Also described herein are methods for monitoring an algal population in water. These methods can comprise deploying a device described herein in the water (e.g., such that the housing is floating on a surface of the water); emitting light at one or more wavelengths from the light source; detecting light transmitted from the light source along the measurement path passing through sample of the water comprising the algal population using the sensor; and analyzing the detected transmitted light to determine information regarding the algal population in the water.

[0163] The deployment process may involve positioning a device within the algal culture at a predetermined depth that provides representative sampling of cultivation conditions while maintaining optical measurement accuracy. The deployment may utilize housing configurations that maintain consistent measurement positioning relative to the water surface while accommodating variations in water level during cultivation cycles.

[0164] The deployment process may involve securing the device at a fixed depth within the algal culture through anchoring systems that prevent movement while maintaining optical alignment. Fixed depth deployment may provide enhanced measurement repeatability by eliminating variations in measurement position caused by water level changes or surface disturbances. The anchoring system may accommodate different pond depths and bottom conditions while maintaining the optical measurement path at the desired depth below the water surface.

[0165] Alternative deployment configurations may utilize floating devices that maintain measurement positioning relative to the water surface through housing designs. Floating deployment may automatically accommodate water level variations while maintaining consistent measurement depth relative to the surface conditions where algal growth typically occurs. The floating deployment approach may provide simplified installation procedures while maintaining measurement accuracy across varying water levels during cultivation operations.

[0166] The deployment process may include calibration procedures that establish baseline measurement conditions and reference standards for the specific algal culture and cultivation environment. Calibration may involve acquiring reference measurements from culture Attorney Docket No. 10975-081W01

[0167] samples with known algal biomass concentrations to establish correlation relationships between optical measurements and actual algal biomass content. The calibration process may account for species-specific optical properties while establishing measurement accuracy standards for the deployed device.

[0168] The method may comprise emitting light from the light source through the algal culture at multiple wavelengths including near-infrared wavelengths to provide illumination across the spectral range required for comprehensive optical analysis. The light emission process may involve sequential activation of different illumination sources to enable discrete wavelength measurements while preventing spectral interference between different measurement channels. The multi-wavelength illumination approach may provide for simultaneous assessment of algal biomass concentration and spectral characteristics through a single optical measurement system.

[0169] In some embodiments, the light emission process may involve activating a white LED to provide broadband illumination across the visible spectrum from approximately 400 nm to 700 nm during visible spectrum measurement phases. The white LED activation may provide continuous spectral coverage that enables simultaneous illumination of multiple discrete wavelength channels measured by the multi-channel spectral sensor. The broadband illumination approach may ensure adequate signal levels across all visible wavelength channels while maintaining consistent illumination conditions throughout the measurement cycle.

[0170] In some embodiments, the light emission process may involve activating an infrared LED to provide targeted illumination at approximately 890 nm during near-infrared measurement phases. The infrared LED activation may provide specific wavelength illumination that facilitates algal biomass quantification measurements while minimizing interference from pigment absorption effects that may vary between different algal species. The near-infrared illumination may operate at controlled intensity levels that optimize signal-to-noise ratios while preventing sensor saturation across the expected range of algal biomass concentrations.

[0171] The light emission process may implement pulse width modulation control to regulate illumination intensity and maintain consistent light output characteristics throughout extended measurement cycles. Pulse width modulation may enable precise control of LED brightness levels while providing power efficiency for autonomous operation. 'The Attorney Docket No. 10975-081W01

[0172] modulation control may account for LED thermal characteristics and aging effects to maintain consistent illumination conditions across extended deployment periods.

[0173] The light emission timing may include thermal stabilization periods that allow LED junction temperatures to reach steady-state conditions before measurement acquisition begins. Thermal stabilization may prevent measurement variations caused by LED brightness changes during warm-up periods or thermal drift effects during continuous operation. The stabilization timing may be optimized based on LED specifications and ambient temperature conditions to ensure consistent illumination characteristics.

[0174] The light detection process may implement integration time control that optimizes signal accumulation for accurate measurements while maintaining measurement speed suitable for continuous monitoring. Integration time optimization may balance signal-to-noise ratios against measurement frequency requirements to provide optimal detection performance. The integration time may be adjustable to accommodate different algal culture densities, ambient light conditions, or measurement accuracy requirements through programmable sensor parameters.

[0175] The light detection process may include multiple measurement acquisitions during each detection cycle to enable statistical analysis and improve measurement precision.

[0176] Multiple measurements may be averaged to reduce random noise effects while providing quantitative assessment of measurement uncertainty. The statistical analysis approach may identify outlier measurements that may indicate temporary interference or environmental disturbances while maintaining measurement reliability.

[0177] In some embodiments, analyzing the detected transmitted light comprises determining absorbance values at one or more wavelengths from the detected transmitted light. In some embodiments, the determining absorbance values comprises performing background subtraction to remove ambient light and sensor dark current effects. In certain embodiments, the performing background subtraction comprises collecting measurements with the light source disabled and subtracting averaged background values from sample measurements.

[0178] In some embodiments, the one or more wavelengths comprise one or more near-infrared wavelengths; and analyzing the detected transmitted light comprises determining absorbance values at one or more near-infrared wavelengths and determining an algal biomass concentration in the water from the absorbance values at one or more near-infrared wavelengths. Attorney Docket No. 10975-081W01

[0179] In some embodiments, the one or more wavelengths comprise one or more visible wavelengths; and analyzing the detected transmitted light comprises determining absorbance values at one or more visible wavelengths and determining spectral characteristics from the absorbance values at one or more visible wavelengths. In some embodiments, the method further comprises detecting contamination events by comparing the spectral characteristics to reference spectral patterns for healthy cultures.

[0180] In some embodiments, the method further comprises performing sum normalization by dividing the absorbance value at each of the one or more visible wavelengths by total visible spectrum absorbance to allow for species identification independent of algal biomass concentration. In some embodiments, the method further comprises identifying algal species in the water based on sum-normalized spectral patterns.

[0181] Systems

[0182] Referring again to Figure 33A, described herein are systems for monitoring an algal population in a body of water. These systems can comprise one or more of the devices described herein (100); a data collection platform (132) configured to receive transmissions from the one or more devices described herein; and a processing system operatively (134) connected to the data collection platform and configured to analyze received measurement data from the one or more sensors, measurement data processed by the one or more controllers, or a combination thereof.

[0183] In some embodiments, the one or more devices are deployed in an open raceway pond (e.g., to monitor and / or autonomously control commercial algal production). In certain embodiments, the devices are deployed in a body of water (e.g., an open raceway pond) having a surface area of at least 0.25 acres (e.g., at least 0.5 acres, at least 0.75 acres, at least 1 acre, at least 1.5 acres, or more). In other embodiments, the one or more devices are deployed in an natural body of water (e.g., a pond, lake, river, estuary, or ocean to monitor potentially harmful algal blooms).

[0184] In some embodiments, the system comprises two or more devices deployed at different locations in the body of water that each transmit measurement data from the one or more sensors, measurement data processed by the one or more controllers, or a combination thereof to the data collection platform, to each other, or a combination thereof. In some of these embodiments, the processing system can be configured to analyze received measurement data and correlate the received measurement data with a location of each of the Attorney Docket No. 10975-081W01

[0185] one or more sensors to obtain information regarding the algal population at different locations in the body of water. In some of these embodiments, the processing system is configured to analyze received measurement data and analyze the received measurement data to obtain a collective understanding of the algal population in the body of water.

[0186] In some embodiments, the processing system is configured to analyze received measurement data from the one or more sensors, measurement data processed by the one or more controllers, or a combination thereof to determine algal biomass concentration in the body of water, assess algal health in the body of water, identify an algal species in the body of water, detecting contamination in the body of water, or any combination thereof.

[0187] In some embodiments, the system further comprises an automated control system (136) operatively connected to the processing system and configured to adjust cultivation parameters in the body of water based on the determined algal biomass concentration, the assessed algal health status, the identified algal species, the detected contamination, or a combination there. In certain embodiments, the automated control system is configured to control pond pump operation, harvest timing, or a combination thereof based on the determined algal biomass concentration.

[0188] The automated control system may comprise pond pump control, harvest timing optimization, and nutrient addition scheduling based on real-time algal biomass measurements and spectral analysis from the distributed device network. The control system integration may enable responsive cultivation management that maintains optimal conditions across large cultivation facilities while preventing culture losses through early detection and intervention.

[0189] The automated control system may control pond pump operation based on spatial analysis of algal biomass distribution and mixing efficiency determined from measurements obtained from one or more devices present in the body of water. Pump control algorithms may adjust circulation rates, flow patterns, or mixing schedules to optimize biomass distribution and prevent settling or stratification in large cultivation ponds. The pump control system may respond to real-time measurements from the device network to maintain uniform cultivation conditions across large pond areas while minimizing energy consumption and operational costs.

[0190] The automated control system may optimize harvest timing based on biomass concentration measurements and spectral analysis from multiple sensor locations to maximize harvest efficiency and product quality. Harvest timing optimization may consider spatial Attorney Docket No. 10975-081W01

[0191] variations in biomass density, species composition, and cellular health status to determine optimal harvest conditions across large cultivation areas. The harvest control system may coordinate partial harvesting of high-density zones while allowing continued growth in areas with lower biomass concentrations.

[0192] The system may be used for monitoring of commercial ponds spanning 1 acre in surface area or greater through deployment of multiple devices positioned to provide representative sampling across large cultivation areas. The device network may be configured with appropriate device spacing and positioning to capture spatial variations in cultivation conditions while maintaining communication connectivity with the central data collection platform. The large-scale monitoring capability may support commercial cultivation operations where pond sizes exceed the monitoring range of individual devices.

[0193] The device network deployment for large commercial ponds may utilize strategic device positioning that accounts for circulation patterns, mixing zones, and potential areas of stratification or settling within the cultivation system. Device placement optimization may ensure representative sampling of cultivation conditions while minimizing the number of devices required for comprehensive monitoring. The deployment strategy may consider factors including communication range limitations, maintenance accessibility, and potential interference from cultivation equipment or infrastructure.

[0194] The processing system may utilize total organic carbon measurements as an alternative reference standard for biomass quantification that provides enhanced accuracy compared to conventional optical density approaches. Total organic carbon analysis may provide direct quantification of biological material that avoids optical artifacts and temporal instability issues that may affect spectrophotometric reference methods. The total organic carbon reference standard may enable more accurate calibration of optical measurements while providing independent validation of sensor performance across different algal species and cultivation conditions.

[0195] The total organic carbon reference approach may be implemented through periodic sampling and laboratory analysis that establishes correlation relationships between optical measurements and actual biomass content. The total organic carbon analysis may provide ground truth measurements that validate the accuracy of near-infrared absorbance measurements while enabling species-specific calibration optimization. The reference standard approach may support quality assurance protocols that maintain measurement accuracy across extended deployment periods and diverse cultivation conditions. Attorney Docket No. 10975-081W01

[0196] The processing system may implement calibration algorithms that utilize total organic carbon reference measurements to optimize biomass quantification accuracy across different algal species and cultivation conditions. The calibration approach may account for species¬ specific variations in optical properties while maintaining universal applicability of the near-infrared measurement approach. The total organic carbon calibration may provide enhanced accuracy for biomass quantification compared to conventional optical density reference methods while supporting automated calibration updates based on periodic reference sampling.

[0197] The system architecture may provide scalable monitoring capabilities that accommodate cultivation facilities ranging from research-scale operations to large commercial production systems through modular device deployment and processing system configuration. The scalable architecture may enable expansion of monitoring coverage through addition of devices while maintaining centralized data collection and analysis capabilities. The modular system design may support diverse deployment scenarios while maintaining standardized measurement protocols and analytical capabilities across different facility sizes and operational requirements.

[0198] An automated control system may be operatively connected to the processing system and may be configured to adjust cultivation parameters based on the determined biomass concentration and algal health status received from the distributed sensor network. The automated control system may provide responsive cultivation management that maintains optimal growth conditions while preventing culture losses through real-time parameter adjustments based on continuous spectral monitoring data. The automated control system may integrate with existing cultivation infrastructure to enable autonomous operation that reduces manual intervention requirements while improving cultivation efficiency and product quality.

[0199] The automated control system may receive processed spectral data from the processing system including biomass concentration measurements, species identification results, and health status assessments from multiple sensor locations across the cultivation facility. The automated control system may analyze the received data to identify cultivation conditions that require parameter adjustments, including biomass density variations, contamination events, nutrient limitation indicators, or physiological stress conditions detected through spectral analysis. The data integration capability may enable the automated Attorney Docket No. 10975-081W01

[0200] control system to make informed decisions regarding cultivation parameter modifications based on comprehensive biological monitoring information.

[0201] The automated control system may be configured to adjust cultivation parameters including circulation rates, mixing schedules, nutrient addition timing, harvest operations, and environmental controls based on real-time biological feedback from the sensor network. Parameter adjustment capabilities may enable optimization of cultivation conditions in response to changing biological requirements, environmental conditions, or operational objectives. The automated control system may implement control algorithms that balance multiple cultivation parameters simultaneously to maintain optimal growth conditions while preventing adverse interactions between different control actions.

[0202] The automated control system may implement feedback control loops that continuously monitor the effects of parameter adjustments on biomass concentration and algal health status through ongoing spectral measurements from the sensor network.

[0203] Feedback control may enable the automated control system to evaluate the effectiveness of control actions and make additional adjustments as needed to achieve desired cultivation outcomes. The feedback control approach may prevent overcorrection or inappropriate control responses while enabling fine-tuning of cultivation parameters based on observed biological responses.

[0204] The automated control system may be configured to control pond pump operation based on real-time biomass measurements and spectral analysis from multiple sensor locations within the cultivation system. Pond pump control may optimize circulation patterns, mixing efficiency, and biomass distribution across large cultivation areas while preventing settling, stratification, or dead zones that may compromise cultivation performance. The pump control system may adjust circulation rates, flow directions, or operational schedules based on spatial analysis of biomass distribution and mixing effectiveness determined from the distributed sensor measurements.

[0205] The automated control system may implement pump control algorithms that respond to biomass concentration gradients detected across multiple sensor locations by adjusting circulation patterns to promote uniform biomass distribution. The pump control system may increase circulation rates in areas where sensor measurements indicate poor mixing or biomass settling while reducing circulation in areas with adequate mixing to optimize energy consumption. The spatial analysis capability may enable targeted circulation adjustments that Attorney Docket No. 10975-081W01

[0206] address localized mixing problems without affecting areas of the cultivation system that maintain appropriate circulation conditions.

[0207] The automated control system may control pump operation schedules based on diurnal variations in algal physiology detected through temporal analysis of spectral measurements from the sensor network. Pump scheduling may account for daily cycles in algal growth rates, photosynthetic activity, and cellular metabolism that affect optimal circulation requirements throughout day and night periods. The automated control system may increase circulation during periods of high photosynthetic activity while reducing circulation during periods of reduced metabolic activity to optimize energy consumption while maintaining appropriate mixing conditions.

[0208] The automated control system may implement pump control strategies that respond to contamination events detected through spectral analysis by increasing circulation rates to prevent contamination spread or implementing targeted circulation patterns that isolate contaminated areas. Contamination response protocols may include emergency circulation adjustments that minimize contamination impact while enabling targeted treatment or removal of contaminated culture volumes. The automated control system may coordinate pump operations with other control actions including harvest operations or chemical treatment applications to address contamination events effectively.

[0209] The automated control system may be configured to control harvest timing based on real-time algal biomass measurements and spectral analysis that indicate optimal harvest conditions across, for example, one or more ponds present in an algal cultivation facility. Harvest timing control may maximize biomass yield while maintaining culture health and product quality through precise timing of harvest operations based on continuous biological monitoring. The automated control system may coordinate harvest operations across separate ponds to optimize overall facility productivity while maintaining seed culture for continued cultivation cycles.

[0210] The automated control system may implement harvest timing algorithms that analyze algal biomass concentration trends from multiple device locations to predict optimal harvest windows that maximize algal biomass yield while preventing culture degradation. Harvest timing optimization may consider growth rate trends, biomass density projections, and spectral indicators of cellular health to determine the optimal balance between biomass accumulation and culture stability. The predictive harvest timing capability may enable Attorney Docket No. 10975-081W01

[0211] proactive harvest scheduling that prevents culture crashes while maximizing productivity across cultivation cycles.

[0212] The automated control system may control partial harvest operations based on spatial analysis of biomass distribution that enables selective harvesting of high-density areas while allowing continued growth in areas with lower biomass concentrations. Partial harvest control may optimize overall facility productivity by harvesting mature culture areas while maintaining active cultivation in areas that have not reached optimal harvest density. The spatial harvest control capability may enable continuous production operations that maintain consistent product output while maximizing biomass utilization efficiency.

[0213] The automated control system may coordinate harvest timing with species identification results from spectral analysis to ensure harvest operations target the desired algal species while avoiding harvest of contaminated or undesirable culture areas. Speciesspecific harvest control may maintain product purity while preventing contamination of harvested biomass with unwanted organisms or degraded culture material. The automated control system may implement harvest protocols that verify species identity through spectral analysis before initiating harvest operations in specific cultivation zones.

[0214] The automated control system may implement harvest timing strategies that respond to physiological stress indicators detected through spectral analysis by initiating emergency harvest operations to recover biomass before culture viability declines. Emergency harvest protocols may prevent total culture losses when spectral monitoring indicates developing problems that may compromise culture health if harvest operations are delayed. The automated control system may balance emergency harvest timing with biomass yield optimization to maximize biomass recovery while preventing culture degradation.

[0215] The automated control system may provide integration capabilities with existing cultivation facility infrastructure including pumps, valves, harvest equipment, and environmental control systems to enable comprehensive cultivation management through a unified control platform. Infrastructure integration may utilize standard industrial communication protocols and control interfaces to coordinate automated control actions across diverse equipment types and manufacturers. The integration capability may enable retrofitting of existing cultivation facilities with automated control capabilities without requiring complete equipment replacement.

[0216] The automated control system may implement safety protocols and operational limits that prevent control actions that could damage cultivation equipment or compromise culture Attorney Docket No. 10975-081W01

[0217] viability through excessive parameter adjustments. Safety protocols may include maximum circulation rates, harvest volume limits, and parameter change rate restrictions that prevent harmful control actions while enabling effective cultivation optimization. The automated control system may include manual override capabilities that enable operator intervention when automated control actions require modification or emergency shutdown procedures become necessary.

[0218] The automated control system may provide data logging and performance monitoring capabilities that track the effectiveness of control actions and enable continuous improvement of control algorithms based on operational experience. Performance monitoring may analyze the relationship between control actions and cultivation outcomes to optimize control parameters and improve system performance over time. The data logging capability may support regulatory compliance requirements while providing operational data for facility optimization and troubleshooting applications.

[0219] The automated control system may implement machine learning algorithms that adapt control strategies based on historical cultivation data and observed responses to control actions across different seasonal conditions, algal species, and operational scenarios. Machine learning integration may enable continuous improvement of control performance while adapting to changing cultivation conditions or facility modifications. The adaptive control capability may optimize cultivation parameters for specific facility characteristics, environmental conditions, and production objectives through automated learning from operational experience.

[0220] In some embodiments, the system may operate continuously for extended periods without manual intervention, providing temporal resolution that exceeds conventional sampling approaches. Real-time biomass and health monitoring capabilities may facilitate responsive cultivation management, including automated harvest timing and early detection of culture problems before visible manifestation occurs.

[0221] By way of non-limiting illustration, examples of certain embodiments of the present disclosure are given below.

[0222] EXAMPLES

[0223] Example 1. Development of a Device for Real-Time Assessment of Algal Biomass and Health.

[0224] Overview Attorney Docket No. 10975-081W01

[0225] Large-scale microalgae cultivation requires continuous monitoring of both biomass and health, yet existing real-time sensors are often prohibitively expensive or reliant on through-flow designs that demand constant maintenance. This Example introduces an affordable, in-situ optical device capable of measuring algal biomass concentration and health without the need for pumping systems. The device employs near-infrared light to measure algal concentration while using absorbance at eight visible wavelengths for algal health. Operating in various lighting conditions, floating on the water surface, and correcting for background interference, it alleviates many constraints of conventional methods. Data from the sensor are wirelessly transmitted to local receivers and integrated into cloud-based databases, enabling remote access and continuous tracking of cultivation parameters. Because it supplies real-time information on growth patterns, effects of nutrient supply, and environmental influences such as pH or time of day variations, the system can detect small changes that can give new insights. When combined with nutrient sensors, the data provide comprehensive oversight of cultivation conditions, facilitating efficient harvesting strategies. By offering immediate feedback on biomass concentration and algal health, this optical sensing system facilitates cultivation adjustments to optimize growth and reduce resource waste. Its low cost, portability, and straightforward deployment make it an attractive alternative for large-scale operations, especially compared with limited manual measurements. With its robust design, this solution significantly advances cost-effective algae cultivation for commercial and research applications.

[0226] Introduction

[0227] The gulf between laboratory success and commercial viability in microalgae cultivation stems from a simple problem: knowing what is happening in the culture at any given moment. In small laboratory flasks, researchers can visually assess culture health, take frequent samples, and maintain near-perfect environmental control. At commercial scales spanning hectares of cultivation area, these simple approaches become impossible.

[0228] Considering the fundamental questions operators must answer continuously: What is the current biomass concentration? Is the culture growing at expected rates? Are the cells physiologically healthy or stressed? Has contamination by competing organisms begun? Are valuable metabolites accumulating? In most industrial bioprocesses, sophisticated sensor networks provide real-time answers. In microalgae cultivation, operators often rely on methods little changed from decades past — manual sampling, microscopic examination, and intuition gained through experience. Attorney Docket No. 10975-081W01

[0229] This technological gap has profound consequences. Contamination by predatory protozoa or competing microorganisms can destroy entire cultures within days, yet early detection requires careful microscopic analysis rarely performed frequently enough.

[0230] Harvest timing dramatically impacts yield and product quality --too early sacrifices productivity, too late results in culture decline and cell death. Stress-induced metabolite production, such as lipid accumulation for biofuels, requires precise navigation between productive stress and irreversible damage.

[0231] The economic implications compound these technical challenges. Manual sampling and laboratory analysis require skilled personnel and introduce delays between observation and intervention that may be costly. Without adequate monitoring, yields remain unpredictable and losses can be frequent, preventing the cost reductions necessary to justify better monitoring infrastructure.

[0232] A device incorporating an optical measurement system was purpose-built for the unique demands of monitoring large-scale microalgae cultivation. Rather than adapting expensive laboratory instruments or accepting the limitations of basic industrial sensors, this work created a new monitoring paradigm aligned with both the technical requirements and economic constraints of commercial production.

[0233] These devices (as well as systems incorporating these devices) can accomplish one or more of the following:

[0234] Enable continuous, autonomous monitoring through a floating platform that operates independently for long periods (e.g., weeks), collecting data around the clock without operator intervention;

[0235] Provide a comprehensive biological assessment by simultaneously measuring biomass concentration through near-infrared transmission and physiological status through visible spectrum analysis;

[0236] Achieve laboratory-quality measurements in field conditions, maintaining accuracy within 10% of standard spectrophotometric methods despite environmental challenges;

[0237] Detect contamination and stress events through spectral pattern recognition, identifying problems before they become visually apparent or cause irreversible damage;

[0238] Demonstrate practical deployment in working cultivation facilities, validating performance under real-world conditions, including temperature extremes, biofouling, and mechanical stress; Attorney Docket No. 10975-081W01

[0239] Establish economic viability through low-cost components and minimal maintenance requirements, enabling deployment of multiple units for comprehensive spatial coverage;

[0240] Create scalable architecture supporting wireless sensor networks that can expand with facility size, providing the data density necessary for precise process control; and / or Provide a foundation for automatic harvesting by leveraging the precise information integrated with a tuned model, maximizing productivity and yield.

[0241] This Example describes devices and systems from conceptual design through field validation, encompassing hardware engineering, algorithm development, and practical deployment protocols. The work demonstrates that sophisticated optical monitoring need not require an exorbitant price. Careful engineering and modern components enable laborator -quality measurements at costs that are compatible with commercial cultivation.

[0242] The technical approach leverages advances in multi-channel spectroscopic sensors, specifically the AS7341 device validated for analytical applications. The system provides comprehensive monitoring through a single integrated platform by combining visible spectrum analysis for biological assessment with near-infrared measurement at 890 nm for biomass quantification. The 890 nm wavelength selection, based on optical properties of microalgae particles, reduces solar interference issues that would affect measurements at the commonly used wavelength of 750 nm while maintaining measurement sensitivity. In addition, 890 nm provides a more universal biomass estimation regardless of strain.

[0243] Critical scope boundaries focus the research on practical deployment needs. 'The system targets biomass and health monitoring rather than complete compositional analysis, though spectral data provides indicators of pigment ratios and general metabolic state. Real-time data transmission enables process monitoring but stops short of implementing closed-loop control, establishing the sensing foundation for future automation development.

[0244] Field validation in operational cultivation systems ensures practical relevance beyond laboratory demonstrations. Extended deployment trials address real-world challenges, including biofouling, calibration drift, and weather exposure — factors that often doom academically developed sensors when faced with commercial reality.

[0245] Focusing on deployment practicality throughout development, this Example bridges the persistent gap between sensor innovation and industry adoption. Attorney Docket No. 10975-081W01

[0246] The implications extend beyond immediate monitoring improvements. Accessible, biological data transforms cultivation from empirical reliance to quantitative science, enabling optimization approaches impossible with current sparse measurements. As the microalgae industry matures toward its multi-billion-dollar potential, sophisticated monitoring will become beneficial and essential for competitive operation. This Example provides both immediate solutions and a technological foundation for the future.

[0247] Methods

[0248] Herein, we describe a comprehensive methodology for developing, validating, and deploying an autonomous device for real-time microalgae cultivation monitoring. The research approach integrates hardware engineering, advanced data analysis, and systematic experimental validation to create a practical monitoring solution that bridges laboratory capabilities with field deployment requirements. The methodology is organized into three interconnected components: device development and hardware architecture, data processing and analytical methods, and experimental design with validation protocols.

[0249] Overall System Design. The device design prioritizes autonomous operation, measurement accuracy, and deployment practicality for extended field use in challenging cultivation environments. The floating platform architecture enables continuous monitoring without infrastructure requirements while maintaining laboratory-quality measurements through careful optical design and environmental compensation strategies. The circular buoyant design ensures stable positioning at the water surface while allowing free water circulation through the measurement volume, providing represen ative sampling during pond circulation cycles without disturbing cultivation operations.

[0250] Electronic Components and Processing Architecture. The device utilizes an ESP32-C3 Mini development board as the central processing unit, selected for its integrated wireless capabilities, low power consumption characteristics, and robust 12 -bit analog-to- digital conversion. The microcontroller operates at a reduced clock frequency of 80MHz to minimize power consumption during active measurement cycles while maintaining sufficient computational capability for real-time statistical processing and wireless data transmission. Deep sleep functionality enables power conservation between measurement cycles, with automatic wake-up and sensor reinitialization protocols ensuring seamless autonomous operation over extended deployment periods. Future system iterations will transition to an ESP32-S2 Feather board to leverage integrated battery management systems and enhanced Attorney Docket No. 10975-081W01

[0251] low-power optimization features, eliminating external power management circuitry requirements while improving overall system reliability and reducing component count.

[0252] Power management integration incorporates a precision voltage monitoring system that tracks board operation voltage status through a voltage divider circuit of lOOkQ and 47k£l resistors connected to GPIO 0. The empirically determined calibration factor of 0.911 combined with 12-bit ADC resolution provides 0.8 mV measurement precision, enabling real-time battery monitoring and predictive maintenance scheduling based on discharge characteristics. Primary power is supplied by a 100050 mAh, 3.7 V LiPo battery providing extended autonomous operation spanning multiple weeks under typical measurement schedules. At the same time, the voltage monitoring system enables automatic low-power warnings and graceful shutdown procedures to prevent data loss during power depletion events

[0253] ESP32-C3 Pin Assignments and Component Connections

[0254] ESP32-C3 Pin Component Connection Function

[0255] GPIO 0 Voltage Divider Signal Input Battery Voltage Monitoring GPIO 2 White LED PWM Output Visible Spectrum Illumination GPIO 3 IR LED (890 nm) PWM Output NIR Biomass Measurement GPIO 4 RYLR998 LoRa UART RX LoRa Data Reception GPIO 5 RYLR998 LoRa UART TX LoRa Data Transmission GPIO 8 AS7341 Sensor l2C SDA Spectral Data Communication GPIO 9 AS7341 Sensor PC SCL Spectral Clock Signal 3.3V AS7341, LoRa, LEDs Power Supply Component Power

[0256]

[0257] GND All Components Ground Common Ground Reference

[0258] Long-range data transmission utilizes a RYLR998 LoRa transceiver operating in the ISM band with UART communication at 115200baud, enabling reliable data transmission over distances exceeding 100 meters under typical outdoor conditions. The transceiver connects to ESP32 GPIO pins 4 (RX) and 5 (TX), providing robust wireless communication capabilities for remote monitoring applications. Power management integration allows the transceiver to enter sleep mode via AT+MODE commands during measurement intervals, significantly reducing power consumption while maintaining communication capability for scheduled data transmission cycles. This wireless architecture eliminates infrastructure requirements while providing sufficient range for pond-scale monitoring applications across typical cultivation facility layouts. Attorney Docket No. 10975-081W01

[0259] Spectrophotometry for Algae Cultivation and Biomass Determination. When examining light attenuation at specific wavelengths, concentration demonstrates a linear correlation with apparent absorbance, though in reality, tliis represents a correlation with total light attenuation. Since pure absorbance varies significantly between algae strains due to pigment differences, it is not a reliable universal indicator of biomass concentration. The industry standard for biomass estimation is optical density (OD), which provides measurement of total light attenuation (reduced transmittance) through the sample, dominated by scattering and reflection rather than pure absorption. Whether absorbance or transmittance measurements are employed, linear relationships exist only within specific concentration ranges. At high concentrations, nonlinear regions emerge due to sensor saturation or physical limitations in light transmission through dense suspensions.

[0260] Different optical configurations can be employed for biomass measurement, with the primary approaches being transmission, backscatter, and side-scatter. While 180-degree backscatter measurements provide the most extensive linear range and 90-degree side-scatter offers intermediate performance, transmission measurements remain the preferred method for quantitative biomass determination despite having the most limited linear range.

[0261] The preference for transmission measurements stems from their fundamental advantages in practical applications. Transmission geometry maintains constant optical path length, enabling simplified calculations and consistent calibration across different measurement conditions. Transmission measurements exhibit the least measurement error compared to scattering-based approaches and maintain direct relationship to Beer-Lambert principles. This reduced sensitivity to particle size variations and environmental factors makes transmission measurements particularly valuable for field deployment in variable conditions.

[0262] The wavelength 750 nm has become the standard for measuring optical density of microalgae because it avoids absorption from biological components that interact with visible light. At this wavelength, light attenuation derives from physical biomass presence through scattering and reflection rather than molecular absorption. This approach allows relatively universal biomass estimation across different strains and cultivation conditions. However, strain-specific calibration against dry weight measurements remains necessary due to variations in cell size and morphology.

[0263] However, a significant limitation of 750 nm measurements is ambient light interference at tliis wavelength, requiring dark measurement environments for accurate Attorney Docket No. 10975-081W01

[0264] results. This Example investigates 890 nm as an improved alternative for optical density measurements. The 890 nm wavelength was specifically chosen because it represents the optimal balance of several critical factors: it corresponds to peak responsivity of the chosen photodetector while avoiding the interfering effects of water absorption that occur above 1000 nm.

[0265] Utilizing near-infrared at 890 nm provides practical advantages, including reduced solar interference that enables measurements in ambient light conditions, and even greater detachment from pigment-based variations compared to 750 nm. These characteristics make the 890 nm approach particularly suitable for field deployment while maintaining the fundamental linear relationship between optical density and biomass concentration.

[0266] Absorbance spectroscopy also facilitates precise, non -destructive analysis of pigment content and composition which are indicators of microalgal physiological state, photosynthetic efficiency, and stress conditions. Chlorophyll represents a fundamental target for algae spectroscopic analysis, exhibiting characteristic absorption peaks that provide insights into culture health and photosynthetic capacity.

[0267] Clear absorption peaks appear around 425 nm and 675 nm, corresponding to chlorophyll a absorption maxima. Notably, minimal chlorophyll absorption occurs at 750 nm, supporting its use for biomass measurements independent of pigment variations. Chlorophyll absorbance at approximately 680 nm directly relates to photosynthetic capacity and cellular health, while carotenoids and accessory pigments (absorbing broadly between 400-550 nm) can indicate stress responses or adaptation states.

[0268] Finally, looking at the absorption spectra of various microalgae reveals differences. Not all microalgae utilize chlorophyll for photosynthesis. In addition, each species may have other factors contributing to its absorbance profile. Even within species, the spectra can vary greatly, as seen below in the Synechococcus red and blue-green wavelengths.

[0269] This spectral diversity stems from evolutionary adaptations that minimize competition for light resources between species. Such species-specific signatures enable identification and monitoring of culture purity, early contamination detection, and optimization of cultivation conditions for target organisms. Furthermore, absorbancebased methods like pulse-amplitude modulation fluorescence can indirectly measure photosystem II efficiency, providing rapid insights into nutrient limitation, photoinhibition, or oxidative stress. Attorney Docket No. 10975-081W01

[0270] Spectroscopic analysis provides sensitive indicators of culture health and physiological stress. Stress-induced changes in pigment composition, particularly alterations in carotenoid-to-chlorophyll ratios, can be detected through systematic spectral analysis before visible manifestation of culture problems.

[0271] The dramatic spectral changes observed when photosynthetic systems shut down or cells die demonstrate the sensitivity of spectroscopic monitoring. Absorption peaks disappear, leaving only baseline light attenuation from cell structures. The gradual spectral decline across wavelengths partially reflects increased light scattering at shorter (blue) wavelengths. Significantly, while no absorption peaks exist at 750 nm, bleached cells still affect optical density measurements, though this effect is reduced at 890 nm, providing additional justification for the longer wavelength approach.

[0272] Spectroscopic analysis extends beyond simple biomass quantification to provide critical insights into culture quality, physiological state, and production optimization.

[0273] Stress-induced changes in algae metabolism often manifest as detectable alterations in pigment composition before conventional indicators show problems. For instance, lipid accumulation during nitrogen deprivation stress correlates with increased carotenoid-to- chlorophyll ratios that can be monitored through systematic spectral analysis. Depending on operational objectives, this capability enables dynamic optimization of cultivation conditions to maximize either biomass production or high-value compound synthesis.

[0274] Similarly, contamination by competing microorganisms represents a significant risk factor in large-scale cultivation systems, but different species exhibit distinct spectral signatures that enable early detection. The ability to distinguish between cyanobacteria and green algae through optical measurements provides operators with early warning systems that can prevent culture losses and maintain product quality. Utilizing statistical analysis and emerging machine learning models, these subtle changes can be detected and addressed earlier than previously achieved.

[0275] Optical Sensing System and Illumination Design, Optical sensing utilized an Adafruit AS7341 11-channel spectral color sensor, providing comprehensive spectral analysis from 415 nm to 890 nm across ten distinct measurement channels. This multi-channel approach allows for simultaneous biomass quantification and spectral characterization for species identification and physiological monitoring, representing a significant advancement over single-wavelength turbidity sensors commonly employed in cultivation monitoring. The Attorney Docket No. 10975-081W01

[0276] sensor configuration employs an integration time (ATIME) of 50 units and a step count (ASTEP) of 999 steps, resulting in a total integration time calculated as

[0277] Integration Time = (ATIME + 1) x (ASTEP + 1) x 2.78μs = 51 x 1000 x 2.78μs = 141.78ms

[0278] A gain setting of 128 provides maximum sensitivity for low-light conditions, while communication with the sensor occurs via I2C using pins 8 (SDA) and 9 (SCL). The AS7341 provides direct digital output through its integrated analog-to-digital converter, eliminating external signal conditioning requirements and reducing system complexity while maintaining measurement precision suitable for quantitative analytical applications.

[0279] The near-infrared channel allows for precise 890 nm measurements optimized for biomass quantification independent of pigment variations, addressing a limitation of conventional optical density measurements that suffer from species-specific interference. The 890 nm wavelength selection represents a balance between minimal biological absorption, reduced solar interference compared to the standard 750 nm measurement, and peak responsivity of silicon photodetectors while avoiding water absorption effects that dominate above 1000 nm. The manufacturer provides each channel’s complete spectral response characteristics, including center wavelengths and Full Width at Half Maximum (FWHM) values, and forms the foundation for the Gaussian spectral reconstruction methodology detailed in subsequent sections. The AS7341 sensor channels exhibit FWHM values ranging from 26 nm for the violet channel (415 nm) to 52 nm for the red channel (680 nm), reflecting the trade-offs between spectral resolution and sensitivity inherent in multi-channel sensor design. These specifications define the fundamental spectral resolution limits of the sensor system while enabling sophisticated analytical capabilities through appropriate data processing techniques.

[0280] The dual-wavelength illumination system provides optimized light sources for biomass measurement and spectral analysis applications through carefully controlled LED brightness levels that prevent photodetector saturation while maintaining sufficient signal-to-noise ratios across the operational concentration range. A generic white LED operates at 1000Hz PWM frequency with 200 / 255 duty cycle (approximately 78% brightness) controlled via GPIO 2, providing broadband illumination for multi-wavelength spectral measurements across the visible spectrum. The QUD123 infrared LED centered at 890 nm operates at identical PWM frequency but reduced 15 / 255 duty cycle (approximately 6% brightness) Attorney Docket No. 10975-081W01

[0281] controlled via GPIO 3, explicitly optimized for biomass measurements that require high sensitivity without sensor saturation at elevated algae concentrations.

[0282] The LED drive specifications include approximately 3.2V forward voltage and 10- 15mA drive current for the white LED, while the IR LED operates at approximately 1.4V forward voltage and 2-3 mA drive current. PWM control utilizes 8-bit resolution at 1kHz base frequency using the ESP32 LEDC peripheral, providing precise brightness control essential for maintaining consistent measurement conditions across varying environmental temperatures and power supply fluctuations. The reduced IR LED brightness prevents photodetector saturation while maintaining sufficient signal-to-noise ratio for biomass measurements across the operational concentration range, enabling accurate quantification from dilute inoculation concentrations through harvest-ready densities.

[0283] Mechanical Design and Waterproof Housing. The waterproof housing featured a circular buoyant design fabricated from Delrin (polyoxymethylene) and machined stainless steel components optimized for stable floating operation and representative water sampling under realistic pond circulation conditions. Two precision measurement legs extend downward from the main housing, each containing high-grade quartz optical windows that provide complete optical transmission across the measurement spectrum while maintaining structural integrity under submersion pressures and thermal cycling typical of outdoor deployment environments. The mechanical design accommodates deployment depths from 1 to 6 inches below the water surface, enabling operation across pond systems with varying water levels and circulation patterns while maintaining consistent optical measurement geometry.

[0284] The optical path configuration maintains a precisely controlled 1.0 cm path length through the water column, matching standard spectrophotometric cuvette dimensions to facilitate direct correlation with laboratory reference measurements. High-grade quartz windows provide complete UV-NIR transmission with minimal optical distortion. O-ring seals with threaded mechanical compression achieve waterproof operation rated for continuous submersion under pond operating conditions. The open flow design allows free water circulation through the measurement volume during pond operation, ensuring representative sampling of the algae culture while maintaining direct line-of-sight optical measurement across the water gap without flow restrictions that could affect local mixing patterns.

[0285] Data Processing and Analytical Methods Attorney Docket No. 10975-081W01

[0286] Measurement Protocols and Timing Architecture. The sensor implements a three-phase measurement protocol designed as a finite state machine to ensure consistent measurement conditions and accurate background subtraction while accounting for LED thermal transients and environmental variations that could affect measurement accuracy. This systematic approach provides a comprehensive statistical analysis of measurement quality through redundant sampling and real-time error quantification, essential for autonomous operation under variable field conditions.

[0287] The first phase performs background measurement with all LEDs disabled to quantify ambient light and sensor dark current contributions that must be subtracted from sample measurements to isolate algae-specific optical signals. Eight total readings are collected, with the first two discarded for sensor stabilization and the remaining six averaged for statistical analysis, providing robust baseline characterization while minimizing measurement time requirements. The second phase activates the white LED at 78% duty cycle following a 750 ms thermal stabilization delay, again collecting eight readings using identical warmup and averaging protocol for multi-wavelength spectral characterization across all visible channels. The third phase follows the same pattern with the IR LED activated at 6% duty cycle with equivalent stabilization and measurement protocols for biomass quantification independent of pigment interference, ensuring consistent measurement conditions across all optical channels.

[0288] Each measurement phase incorporates 175 ms integration delays between LED activation and sensor reading to ensure thermal equilibrium in LED output characteristics. This proves critical for measurement repeatability under varying environmental temperatures. Inter-phase delays of 750 ms provide LED stabilization between state transitions, eliminating thermal crosstalk effects that could introduce systematic measurement errors during continuous operation. The complete three-phase measurement cycle requires approximately 30seconds, followed by 20-minute deep sleep intervals that conserve battery power while providing sufficient temporal resolution for cultivation monitoring applications Buoyancy characteristics are optimized to maintain stable floating orientation regardless of measurement leg immersion depth. At the same time, the circular geometry provides rotational stability under wind loading and wave action typical of outdoor pond environments. The housing design incorporates access ports for maintenance and component replacement while maintaining waterproof integrity during regular operation. Material selection prioritizes chemical resistance to cultivation media, UV stability under outdoor Attorney Docket No. 10975-081W01

[0289] exposure, and thermal expansion compatibility across operational temperature ranges spanning typical outdoor cultivation conditions.

[0290] Signal Processing and Absorbance Calculations. Raw sensor measurements undergo systematic processing to eliminate environmental interference and extract quantitative biological information through established optical principles adapted for the discrete multi-channel measurement approach. Background subtraction removes ambient light contamination and sensor dark current effects by calculating the corrected signal as the raw measurement minus the background average, applied independently to each of the 12 spectral channels to ensure accurate spectral analysis and eliminate systematic measurement errors that could compromise species identification or biomass quantification accuracy.

[0291] Following background correction, apparent absorbance values are calculated using the Beer-Lambert relationship, representing the fundamental optical principle underlying quantitative spectrophotometry. This calculation represents the background-corrected signal with algae present and includes terms representing the background-corrected signal through transparent medium and the dimensionless absorbance suitable for concentration correlation across different species and cultivation conditions. This approach maintains consistency with established analytical chemistry principles while accommodating the discrete wavelength measurements provided by the multi-channel sensor system. Measurement uncertainty propagates through these calculations using the standard error followed by absorbance error propagation combination. This provides quantitative uncertainty estimates for each spectral channel, enabling statistical validation of measurement reliability and confidence interval determination across varying environmental conditions and biological states. Real-time statistical analysis employs Welford’s algorithm for numerically stable online variance calculation without storing individual measurements, essential for microcontroller memory constraints, while maintaining the statistical rigor required for quantitative analytical applications.

[0292] Spectral Analysis and Pattern Recognition Methods. Multi-wavelength spectral data undergoes comprehensive processing to extract species identification signatures and physiological state indicators beyond simple biomass quantification. This represents a significant advancement over conventional single-wavelength turbidity measurements that provide limited biological information. Raw discrete spectral analysis involves directly plotting eight-channel absorbance measurements to create characteristic spectral signatures Attorney Docket No. 10975-081W01

[0293] suitable for species identification and temporal monitoring of pigment composition changes that indicate culture health status or contamination events.

[0294] Sum-normalized spectral processing divides each wavelength measurement by the total visible spectrum absorbance (415 -680 nm) to eliminate concentration effects while preserving species-specific spectral patterns, enabling direct comparison across different biomass levels and cultivation conditions. This normalization provides robust pattern recognition capabilities independent of absolute biomass concentration, which is critical for species identification applications where concentration variations could mask fundamental spectral characteristics. The normalization approach proves particularly valuable for contamination detection applications where identifying competing organisms requires discrimination based on spectral patterns rather than absolute absorbance levels.

[0295] Gaussian Distribution. Gaussian distribution spectral reconstruction represents a sophisticated approach to generating continuous spectral curves from the discrete eight¬ channel measurements, enabling direct comparison with traditional high-resolution spectrophotometry while preserving the fundamental measurement accuracy of the simplified sensor system. This reconstruction method addresses the inherent limitation that discrete multi-channel sensors produce stepped spectral profiles rather than the smooth, continuous spectra characteristic of scanning spectrophotometers used in most literature references and spectral databases.

[0296] The reconstruction process utilizes the known spectral response characteristics of each AS7341 sensor channel, specifically the FWHM specifications provided by the manufacturer for each wavelength band. These FWHM values are converted to Gaussian standard deviation parameters using the relationship σ = FWHM / 2.355, where the factor 2.355 represents the mathematical conversion between FWHM and standard deviation for Gaussian distributions. Each of the eight discrete absorbance measurements is then modeled as a Gaussian peak centered at the corresponding channel wavelength, with the peak intensity equal to the measured absorbance value and the peak width determined by the sensor’s spectral response function.

[0297] The mathematical implementation creates a fine wavelength grid spanning 415 to 680 nm at 2 nm intervals, providing sufficient resolution to capture spectral features while maintaining computational efficiency. For each discrete measurement point, a Gaussian contribution is calculated. The complete reconstructed spectrum can then be obtained by summing the Gaussian contributions from all eight channels. Attorney Docket No. 10975-081W01

[0298] This approach provides several advantages for spectral analysis applications. The reconstruction accounts for the finite spectral bandwidth of each sensor channel rather than treating measurements as infinitely narrow spectral lines, resulting in physically realistic spectral profiles that reflect the actual sensor response characteristics. The smooth, continuous output facilitates direct comparison with literature spectra and facilitates application of traditional spectroscopic analysis techniques developed for high- resolution instruments. Additionally, the method provides meaningful interpolation between discrete measurement points, revealing spectral features that might be obscured in simple discrete plotting while avoiding the artifacts that could arise from linear interpolation or spline fitting approaches that do not account for the underlying sensor physics.

[0299] The Gaussian reconstruction methodology is applied to raw absorbance measurements and sum-normalized spectral data, creating two complementary analytical presentations. Raw Gaussian spectra preserve absolute absorbance magnitudes and concentration-dependent scaling, enabling quantitative analysis and biomass correlation studies similar to traditional spectrophotometry. Sum-normalized Gaussian spectra eliminate concentration effects while maintaining spectral shape characteristics, providing optimal conditions for species identification and contamination detection applications where relative pigment patterns are more informative than absolute absorption magnitudes. This dual approach maximizes the analytical utility of the eight-channel measurements by providing both quantitative and qualitative spectral analysis capabilities within a single integrated sensing platform. In contrast, the underlying discrete measurements remain available for applications requiring direct sensor output without mathematical processing.

[0300] PCA. Principal component analysis processes normalized spectral data to identify underlying patterns distinguishing species, physiological states, and culture conditions through dimensionality reduction techniques that preserve maximum variance while eliminating measurement noise and environmental artifacts. Both sum-normalized and ratio¬ based approaches are implemented to maximize discrimination capability while reducing environmental sensitivity, enabling automated classification algorithms suitable for operational deployment in variable field conditions. The PCA framework provides a foundation for machine learning approaches that adapt to new species or cultivation conditions through continued data collection and model refinement. Attorney Docket No. 10975-081W01

[0301] Communication Protocols and Data Management. The LoRa data transmission system employs two distinct message types transmitted using AT command protocol with automatic payload length calculation and acknowledgment verification to ensure reliable data delivery under challenging RF environments typical of industrial cultivation facilities.

[0302] Primary data messages follow the format:

[0303] D:violet, blue, cyan, green, yellowgreen, yellow, orangered, red,

[0304] N 1R2, voltage, background

[0305] Providing complete spectral measurement data plus system status information is essential for remote monitoring and automated alert generation. Error data messages (standard deviation) use the format:

[0306] E:stdv violet, stdv blue, stdv cyan, stdv green, stdv yellowgreen,stdv yellow, stdv orangered, stdv red, stdv NIR2

[0307] To provide complete measurement uncertainty information for each spectral channel, enabling real-time assessment of data quality and identification of measurement conditions that might compromise analytical accuracy.

[0308] Power management cycles coordinate measurement activities with wireless communication to optimize battery life while maintaining adequate temporal resolution for cultivation monitoring applications. Sleep protocols include ESP32 deep sleep with timer wake-up, AS7341 power-down via PON bit clearing, and RYLR998 sleep mode activation, reducing total power consumption below 1 mA during inactive periods while ensuring automatic sensor reinitialization upon timer expiration. This power management architecture enables autonomous operation spanning multiple weeks under typical measurement schedules without manual intervention, which is critical for practical deployment in commercial cultivation facilities where daily maintenance requirements would compromise economic viability.

[0309] The data acquisition system utilizes a Python-based collection platform that receives LoRa transmissions and logs comprehensive data to CSV format with automated graphing and visualization accessible through ngrok for remote monitoring capability. Complete data logging includes all 12 spectral channels plus metadata encompassing voltage, temperature, error estimates, and timestamp information, enabling comprehensive post-processing analysis and system performance evaluation over extended deployment periods. Statistical validation frameworks employ correlation analysis through linear regression with confidence intervals, accuracy assessment via residual analysis and systematic bias evaluation, and performance Attorney Docket No. 10975-081W01

[0310] metrics including R2, RMSE, mean absolute error, and percentage of measurements within specification limits to quantify sensor performance relative to established analytical standards.

[0311] Experimental Design and Validation Methodology. The experimental validation framework was designed to systematically evaluate sensor performance across three distinct operational environments: controlled laboratory conditions for establishing fundamental measurement relationships, semi-realistic pond systems for validating performance under controlled cultivation scenarios, and full-scale field deployments for demonstrating operational capability under commercial conditions. This progressive validation approach ensures comprehensive assessment while identifying performance limitations and calibration requirements across the intended deployment spectrum.

[0312] Calibration Protocol Development. The calibration methodology establishes quantitative relationships between sensor measurements and reference standards through systematic dilution studies that span the operational concentration range of commercial algae cultivation systems. Serial dilutions were prepared from exponentially growing cultures maintained under standardized laboratory conditions, with each species cultivated in appropriate growth medium at optimal temperature and light conditions to ensure physiological consistency during calibration procedures.

[0313] Reference measurements utilized a NanoDrop spectrophotometer configured for 750 nm optical density measurements with 1cm path length cuvettes to match the sensor’s optical geometry. While this particular instrument consistently reads 2-3 times higher than conventional benchtop spectrophotometers, the linear relationship remains valid for calibration piuposes, with all comparative analyses maintaining consistent reference standards. Calibration points typically spanned concentrations from OD 0.05 to 4.0, encompassing the full operational range of pond cultivation systems from inoculation through harvest-ready densities.

[0314] The dilution protocol employed volumetric techniques with precision pipettes to ensure accurate concentration ratios, while timing constraints required completion of all measurements within 15minutes of sample preparation to minimize settling effects that could compromise measurement accuracy. For species exhibiting temporal instability in optical density measurements, alternative calibration approaches utilized known dilution ratios from characterized stock cultures combined with initial OD measurements before degradation occurred. This enables accurate calibration despite reference measurement limitations. Attorney Docket No. 10975-081W01

[0315] Statistical validation of calibration relationships employed linear regression analysis with confidence interval calculation to quantify measurement uncertainty and establish operational limits. The calibration process was repeated across multiple independent culture batches to verify consistency and identify potential sources of systematic variation that might affect long-term deployment accuracy.

[0316] Controlled Physiological Response Studies. The controlled degradation experimental framework was designed to evaluate sensor sensitivity to culture health changes and establish detection thresholds for identifying physiological stress or contamination events early. Sodium hypochlorite bleaching studies provided systematic photosynthetic system degradation across a range of severity levels, enabling quantification of spectral response to cellular damage progression.

[0317] Experimental cultures oiNannochloropsis oceanica were prepared at standardized density and divided into treatment groups receiving bleach concentrations of 0.02%, 0.2%, 5%, and 10% sodium hypochlorite, with untreated controls maintained for comparison. These concentration levels were selected to span from minimal stress responses detectable only through spectral analysis to complete photosynthetic system destruction visible through color change, providing a comprehensive assessment of sensor sensitivity across the physiological response spectrum.

[0318] Temporal measurement protocols captured spectral evolution during degradation through automated data collection at 7-minute intervals throughout the 28-minute experimental period. This high-frequency sampling identified early spectral changes that precede visible manifestation of cellular damage, which is critical for developing early warning capabilities in operational cultivation systems. Parallel visual documentation through standardized photography provided correlation between spectral changes and observable color transitions, validating the biological relevance of measured spectral evolution.

[0319] The experimental design incorporated statistical controls, including replicate treatments, consistent sample volumes, standardized mixing procedures, and environmental control to minimize confounding variables that could affect the interpretation of spectral response patterns. Data analysis focused on identifying minimal detectable changes in spectral characteristics that could serve as early indicators of culture health deterioration in operational monitoring applications.

[0320] Semi-Controlled Cultivation System Validation. Validation studies in controlled pond systems at Colorado State University provided realistic cultivation conditions while Attorney Docket No. 10975-081W01

[0321] maintaining experimental control over key environmental variables that affect sensor performance. The 75-liter cultivation systems incorporated automated pH control, programmable LED illumination systems, and circulation pumps to simulate commercial pond operations while enabling systematic performance evaluation.

[0322] Environmental control protocols established standardized cultivation conditions, including pH maintenance through automated CO2 injection, and diurnal lighting cycles programmed to transition gradually from 0% to 100% intensity over realistic time frames mimicking natural outdoor conditions. These controlled variables enabled systematic evaluation of sensor response to biological changes while minimizing environmental artifacts that could confound performance assessment.

[0323] The validation protocol incorporated continuous sensor monitoring and periodic reference sampling to establish measurement accuracy under dynamic cultivation conditions. Reference samples of 50 mL were collected every 24 hours using sterile technique and stored under refrigeration until spectrophotometric analysis could be performed, typically within 12 hours of collection to minimize sample degradation. Cross-validation with total organic carbon analysis provided independent biomass verification that does not rely on optical measurement principles, enabling assessment of fundamental measurement validity.

[0324] Multiple sequential cultivation runs with identical species and culture conditions enabled statistical validation of measurement consistency while identifying potential sources of systematic error or calibration drift over extended operational periods. Each cultivation cycle spanned 5-10days from inoculation through harvest, providing comprehensive sensor performance assessment across complete growth cycles, including exponential growth, stationary phase transitions, and harvest timing optimization.

[0325] Field Deployment Validation Methodology. A field deployment study in operational cultivation facilities provided insight into sensor performance under realistic commercial conditions, including environmental variability, contamination risks, and operational constraints typical of industrial algae production systems.

[0326] Deployment protocols established standardized installation procedures, including sensor positioning optimization for representative sampling, daily maintenance routines for optical window cleaning, and data collection schedules balancing information requirements with operational constraints.

[0327] Performance validation incorporated a systematic comparison between continuous sensor measurements and periodic reference sampling using portable spectrophotometric Attorney Docket No. 10975-081W01

[0328] equipment calibrated against laboratory standards. Environmental monitoring through weather station data and cultivation system sensors provided a correlation between measurement variations and external factors, including ambient light intensity, temperature fluctuations, and operational activities that might affect sensor performance.

[0329] Long-term deployment studies spanning multiple weeks enabled assessment of calibration stability, environmental resilience, and maintenance requirements under realistic operational conditions. Systematic documentation of fouling accumulation rates, cleaning effectiveness, and measurement drift patterns provided essential information for developing operational protocols and determining maintenance intervals for commercial deployment applications.

[0330] Environmental Interference Characterization. Systematic evaluation of environmental factors affecting measurement accuracy enabled the development of correction algorithms and the establishment of operational limitations under field conditions. Ambient light interference studies utilized controlled illumination sources spanning intensity ranges from darkness through full sunlight to quantify channel-specific sensitivity and develop mitigation strategies for continuous outdoor operation.

[0331] Voltage stability testing examined the relationship between power supply variations and measurement accuracy through systematically evaluating battery discharge effects, temperature induced voltage changes, and external power supply fluctuations on LED brightness control and sensor response characteristics. These studies established requirements for power management systems and identified correction factors needed to maintain measurement accuracy across operational voltage ranges.

[0332] Results

[0333] The comprehensive validation of the optical sensor system demonstrates exceptional performance across the full spectrum of operational requirements for real-time microalgae cultivation monitoring. This Example presents the results of the systematic evaluation, establishing the sensor’s capabilities for quantitative biomass measurement, species identification, physiological monitoring, and autonomous field deployment under realistic commercial conditions. The validation framework progresses from fundamental optical measurements through sophisticated spectral analysis to extended field trials, culminating in demonstrated operational capability that bridges laboratory precision with practical deployment requirements. Attorney Docket No. 10975-081W01

[0334] The experimental validation encompasses six interconnected analytical domains that establish sensor performance and practical utility.

[0335] Optical density estimation validation demonstrates universal algal biomass quantification capability across diverse species while identifying optimal operational parameters for maximum accuracy.

[0336] Multi-wavelength spectral analysis reveals species-specific signatures and physiological indicators that enable automated identification and health monitoring.

[0337] Controlled degradation studies establish sensitivity thresholds for detecting culture stress and contamination events early.

[0338] Principal component analysis validates pattern recognition capabilities essential for automated classification systems.

[0339] Extended pond deployment trials demonstrate sustained accuracy and reliability under realistic cultivation conditions.

[0340] Environmental interference characterization quantifies measurement robustness and identifies correction strategies for optimal field performance.

[0341] Optical Density Measurement Validation. The validation of optical density measurement capabilities establishes the fundamental quantitative accuracy required for algal biomass monitoring applications while demonstrating universal applicability across diverse microalgae species. All calibration measurements utilized a NanoDrop spectrophotometer as the reference standard, which consistently provides readings approximately 2-3 times higher than conventional benchtop spectrophotometers. While this scaling factor affects absolute values, the linear relationships and correlation coefficients remain valid for sensor validation purposes. This ensures that all comparative analyses maintain consistent reference standards throughout the evaluation process.

[0342] Near-Infrared Optimization and Performance. The systematic optimization of infrared LED brightness revealed critical operational parameters that maximize measurement sensitivity while preventing photodetector saturation across the biomass concentration ranges encountered in commercial cultivation systems. Figure 4A demonstrates the relationship between PWM duty cycle and calibration performance for Nannochloropsis oceanica, revealing that a 15% duty cycle provides optimal performance with R2= 0.99, slope = 0.203 ± 0.003. Higher duty cycles rapidly approached saturation limits, while lower settings compromised signal-to-noise ratios and measurement precision. Attorney Docket No. 10975-081W01

[0343] The 15% duty cycle setting was adopted for all subsequent measurements, establishing the operational standard “NIR 15” throughout this research. This optimization represents a balance between measurement sensitivity and operational range, allowing for accurate quantification from dilute inoculation concentrations through harvest-ready densities without dynamic brightness adjustment. While higher optical densities remain untested, the systematic relationship between duty cycle and operational range suggests that multiple brightness levels could extend measurement capability for specialized applications requiring quantifying exceptionally dense cultures.

[0344] The species-specific validation demonstrates remarkable consistency in near-infrared optical density relationships across taxonomically diverse microalgae, supporting the development of universal calibration approaches while maintaining species-specific optimization capabilities. Figure 4B presents calibration results for Monoraphidium minutum, achieving R2= 0.998 with slope = 0.293, confirming the universal applicability of the 890nm measurement principle while revealing the species-specific variations that reflect natural differences in cell morphology and internal structure.

[0345] The validation with Phaeodactylum tricornutum (Figure 4C) presented unique challenges due to the limited concentration range achievable with this diatom species and its tendency toward temporal instability in optical measurements. Despite operating within a constrained OD range of 0.05-0.2, the linear relationship remained evident R2= 0.953, with increased uncertainty reflected in larger confidence intervals. The temporal instability observed after 15 minutes of measurement suggests species- specific settling or aggregation behavior characteristic of diatom morphology, highlighting the importance of standardized measurement timing protocols for specific taxonomic groups.

[0346] The Tetraselmis suecica validation revealed fundamental limitations in conventional spectrophotometric reference measurements for particular species, necessitating alternative calibration approaches that demonstrate sensor reliability even when traditional reference methods fail. Initial calibration attempts using direct OD 750 measurements yielded unsatisfactory R2= 0.918 (Figure 4D), with measurement reliability degrading after approximately 15 minutes into the process due to settling effects commonly observed with this species at the facility where measurements were conducted.

[0347] An alternative dilution-based estimation approach demonstrated superior performance, achieving R2= 0.961 with reduced scatter and improved consistency (Figure 4E). This method utilized known dilution ratios from characterized stock cultures combined with Attorney Docket No. 10975-081W01

[0348] initial OD measurements collected before degradation occurred, enabling accurate calibration despite reference measurement limitations. The comparison between conventional and dilution-based approaches (Figure 4F) demonstrates the improved linearity achieved through the alternative methodology, validating the sensor’s fundamental accuracy while highlighting the importance of appropriate reference measurement protocols.

[0349] The comprehensive cross-species analysis reveals remarkable consistency in NIR- based optical density measurements, with calibration slopes ranging from 0.293 to 0.348 and all correlations exceeding R² > 0.95 across taxonomically diverse species (Figure 5). This consistency supports the development of universal calibration approaches for multi-species cultivation facilities while maintaining the capability for species-specific optimization when maximum accuracy is required. The narrow range of slopes demonstrates that fundamental light scattering properties remain consistent across diverse cell morphologies, validating the sensor’s core measurement principle while revealing that species-specific variations reflect genuine biological differences rather than measurement artifacts.

[0350] Table 1 quantifies the exceptional performance achieved across all tested species, with RMSE values consistently below 0.05 OD units and p-values indicating highly significant linear relationships (p < 0.001) for all calibration datasets. The consistency of these performance metrics across taxonomically diverse organisms validates the fundamental measurement approach while establishing quantitative accuracy benchmarks for operational deployment. These results demonstrate that the NIR approach achieves the measurement precision required for quantitative cultivation monitoring while maintaining broad applicability across the species diversity encountered in commercial operations.

[0351] Table 1. Performance Summary for NIR-Based OD Estimation

[0352] Strain Slope Slope SE R2 p-value RMSE Nannochloropsis oceanica 0.293 0.0448 0.998 < 0.001 0.0456 Monoraphidium minutum 0.338 0.0234 0.9989 < 0.001 0.0234 Phaeodactylum tricornutum 0.325 0.0367 0.953 < 0.001 0.0345 Tetraselmis suecica 0.341 0.0412 0.9912 < 0.001 0.0289

[0353] Visible Spectrum Summation as an Alternative Methodology. The visible spectrum summation approach provides valuable backup capability for biomass estimation Attorney Docket No. 10975-081W01

[0354] while demonstrating the analytical potential of multi-wavelength measurements beyond simple near-infrared quantification. Individual channel analysis for Nannochloropsis oceanica (Figure 6A) reveals strong linear correlations (R2> 0.99) across most wavelength channels, with violet (415 nm) and red (680 nm) channels showing slightly elevated variability likely reflecting sensor response characteristics at spectral boundaries

[0355] After more testing, the ideal channel or combination of channels could be

[0356] determined for each individual strain. For the current test and to serve as a proof of

[0357] concept absorbance values from channels 2-7 (blue through orange-red, 445-630 nm), excluding edge channels that exhibit higher variability while preserving the core spectral information essential for biomass correlation.

[0358] Testing across multiple LED brightness levels (duty cycles 100, 150, and 200) demonstrates minimal sensitivity to illumination variations within the operational range (Figure 6B), confirming measurement robustness compared to individual channel approaches that might be more susceptible to LED brightness fluctuations.

[0359] Cross-species validation reveals greater slope variability in the summation

[0360] approach than NIR measurements, reflecting the fundamental difference between wavelength-dependent pigment absorption and wavelength-independent scattering phenomena. While each species maintains excellent linearity (Figure 6C), the method

[0361] proves less robust for universal calibration applications due to species-specific pigment composition effects on visible wavelength absorption.

[0362] Table 2 presents comprehensive performance metrics demonstrating that while slopes vary significantly between species (3.08-5.50), correlation coefficients remain consistently high R2> 0.96), supporting the method’s reliability as an alternative measurement approach when properly calibrated for specific species or cultivation conditions.

[0363] Table 2. Performance Comparison for Visible Spectrum Summation Method.

[0364] Strain Brightness Slope Slope SE R2 p-value RMSE Monoraphidium minutum 200 3.6881 0.0333 0.9991 < 0.001 0.1449 Nannochloropsis oceanica 100 3.0795 0.0589 0.9953 < 0.001 0.3034 Nannochloropsis oceanica 150 3.0936 0.0305 0.9967 < 0.001 0.1646 Nannochloropsis oceanica 200 3.0781 0.0584 0.9953 < 0.001 0.3010 Attorney Docket No. 10975-081W01

[0365] Phaeodactylum tricornutum 200 5.5034 0.4145 0.9618 < 0.001 0.1671 Synechocystis 200 3.9854 0.0520 0.9990 < 0.001 0.0840 Tetraselmis suecica 200 5.3011 0.1892 0.9837 < 0.001 0.1059

[0366] Multi-Wavelength Spectral Analysis. Building upon the quantitative biomass measurement capabilities established through optical density validation, the multi-wavelength spectral analysis demonstrates sophisticated analytical potential beyond simple biomass quantification. The species-specific pigment composition effects observed in the visible spectrum summation approach reveal the underlying spectral information content that enables species identification, physiological monitoring, and culture health assessment by systematically evaluating absorption patterns across the visible spectrum. Three complementary analytical approaches reveal different aspects of this spectral information content: raw discrete-point absorbance values provide direct measurement interpretation, sum-normalized spectra enable species-independent pattern recognition, and Gaussian- distributed continuous spectra facilitate comparison with traditional spectrophotometric databases.

[0367] Raw Absorbance Spectral Characteristics. Applying Beer-Lambert principles to discrete eight-channel measurements yields quantitative spectral information that captures essential absorption features while maintaining practical measurement capability with simplified hardware. Figure 7A demonstrates representative spectra for Nannochloropsis oceanica across varying biomass concentrations, revealing precise concentration-dependent scaling while preserving consistent spectral shape characteristics essential for species identification applications.

[0368] The discrete nature of the eight-channel measurement produces characteristic stepped spectra that successfully capture essential algae absorption features, including chlorophyll absorption peaks around 445 nm and 680nm, along with the characteristic minimum around 550 nm corresponding to reduced absorption in the green spectral region. These fundamental spectral characteristics remain distinguishable despite the simplified measurement approach, validating the analytical capability for practical monitoring applications. The stepped spectral profile demonstrates that strategic wavelength selection preserves the essential information content required for species identification while maintaining the hardware simplicity necessary for cost-effective deployment. Attorney Docket No. 10975-081W01

[0369] Cross-species spectral comparison reveals distinct species-specific signatures that enable identification and monitoring applications across taxonomically diverse organisms. Figure 7B presents comprehensive spectral comparison demonstrating that Nannochloropsis oceanica exhibits the classic green algae pattern with intense blue and elevated red channel absorption, while Synechocystis demonstrates broader absorption characteristics of cyanobacteria with enhanced orange-red absorption due to phycobiliprotein content.

[0370] The overlaid presentation (Figure 7C) facilitates direct comparison of spectral characteristics, revealing universal features shared across photosynthetic organisms and species specific variations that provide discrimination capability. The consistent presence of chlorophyll absorption features validates the fundamental measurement approach, while distinct patterns in accessory pigment regions (450-550 nm) provide the discrimination capability essential for automated species identification systems. These results demonstrate that the eight-channel approach captures sufficient spectral information to distinguish between taxonomically diverse organisms while maintaining the measurement simplicity required for autonomous field deployment. The clear differentiation between green algae, cyanobacteria, and diatoms through discrete spectral measurements establishes the foundation for automated classification algorithms that could operate without human interpretation.

[0371] Sum-Normalized Spectral Analysis. While raw absorbance spectra provide quantitative measurement capability, sum normalization eliminates concentration effects while preserving species-specific spectral patterns. This facilitates direct pattern comparison across different organisms and biomass levels by expressing each wavelength’s contribution as a fraction of visible absorption. This normalization approach addresses the concentration-dependent scaling observed in raw measurements while amplifying the subtle spectral differences that enable species discrimination independent of biomass levels. Figure 8A demonstrates the transformation from raw absorbance to normalized spectra for Nannochloropsis oceanica, showing how concentration effects are eliminated while essential spectral features remain preserved for analytical applications. This allows for identification and relevant information across the entire range of concentrations. The spread of the violet channel indicates inconsistencies on the edge of sensor resolution that may need to be addressed in future iterations.

[0372] The normalization process amplifies subtle differences in pigment ratios that may be obscured by biomass concentration variations in raw measurements, enabling enhanced discrimination capability that is particularly valuable for early contamination detection and Attorney Docket No. 10975-081W01

[0373] physiological monitoring applications. Figure 8B presents normalized spectra across all measured species, demonstrating enhanced discrimination of pigment composition patterns while maintaining the fundamental spectral characteristics required for species identification.

[0374] The overlaid normalized spectra (Figure 8C) provide clear visualization of inter-species differences while highlighting standard photosynthetic features, facilitating rapid species identification and enabling monitoring of temporal changes in pigment composition that may indicate physiological stress, contamination events, or culture state transitions critical for operational cultivation management. The enhanced discrimination capability revealed through normalization demonstrates that relative pigment ratios provide more robust identification signatures than absolute absorption values. This facilitates species identification across the range of biomass concentrations encountered during cultivation cycles from inoculation through harvest.

[0375] Gaussian Spectral Reconstruction. While raw and normalized discrete measurements demonstrate clear species discrimination capability, the Gaussian distribution approach creates continuous spectra from discrete eight channel measurements through sophisticated mathematical reconstruction that accounts for each sensor channel’s known spectral response characteristics as described herein. This methodology facilitates direct comparison with traditional spectrophotometric databases while maintaining the fundamental accuracy of the simplified sensor measurements, bridging the gap between hardware simplification and established analytical protocols.

[0376] Figure 9 A presents representative continuous spectra generated for multiple species, demonstrating good agreement with expected spectral characteristics while preserving the essential measurement information obtained from the discrete eight-channel approach. The smooth spectral curves enable comparison with traditional spectrophotometry while maintaining species-specific features crucial to identification applications.

[0377] The overlaid Gaussian spectra (Figure 9B ) demonstrate that species discrimination capability remains robust in the reconstructed format, with some spectral features becoming more clearly defined. In contrast, others show reduced resolution compared to the discrete measurement approach. This trade-off reflects the inherent balance between spectral smoothing and feature preservation in the reconstruction process.

[0378] Combining Gaussian reconstruction with sum normalization (Figure 9C) produces spectra that closely resemble traditional spectrophotometric presentations while maintaining the analytical advantages of normalized measurements. This approach provides optimal Attorney Docket No. 10975-081W01

[0379] compatibility with existing spectral databases and established analysis protocols while preserving the species discrimination capability essential for practical monitoring applications.

[0380] The direct comparison with literature references (Figure 9D) validates the reconstruction approach through demonstrated agreement in major spectral features, though the limited spectral resolution inherent in the eight-channel approach results in some loss of fine spectral detail compared to high-resolution spectrophotometry. The second chlorophyll absorption peak around 675-680 nm is partially resolved but shows reduced amplitude compared to high-resolution measurements, reflecting bandwidth limitations while maintaining sufficient detail for practical species identification and monitoring applications.

[0381] The multi -wavelength spectral analysis establishes sophisticated analytical capabilities that transform the sensor from a simple biomass monitor into a comprehensive biological assessment platform. The three complementary analytical approaches demonstrate that strategic wavelength selection and appropriate data processing can extract meaningful biological information while maintaining hardware simplicity. Raw spectra provide quantitative measurements suitable for biomass correlation, normalized spectra enable species identification independent of concentration effects, and Gaussian reconstruction facilitates integration with existing analytical protocols. These capabilities provide the foundation for monitoring physiological changes and culture health beyond simple biomass quantification.

[0382] Controlled Physiological Response Studies. The spectral analysis capabilities demonstrated across healthy cultures provide the foundation for detecting physiological changes and distinguishing between healthy and compromised cultures. The controlled bleaching experiments systematically evaluate the sensor’s sensitivity to culture health changes through progressive photosynthetic system degradation, establishing detection thresholds for early identification of physiological stress or contamination events in operational cultivation systems. The experimental framework utilized Nannochloropsis oceanica cultures treated with sodium hypochlorite concentrations spanning 0.02% to 10%, enabling assessment of detection sensitivity across the complete physiological response spectrum from minimal stress to complete system destruction.

[0383] A visual progression from healthy green coloration to complete pigment loss was observed after 28 minutes of exposure, providing direct correlation between spectral Attorney Docket No. 10975-081W01

[0384] measurements and observable physiological changes that validate the biological relevance of the sensor response.

[0385] The continuous spectral monitoring reveals systematic changes in absorption patterns that provide a quantitative assessment of photosynthetic system degradation. Figure 10A demonstrates the complete spectral evolution during the bleaching process, with clear reductions in chlorophyll absorption peaks and gradual baseline shifts that reflect progressive cellular disruption across the temporal measurement series.

[0386] Sum normalization reveals subtle changes in relative pigment composition that precede complete pigment loss, demonstrating enhanced sensitivity for early detection applications. Figure 10B shows how normalized spectra can detect physiological stress before visible changes become apparent. 'This supports early warning capabilities essential for operational cultivation management, where intervention timing proves critical for preventing culture losses.

[0387] The temporal analysis demonstrates detection capability at concentrations as low as 0.02% sodium hypochlorite (Figure 10C), though this sensitivity level likely exceeds practical requirements for pond deployment applications where environmental noise would mask such subtle changes. However, the demonstrated sensitivity establishes detection thresholds that enable early identification of culture deterioration under controlled conditions where intervention timing proves critical for maintaining culture health. Detecting physiological stress before visible manifestation provides the early warning capability essential for preventing culture losses in commercial operations where rapid intervention can salvage compromised cultures.

[0388] The near-infrared channel response (Figure 10D) reveals approximately 15% reduction in signal intensity as cells lose physiological integrity, indicating continued sensitivity to physical biomass presence even after photosynthetic system destruction.

[0389] While less dramatic than the 50% reductions reported in literature, this response demonstrates that the fundamental measurement principle remains valid across the complete physiological spectrum from healthy cultures through complete system failure.

[0390] The controlled physiological response studies validate the sensor’s capability to detect culture health changes through visible spectrum evolution and near-infrared biomass measurements. The systematic spectral changes during photosynthetic system degradation establish sensitivity thresholds while demonstrating that early detection of physiological stress is achievable before visible manifestation. These findings confirm that the multi- Attorney Docket No. 10975-081W01

[0391] wavelength approach provides comprehensive culture assessment capability beyond simple biomass quantification to enable health monitoring essential for operational cultivation management. Combining spectral pattern changes and absolute signal variations provides multiple independent indicators of culture status, enhancing reliability for autonomous monitoring applications.

[0392] Principal Component Analysis and Pattern Recognition. The comprehensive spectral datasets from species identification and physiological monitoring provide the foundation for automated pattern recognition and classification systems. Principal component analysis reveals underlying spectral patterns that enable automated classification and monitoring applications through systematic dimensionality reduction that preserves maximum variance while eliminating measurement noise and environmental artifacts. The study demonstrates apparent species clustering and physiological state discrimination, supporting automated identification systems and early warning capabilities essential for operational cultivation management.

[0393] The sum-normalized PCA results (Figure 11 A) demonstrate apparent clustering of samples by species with distinct separation between major taxonomic groups.

[0394] Nannochloropsis oceanica forms a tight cluster, indicating consistent spectral characteristics. At the same time, Synechocystis occupies a distinctly different region, reflecting fundamental differences in pigment composition and photosynthetic apparatus organization between cyanobacteria and eukaryotic algae.

[0395] The bleaching experiments create distinct trajectories showing progression from healthy to degraded culture states, validating the PCA approach for health monitoring applications where early detection of physiological changes enables intervention before irreversible culture damage occurs. This temporal progression provides a quantitative assessment of spectral changes associated with culture decline while demonstrating the analytical framework necessary for automated health monitoring systems.

[0396] The ratio-based PCA approach (Figure 11B) demonstrates enhanced discrimination capability and reduced sensitivity to environmental variations compared to absolute spectral measurements. The improved clustering performance reflects reduced sensitivity to illumination variations, sensor drift, and measurement geometry changes that can affect absolute measurements, making this approach particularly suitable for field deployment applications where environmental conditions vary significantly Attorney Docket No. 10975-081W01

[0397] Figure 11C presents enhanced visualization showing distinct groupings for experimental categories including bleach experiments, calibration samples, and different strain variations. The clear separation between these categories demonstrates the PCA approach’s capability for automated classification while highlighting the potential for machine learning applications that could adapt to new species or cultivation conditions through continued data collection and model refinement.

[0398] The principal component analysis validates the pattern recognition capabilities essential for autonomous monitoring systems by demonstrating clear separation between species, physiological states, and experimental conditions. The enhanced discrimination achieved through ratio-based approaches provides robust classification capability suitable for field deployment, where environmental variations could affect absolute measurements. These results establish the analytical framework necessary for automated species identification and health monitoring that operates without human interpretation, transforming complex spectral data into actionable information for cultivation management. The clear clustering patterns and physiological state trajectories provide the foundation for machine learning algorithms that could continuously improve classification accuracy through operational experience.

[0399] Pond Deployment Validation. The comprehensive laboratory validation encompassing quantitative biomass measurement, species-specific spectral analysis, physiological monitoring, and automated pattern recognition provides the foundation for evaluating sensor performance under realistic cultivation conditions. The controlled mini¬ pond validation at Colorado State University systematically evaluates the integration of all sensor capabilities under semi-realistic conditions while maintaining experimental control over key variables essential for quantitative performance assessment. The 75 L cultivation systems incorporated automated pH control, programmable LED illumination, and circulation systems that simulate commercial pond operations while enabling precise performance evaluation under controlled conditions.

[0400] A programmable LED illumination system provided realistic diurnal cycles with gradual transitions from 0% to 100% intensity, enabling systematic evaluation of sensor response to biological changes while minimizing environmental artifacts that could confound performance assessment. This controlled environment proved essential for establishing baseline performance characteristics before field deployment trials.

[0401] Multi-Run Performance Validation. 'The comprehensive dataset from five sequential cultivation Runs (Figure 12 A) demonstrates consistent sensor performance across Attorney Docket No. 10975-081W01

[0402] multiple independent experiments, with each run spanning 5-10 days and exhibiting clear growth phases and harvest cycles that validate measurement reliability under dynamic biological conditions.

[0403] The detailed analysis of the first 100 hours of cultivation (Figure 12B) reveals growth dynamics impossible to capture through manual sampling approaches, with clear correlation between sensor measurements and programmed light cycles that validates biological sensitivity and demonstrates the temporal resolution advantages of continuous monitoring over conventional discrete sampling protocols.

[0404] The systematic comparison between sensor measurements and benchtop spectrophotometer references (Figure 12C) demonstrates excellent correlation for both NIR and visible summation approaches, with the NIR method showing superior consistency and reduced variability compared to the summation approach. Reference samples collected every 24 hours and stored under refrigeration until analysis systematically validated sensor accuracy under dynamic cultivation conditions.

[0405] Performance evaluation against the target accuracy specification of ±10% relative to reference measurements reveals that 73.9% of measurements fall within the target range, with mean absolute percent difference of 6.12% and maximum deviation of only 12.8%. This performance demonstrates sensor accuracy that exceeds typical requirements for cultivation monitoring while maintaining operational simplicity essential for practical deployment. / Achieving sub-10% average error under dynamic cultivation conditions validates the sensor’s capability for quantitative monitoring that approaches laboratory-quality measurements in field environments.

[0406] The overall correlation analysis (Figure 12D) confirms exceptional sensor performance withR“ = 0.987 and slope = 1.042, indicating minimal systematic bias and near¬ perfect linearity across the operational range in a controlled setting. The slope near a value of 1 demonstrates that calibration relationships established under controlled conditions remain valid during dynamic cultivation scenarios.

[0407] The residual analysis (Figure 12E) usually demonstrates distributed measurement errors with mean near zero and absence of systematic trends, confirming the appropriateness of the linear calibration model and validating sensor performance consistency across varying cultivation conditions and biomass levels.

[0408] The independent validation through total organic carbon analysis (Figure 12F ) provides critical verification that optical measurements accurately reflect actual biomass Attorney Docket No. 10975-081W01

[0409] content rather than responding to environmental artifacts or measurement geometry effects. The strong correlation (R2= 0.954) between NIR absorbance and TOC confirms the fundamental validity of the optical measurement approach for biomass assessment applications.

[0410] Spectral Analysis Under Cultivation Conditions. The comprehensive spectral analysis during active cultivation reveals temporal changes in pigment composition and culture health that laboratory studies cannot observe. Raw absorbance spectra measured across different optical densities during cultivation (Figure 13A) demonstrate consistent spectral characteristics across different pond runs and growth phases, validating measurement repeatability and biological relevance under dynamic conditions

[0411] Sum-normalized spectral patterns (Figure 13B) demonstrate relatively stable spectral characteristics throughout cultivation cycles, indicating healthy culture conditions without significant contamination or stress events that would manifest as systematic spectral evolution patterns.

[0412] The comparison across all pond runs (Figure 13C) demonstrates good consistency between independent cultivation experiments, validating the normalization approach for field applications while establishing baseline spectral patterns that could serve as reference standards for automated monitoring systems.

[0413] Environmental Interference Characterization. The systematic evaluation of environmental factors affecting measurement accuracy establishes operational limitations while identifying correction strategies essential for reliable field deployment. Ambient light interference represents the primary ecological challenge, with channel-specific sensitivities correlating with solar spectral characteristics requiring targeted mitigation approaches.

[0414] Individual channel analysis (Figure 14) reveals that blue and violet channels exhibit higher sensitivity to ambient light interference. In contrast, longer wavelength channels demonstrate enhanced stability for continuous outdoor operation. This wavelength-dependent sensitivity pattern aligns with expected solar spectral characteristics and validates the strategic wavelength selection for the measurement system. However, overall, there is no clear impact of ambient light interference on the individual channels.

[0415] The NIR channel performance demonstrates minimal sensitivity to ambient light variations, validating the 890 nm wavelength selection for continuous outdoor monitoring applications where solar interference could compromise measurement accuracy with conventional 750 nm approaches. Attorney Docket No. 10975-081W01

[0416] Multi-day field deployment analysis (Figure 15A) reveals systematic diurnal patterns correlating with overhead light intensity cycles, demonstrating predictable interference patterns that enable algorithmic correction for enhanced measurement accuracy. However, violet and cyan channels exhibit behavior patterns distinct from other channels, suggesting additional interference mechanisms beyond simple solar spectrum effects.

[0417] The correlation analysis between light intensity and NIR signal variations (Figure 15B) indicates that measurement variability derives primarily from sources other than overhead illumination, with voltage fluctuations representing a more significant influence on measurement stability than environmental light interference.

[0418] Voltage stability analysis (Figure 15C) reveals systematic patterns during 24-hour deployment cycles that likely reflect wall outlet fluctuations rather than battery discharge effects. These predictable patterns enable the development of voltage-based correction algorithms that could enhance measurement stability during extended deployment periods.

[0419] The brightness-dependent error analysis (Figure 15D) demonstrates minimal magnitude effects from LED brightness variations within the operational range, confirming that the optimized brightness settings provide robust performance. In contrast, the benefits of lower brightness levels outweigh potential accuracy compromises from reduced signal intensity.

[0420] The comprehensive environmental interference analysis demonstrates that interference effects remain manageable through appropriate sensor design and algorithmic correction approaches. The systematic nature of most interference sources enables predictive correction strategies that maintain measurement accuracy under realistic deployment conditions while identifying the operational boundaries that define reliable performance limits.

[0421] The comprehensive validation demonstrates exceptional sensor performance across operational requirements for autonomous microalgae cultivation monitoring. The progression from fundamental optical measurements through sophisticated spectral analysis to extended field deployment establishes measurement capabilities that bridge laboratory precision with practical deployment requirements. Near-infrared biomass quantification achieves universal applicability across diverse species with accuracy approaching laboratory standards, while multi-wavelength spectral analysis enables species identification and physiological monitoring beyond simple biomass measurement. The controlled degradation studies confirm early detection capabilities for culture health monitoring, while principal component analysis Attorney Docket No. 10975-081W01

[0422] validates automated classification potential essential for autonomous operation. Extended pond deployment trials demonstrate sustained accuracy and reliability under realistic cultivation conditions, confirming that sophisticated monitoring need not require prohibitive complexity or cost. The systematic characterization of environmental interference effects establishes operational boundaries while identifying correction strategies that maintain measurement accuracy across diverse deployment scenarios. These results demonstrate that the sensor system achieves the analytical capabilities necessary to transform microalgae cultivation from empirical management to quantitative optimization through continuous biological feedback.

[0423] Discussion

[0424] Physical Basis of Universal Near-Infrared Response. The consistency in near-infrared optical density relationships across taxonomically diverse microalgae reveals insights into the physical basis oflight-biomass interactions that may be independent of biological complexity. The narrow calibration slope range (0.29-0.35) observed across green algae, cyanobacteria, and diatoms suggests that near-infrared light scattering properties remain appropriately uniform despite vast differences in cell architecture, size distributions, and internal organization that have evolved over billions of years.

[0425] This universality can be understood through Mie scattering theory, where particles similar in size to the wavelength of incident light produce scattering patterns primarily dependent on the refractive index difference between cells and the surrounding medium rather than specific cellular components. At 890 nm, the wavelength approaches the size scale of many microalgae cells (1-20μm), placing the interaction in the Mie scattering regime where geometric cross-section becomes the dominant factor determining light attenuation. The relative insensitivity to internal cellular structure explains why organisms with fundamentally different photosynthetic apparatus - from the elaborate chloroplast architecture of eukaryotic algae to the simpler thylakoid arrangements of cyanobacteria -exhibit similar optical density relationships.

[0426] While small in absolute terms, the species-specific slope variations likely reflect fundamental differences in average cell size, shape factors, and refractive index properties between species. Monoraphidium minuium, with its smaller cell morphology, shows a slightly higher slope (0.338) compared to the more spherical Nannochloropsis oceanica (0.293), consistent with increased scattering cross-section from non-spherical particles. These variations validate that the sensor responds to genuine biological differences rather than Attorney Docket No. 10975-081W01

[0427] measurement artifacts, while remaining sufficiently small to enable universal calibration approaches across diverse cultivation scenarios.

[0428] The minimal influence of pigment composition on near-infrared measurements represents a critical advantage over visible wavelength approaches. While chlorophyll and accessory pigments strongly absorb visible light through specific electronic transitions, their influence diminishes dramatically at 890 nm where electronic absorption becomes negligible. This wavelength selection isolates the physical scattering component of light attenuation from the absorption effects that introduce species-specific variability in conventional 750 nm optical density measurements.

[0429] Pigment-Dependent Spectral Signatures and Evolutionary Context. The greater variation (nearly 2x) in visible spectrum summation slopes (3.08-5.50) compared to the narrow range observed in near-infrared measurements illuminates the profound influence of pigment evolution on algae’s optical properties. This variability reflects millions of years of evolutionary pressure that shaped light-harvesting strategies to minimize competitive overlap between coexisting species while maximizing photosynthetic efficiency under specific environmental conditions. Under optimal conditions, isolated wavelengths can be used as a biomass proxy. Generally, NIR should be utilized when measuring biomass, while absorbance is used to demonstrate health and productivity.

[0430] The spectral signatures observed across different taxonomic groups reveal the underlying pigment architectures that define each organism’s ecological niche. Synechocystis demonstrates enhanced absorption in the orange-red region (630-680 nm) characteristic of phycobiliprotein-containing organisms, reflecting the evolutionary adaptation of cyanobacteria to utilize wavelengths penetrating deeper into aquatic environments where green light becomes limited. This contrasts sharply with the blue and red chlorophyll dominance observed in green algae like Nannochloropsis oceanica, optimized for surface waters where the full solar spectrum remains available.

[0431] The spectral characteristics of Phaeodactylum tricornutum reflect the evolutionary history of diatoms, which acquired chloroplasts through secondary endosymbiosis involving a red algae ancestor. The resulting pigment complement combines chlorophyll a and c with fucoxanthin, creating spectral signatures distinct from primary endosymbiotic algae and cyanobacteria. This evolutionary complexity manifests in the sensor measurements as intermediate absorption patterns, which enable taxonomic discrimination while providing insights into the organization of the photosynthetic apparatus. Attorney Docket No. 10975-081W01

[0432] The sum normalization approach proves particularly revealing for understanding these evolutionary relationships, as it eliminates biomass effects while preserving relative pigment ratios. The enhanced species discrimination achieved through normalization suggests that pigment stoichiometry provides more robust taxonomic indicators than absolute absorption magnitudes, consistent with the evolutionary constraint that pigment ratios must remain optimized for photosynthetic efficiency regardless of overall cellular pigment content.

[0433] Methodological Insights from Reference Measurement Limitations. The calibration challenges encountered with particular species, particularly Tetraselmis suecica and Phaeodactylum tricornutum, provide critical insights into the limitations of conventional reference measurement approaches and highlight important considerations for sensor validation methodologies. The superior performance achieved through dilution-based estimation compared to direct spectrophotometric measurements reveals fundamental problems with the assumed reliability of OD 750 measurements as reference standards.

[0434] The temporal instability observed in spectrophotometric measurements for these species likely reflects settling and aggregation behaviors that vary significantly between taxonomic groups. Diatoms like Phaeodactylum tricornutum possess dense silica frustules that promote rapid settling. At the same time, flagellated species such as Tetraselmis suecica may exhibit aggregation behaviors that alter optical properties relevant to measurement protocols over time scales. These biological factors introduce systematic errors in reference measurements, which can artificially constrain the apparent sensor performance.

[0435] Observing that sensor measurements often prove more consistent than reference methods raises essential questions about analytical validation protocols in biological systems. Traditional analytical chemistry approaches assume that reference methods provide a ground truth against which new techniques are evaluated; however, biological systems introduce complexities that can compromise the reliability of reference measurements. The demonstration that innovative sensor technologies can exceed the stability of conventional reference methods suggests a need for alternative validation approaches that account for biological variability and temporal stability limitations.

[0436] The successful application of total organic carbon analysis as an independent validation method provides a path forward for establishing more robust reference standards. While requiring sample processing and laboratory analysis, TOC measurements provide direct quantification of biological material that avoids optical artifacts and temporal instability issues affecting spectrophotometric approaches. The strong correlation observed Attorney Docket No. 10975-081W01

[0437] between sensor measurements and TOC (R2= 0.954) validates the fundamental accuracy of the optical approach while establishing chemical analysis as a superior reference standard for biological monitoring applications.

[0438] Spectral Reconstruction and Information Content Preservation. The Gaussian reconstruction methodology provides valuable insights into the relationship between spectral resolution and biological information content, demonstrating that strategic wavelength selection can preserve essential identification capabilities while dramatically reducing measurement complexity. The successful generation of meaningful continuous spectra from eight discrete measurements validates the hypothesis that biological identification relies on broad spectral features rather than fine spectral detail.

[0439] The mathematical framework underlying Gaussian reconstruction considers the finite bandwidth characteristics of each sensor channel, resulting in physically realistic spectral profiles that accurately reflect the actual measurement process, rather than idealized, infinitely narrow spectral lines. This approach acknowledges the fundamental trade-off between spectral resolution and measurement precision, where broader channel bandwidths increase signal-to-noise ratios at the cost of spectral detail. The preservation of essential chlorophyll absorption features and accessory pigment signatures, despite significant bandwidth limitations, demonstrates that biological discrimination depends more on gross spectral patterns than on fine spectral structure.

[0440] Comparison with literature spectra reveals both the capabilities and limitations of the simplified approach. The primary identification characteristics remain distinguishable while secondary chlorophyll peaks show reduced amplitude and fine spectral features become smoothed. This finding has broader implications for analytical instrumentation design, suggesting that pursuing maximum spectral resolution may be unnecessary for many practical applications where key discrimination features can be captured through strategic wa velength selection.

[0441] The integration of normalized and raw spectral approaches provides complementary information, enhancing the overall analytical capability. Raw measurements preserve quantitative relationships essential for biomass correlation, while normalized patterns eliminate concentration effects that could mask species-specific signatures. The convergent results obtained through different analytical approaches validate the robustness of the measurement system while providing multiple independent sources of biological information. Attorney Docket No. 10975-081W01

[0442] Principal Component Analysis and Biological Pattern Recognition. The apparent clustering observed in principal component analysis reveals the underlying structure of spectral -biological relationships while providing insights into the dimensionality of biological information content in optical measurements. The separation between taxonomic groups in PCA space reflects fundamental differences in pigment organization and photosynthetic apparatus architecture that transcend the specific wavelength channels employed in the measurement system.

[0443] The enhanced discrimination achieved through ratio-based PCA analysis highlights the importance of relative spectral relationships over absolute measurement values. Channel ratios effectively normalize environmental factors that affect all wavelengths proportionally while preserving biological signals that create wavelength-specific variations. This mathematical transformation mirrors biological strategies where pigment ratios remain optimized for photosynthetic efficiency despite variations in absolute pigment concentrations due to environmental conditions or growth phase effects.

[0444] The absorbance degradation observed during controlled bleaching experiments provides particularly valuable insights into the relationship between spectral patterns and physiological states. The progression from healthy clusters toward degraded regions in PCA space demonstrates that physiological stress manifests as systematic shifts in spectral patterns rather than random variations. This finding validates the potential for automated health monitoring based on pattern recognition approaches that can detect physiological changes before they become visible.

[0445] The clustering of different experimental categories — calibration samples, bleaching experiments, and pond deployments — reveals the influence of measurement conditions and biological states on spectral patterns. The clear separation between these categories demonstrates that PCA can distinguish between normal biological variations and experimental artifacts, providing the foundation for quality control algorithms that could identify measurement anomalies in autonomous monitoring systems. In the future, machine learning will provide a more universal statistical analysis independent of PCA space.

[0446] Environmental Interference Mechanisms and Correction Strategies. The systematic characterization of environmental interference effects reveals the physical mechanisms underlying measurement variability while establishing the boundaries of reliable sensor performance under field conditions. The wavelength-dependent sensitivity to ambient Attorney Docket No. 10975-081W01

[0447] light interference reflects the spectral characteristics of solar radiation and validates theoretical predictions about outdoor measurement challenges.

[0448] The minimal interference observed in the near-infrared channel confirms the strategic advantage of the 890 nm wavelength selection for continuous outdoor operation. Solar spectral irradiance decreases significantly in the near-infrared region compared to visible wavelengths, while atmospheric scattering effects that contribute to background light interference also diminish at longer wavelengths. This natural filtering effect enables reliable measurements during daylight operation that would be impossible with conventional 750 nm approaches without sophisticated background subtraction algorithms.

[0449] Identifying voltage fluctuations as a more significant error source than ambient light interference highlights the importance of power management in autonomous sensor systems. The systematic patterns observed during deployment cycles suggest that voltage variations affect LED brightness and sensor response in predictable ways that could be corrected through algorithmic compensation. However, the magnitude of these effects relative to biological signals indicates that hardware improvements might provide more robust solutions for commercial deployment applications.

[0450] The predictable nature of most interference patterns enables the development of correction algorithms that maintain measurement accuracy under variable environmental conditions. The diurnal cycles observed in ambient light effects provide sufficient structure for algorithmic compensation. At the same time, the systematic voltage variations suggest that power monitoring could enable real-time correction of measurement drift. These findings establish the feasibility of autonomous operation under realistic field conditions while identifying specific engineering improvements that could enhance measurement stability.

[0451] Physiological Monitoring and Early Detection Mechanisms. The controlled degradation studies provide fundamental insights into the relationship between cellular physiology and optical properties while establishing the sensitivity limits for early detection of culture health changes. The spectral evolution observed during photosynthetic system destruction reveals the temporal sequence of physiological changes that affect optical measurements, enabling the identification of early indicators that precedes visible manifestation of culture problems.

[0452] The progressive loss of chlorophyll absorption features during bleaching reflects the systematic destruction of photosystem components, beginning with the more vulnerable accessory pigments and proceeding to the core chlorophyll molecules. The temporal sequence Attorney Docket No. 10975-081W01

[0453] of these changes provides insights into the biochemical pathways of oxidative stress while establishing spectral markers that could indicate specific types of cellular damage. The detection of spectral changes at sodium hypochlorite concentrations as low as 0.02% demonstrates sensitivity to oxidative stress levels that might occur during cultivation stress events.

[0454] While less dramatic than the visible spectrum changes, the NIR response to cell death provides essential information about the relationship between cellular integrity and light scattering properties. The 15% reduction in signal intensity despite complete photosynthetic system destruction indicates that physical biomass remains detectable even after cellular death, enabling accurate assessment of culture recovery potential and harvest timing optimization. This finding has practical implications for cultivation management, where distinguishing between reversible stress and irreversible damage determines appropriate intervention strategies.

[0455] The enhanced sensitivity achieved through sum normalization demonstrates the value of mathematical processing for amplifying subtle physiological signals. Eliminating concentration effects enables the detection of relative pigment changes that biomass variations in raw measurements might mask. This capability provides early warning of physiological stress before biomass effects become apparent, enabling intervention strategies that could prevent culture losses in commercial operations.

[0456] Implications for Design Philosophy. The comprehensive validation demonstrates that sophisticated analytical capabilities can emerge from strategic simplification when underlying physical principles are correctly understood and exploited. The multi-modal measurement approach validates the design philosophy that integrated capabilities provide greater analytical value than specialized single function sensors, while maintaining the hardware simplicity essential for practical deployment.

[0457] The successful implementation of multiple analytical methods within a single sensor platform challenges conventional assumptions about the cost and instruments required for comprehensive biological monitoring. Traditional approaches often employ separate instruments for biomass measurement, species identification, and health monitoring, increasing cost and operational complexity. The demonstration that strategic wavelength selection combined with appropriate data processing can provide integrated analytical capabilities suggests pathways for consolidating multiple monitoring functions while reducing overall system information. Attorney Docket No. 10975-081W01

[0458] The validation of discrete measurement approaches compared to high-resolution spectroscopy has broader implications for analytical instrumentation design across biological applications. The preservation of essential information content despite a dramatic reduction in spectral resolution suggests that many analytical applications may be over-instrumented relative to actual information requirements. This finding could inform the development of simplified analytical platforms that provide adequate analytical capability at low cost and are compatible with widespread deployment.

[0459] The autonomous operation capability validated through extended field trials addresses fundamental barriers preventing widespread adoption of optical monitoring in biological applications. Eliminating daily maintenance requirements and specialized operator expertise enables deployment scenarios previously considered impractical due to operational complexity. This capability has implications beyond algae cultivation for any biological monitoring application where autonomous operation provides advantages over manual sampling protocols.

[0460] Conclusion

[0461] Three fundamental achievements emerge from this work. First, NIR transmission at 890 nm is an effective universal proxy for optical density across multiple algal species, consistently achieving correlation coefficients exceeding R2= 0.95 with measurement accuracy approaching laboratory spectrophotometry standards. Second, meaningful absorbance spectra can be reconstructed from strategically selected discrete wavelength channels, creating a floating spectrophotometer that quantifies algal biomass and assesses biological state. Third, autonomous field deployment under realistic commercial conditions validates weeks of continuous operation with minimal maintenance, precisely the robustness required for industrial adoption.

[0462] The technical innovation transcends simple component integration by developing analytical frameworks that extract maximum biological information from simplified hardware. Multi-modal spectral analysis approaches--encompassing raw absorbance, sum- normalized patterns, and continuous spectral reconstruction--transform discrete measurements into a comprehensive biological assessment capability that rivals traditional laboratory instrumentation.

[0463] Strategic wavelength selection represents an engineering insight facilitating practical deployment. The 890 nm selection minimizes absorbance interference while maintaining biological sensitivity, facilitating continuous outdoor operation impossible with conventional Attorney Docket No. 10975-081W01

[0464] 750 nm approaches. Combined with the AS7341 multi-channel sensor platform, sophisticated analytical capability emerges from careful engineering rather than expensive specialized components.

[0465] Species discrimination through eight-channel spectroscopy validates principles with broader implications for analytical instrumentation design. Essential biological information content can be preserved with dramatically simplified hardware when wavelength selection targets key spectral features and advanced data processing compensates for reduced resolution. This finding suggests pathways for reducing costs across analytical applications without compromising measurement quality.

[0466] The comprehensive validation framework - spanning laboratory calibration, controlled degradation studies, and extended field deployments - establishes methodological approaches for translating academic sensor development into commercial technology.

[0467] Systematic evaluation of error sources, environmental interference, and long-term stability provides essential insights that can be overlooked in laboratory-focused research, yet are critical for practical implementation.

[0468] These results enable a comprehensive cultivation management system. Expanding species-specific calibration databases across commercially relevant strains can allow for broader deployment, while machine learning integration can automate pattern recognition for contamination detection and physiological monitoring. Sensor networks providing spatial coverage across industrial facilities represent the logical scaling from individual sensors to cultivation management systems.

[0469] The ultimate vision encompasses autonomous cultivation management where realtime biological feedback optimizes growth conditions, predicts harvest timing, and prevents culture losses through early intervention. This progression from monitoring to control represents a natural evolution enabled by reliable and comprehensive biological data streams, as demonstrated in this research.

[0470] Broader implications extend beyond algae cultivation to establish principles for bioprocess monitoring that reconcile analytical requirements with deployment constraints. The demonstrated approach provides a framework applicable wherever sophisticated biological measurement must operate under challenging environmental conditions with economic limitations.

[0471] Perhaps most significantly, no existing system successfully integrates autonomous deployment and comprehensive biological assessment at commercial scale. This Example Attorney Docket No. 10975-081W01

[0472] demonstrates that sophisticated monitoring does not require prohibitive costs or complex infrastructure.

[0473] Example 2. Further Investigations of Example Devices and Systems.

[0474] The devices described herein can detect both the biomass concentration and the cellular health of algae through measurement of light absorbance and color. Absorbance intensity is indicative of biomass concentration; whereas stress, starvation, contamination, or disease can impact the color of algal cells grown phototrophically. The devices can monitor more than one visible wavelength to provide information on the state (or health) of the cultivation.

[0475] The device can function as an in-situ photometer that can be deployed in open raceway ponds used for research and commercial cultivation of algae. In some instances, the device can monitor commercial ponds spanning 1 acre in surface area or greater.

[0476] A schematic of an example device is shown in Figure 16. As shown, the optical probe portion of the device is placed within an algae pond, with the ability to make continuous measurements during cultivation. The optical probe can be fixed at a specific pond depth for consistent and repeatable measurements. The data collected by the algae cultivation device can be transmitted wireless for datalogging and monitoring. Measurements can be interfaced with a control system to provide active and real time feedback to control the algae cultivation. For example, the algae cultivation device can provide for automation of pond pump control.

[0477] As shown in Figures 17A-17E, the device can include a white LED as a light source that is directed through a region of the algal culture to a red-green-blue color sensor. An Arduino microcontroller (or any other controller) is used to control the input light and to collect the output of the sensor module. An HC-05 Bluetooth module (or any other Bluetooth module) is used for wireless communication. The casing can be 3D-printed using PLA filament. However, it is envisioned that other methods (including, but not limited to, machining, casting, tooling, or molding) and materials (including, but not limited to, other plastics, polymers, or composites) may be utilized. The device can be powered by a battery pack.

[0478] The device can include an infrared (IR) light source and infrared detector. IR wavelengths are insensitive to cell color and can be utilized to accurately determine / monitor algal cell concentration. Furthermore, cell concentration data can be used to correct color signal data collected by the device Attorney Docket No. 10975-081W01

[0479] A radio system for data transmission can be incorporated to allow the device to transmit data over longer distances, allowing for the device to operate at commercial distances from a data collection or monitoring station.

[0480] Testing of the device indicated that there is a linear relationship between the total absorbance (all sensor channels; Figure 18 A) and IR absorbance (Figure 18B) measured by the device with standard measurement in a benchtop spectrophotometer. Sensor data were corrected utilizing a background light removal algorithm. Notably, the device exhibited a larger linear measurement range than the benchtop spectrophotometer. In addition, as shown in Figure 19, a similar linear response to total absorbance was demonstrated using only the green channel in the dark and as well as in the light, indicating lower sensitivity to ambient light than is the total absorbance.

[0481] Biomass concentration can be monitored by the device as shown in Figure 20. During the cultivation, portions of the culture volume were harvested periodically and replaced with fresh medium, leading to decreases in cell concentration which were accurately tracked by the device.

[0482] As noted previously, algal cultivation health monitoring can be monitored by evaluating changes in color. Algae can lose color upon infection with pathogenic bacteria, fungi, and predators (e.g., rotifers or others). Figures 21-23 show some results of channel comparisons and concentration ranges for various color sensors using the device.

[0483] Notably, different algae have different colors and through the monitorization of various colors, more than one algal population can be monitored, as shown in Figure 24. This example indicates that the device can detect clear differentiation between red, green, and blue sensor signals. Even more, such differentiation can detect algal cultivation contamination as seen in Figure 25. This example indicates the device can detect clear differentiation of sensor signals between normal and contaminated cultures.

[0484] In some cases, algal cultivation health data transmitted to a monitoring station can provide monitors with real-time interpretations of the data. The interpretations can include low nitrogen levels, low carbon dioxide levels, changes (both low and high levels) that indicate contamination by other algal species, among others.

[0485] Two field deployments have been completed using the devices described herein. We have also developed automated algal cultivation system utilizing these devices. In these trials, the sensors were able to demonstrate good capabilities for biomass approximation, detect Attorney Docket No. 10975-081W01

[0486] subtle changes in algal cultures, provide for consistent measurements, and early indications suggest that machine learning and health indication can be found.

[0487] Daily Harvest

[0488] Figure 2.6 illustrates an example automated harvest system including the devices described herein. A flow diagram illustrating the automated harvest system as well as the process employed is shown in Figure 27.

[0489] Figure 28 is a plot showing the sensor-measured biomass expressed as TOC values. The set points are 80, 60, 115, and 110 mg / L in chronological order. The last two set points, 110 and 115 mg / L, indicate the current capabilities after minor adjustments to the algorithm, which increase consistency and accuracy. As shown Figure 28, the system could effectively be used for automated algal cultivation and harvest.

[0490] Field Deployment

[0491] Field deployment in Arizona tested sensor performance with Picochlorum, a smaller organism with different scattering properties than Nannochloropsis. Daily morning harvest to a constant level created repetitive growth cycles, enabling assessment of day-to-day reproducibility and investigation of cell-size-dependent optical signatures

[0492] Figure 29 is a plot illustrating biomass tracking through NIR absorbance measurement. Near-infrared measurements reveal significant cellular respiration patterns not observed in Nannochloropsis deployments. The smaller cell size of Picochlorum alters scattering characteristics, resulting in NIR signals that capture both biomass and metabolic state. Daily oscillations reflect the balance between photosynthetic growth during the day and respiratory carbon consumption at night. Error bars quantify measurement variability, demonstrating increased interference in outdoor cultivation while remaining accurate.

[0493] As shown in Figure 30, for Picochlorum, yellow-green wavelength absorbance (555nm) provides a lower standard deviation in measurement than NIR scattering. This observation highlights the importance of multi -wavelength capability: optimal measurement channels vary by species morphology. The sensor's ability to select appropriate channels post-deployment enables adaptation to different cultivation systems without hardware modifications.

[0494] Direct comparison between NIR-based and absorbance-based biomass proxies reveals differences in channel performance for Picochlorum. The NIR channel may capture additional information - potentially cellular respiration products - that absorbance Attorney Docket No. 10975-081W01

[0495] measurements miss. This species-specific optimization demonstrates the value of multichannel approaches.

[0496] Periodic sampling with benchtop spectrophotometry validates that both NIR and yellow-green channels provide linear responses despite their different biological origins. Yellow-green channel's tighter correlation suggests absorbance-based measurement advantages for small-cell organisms where scattering efficiency is reduced. Nevertheless, both approaches maintain linearity, ensuring reliable quantitative monitoring.

[0497] When comparing both channels, the yellow-green (555nm) channel followed benchtop OD 750 more closely. It is possible that there are differences in the information that absorbance can capture. For especially small organisms like Picochlorum, the access to multiple channels still provides invaluable information.

[0498] Overlaying NIR-derived biomass with sunlight intensity (from location weather data) reveals a relationship: biomass increases follow photon flux with a delay corresponding to photosynthetic processing time. On the final measurement day with very low sunlight, the NIR channel captures lack of growth in real-time, demonstrating the sensor's ability to detect environmental limitation immediately. This real-time feedback enables rapid operational adjustments in response to suboptimal conditions.

[0499] The ratio of blue absorbance (445 nm) to yellow-green absorbance (555 nm) functions as a chlorophyll indicator normalized for biomass. Under typical conditions, this ratio tracks sunlight availability: high during photosynthetic activity, low during respiration. However, the anomalous increase on the final low-light day suggests either physiological stress response (chlorophyll retention despite low productivity) or biofouling interference. This metric provides a health indicator that operates independently of absolute biomass values.

[0500] A temporal overlay of the chlorophyll proxy (absorbance ratio) and biomass estimation revealed their phase relationship. In both the yellow-green absorbance channel and NIR channel, chlorophyll peaks precede biomass peaks, consistent with photosynthetic induction preceding cell division. Yellow-green absorbance shows similar dynamics to the chlorophyll ratio, confirming its utility as a biomass-normalized chlorophyll indicator. The consistency of this pattern across days demonstrates reproducible physiological responses that could inform harvest timing optimization.

[0501] Texas Deployment (Culture Crash)

[0502] Pilot studies at an algae cultivation facility in Texas provided an unanticipated but valuable opportunity to document culture failure progression and distinguish healthy from Attorney Docket No. 10975-081W01

[0503] unhealthy cultures. The sensor was deployed across three ponds, capturing a culture crash in Pond 1 and recovery after transfer to healthy Ponds 2 (briefly) and 3.

[0504] A visual comparison of device deployment in healthy Pond 3 versus declining Pond 2 showed a visible discoloration indicating physiological distress. Environmental particulates present in outdoor systems add complexity to optical measurements, yet the sensor successfully differentiates culture health status despite this interference. This demonstrated robustness under non-ideal commercial conditions.

[0505] A deployed sensor was used to progression through three ponds over 12 days (Figure 32). In Pond 1, downward trend preceded operator-initiated dilution, indicating early stress detection capability. Post-dilution, the culture failed to recover - unusual behavior that suggested irreversible damage. Biomass decreased during daylight hours (respiration exceeding photosynthesis) but paradoxically increased at night, possibly reflecting bacterial contamination or metabolic dysfunction. After 12 hours of data loss, declining trend resumed. Transfer to Pond 2 (also experiencing failure) showed no improvement, prompting final transfer to Pond 3 at higher stable concentration.

[0506] Focused examination of Pond 1 data revealed increased measurement noise (higher standard deviation) and inverse diurnal behavior: biomass dropped during photosynthetic periods. The yellow-green channel mirrored the NIR trend exactly, ruling out species-specific measurement artifacts. This inverse behavior provided immediate diagnostic information distinct from simple low-biomass conditions.

[0507] The 445 / 540 nm health ratio showed expected daytime upticks and nighttime decreases, but with substantially elevated noise levels. Stress-induced variability in pigment content causes ratio instability, making trends difficult to resolve on short timescales.

[0508] Nevertheless, comparing ratio values to those from healthy cultures at equivalent TOC revealed the fundamental difference: healthy cultures maintained higher, more stable ratios. This strategy provides immediate diagnostic information distinct from simple low-biomass conditions.

[0509] When biomass proxy and health ratio are plotted together for Pond 1 and entire run, their almost perfect inverse correlation becomes apparent. In healthy cultures, these metrics typically show positive correlation during growth. This inverse relationship serves as a diagnostic flag: chlorophyll is degrading faster than biomass is declining, possibly indicating cellular dysfunction or contamination. Pond 3 does not show much improvement; however, there is minor growth during the day. Attorney Docket No. 10975-081W01

[0510] Comparing reconstructed spectra from early deployment (Pond 1) to late deployment (Pond 3) revealed significant differences. High concentrations were only present in Pond 3, accounting for large peak spectral lines. Unhealthy pond spectra are flat with minimal peak structure both day and night, indicating a degraded photosynthetic system. When transferred to Pond 3, clear chlorophyll peaks appear immediately. Although a difference in spectral magnitude is expected due to concentration changes, the overall shape should stay the same, which is not observed here. Additionally, there is a notable difference in nighttime versus daytime behavior of algae from unhealthy Pond 1.

[0511] Direct comparison between unhealthy culture and healthy culture at matched total organic carbon levels (-20-60 mg / L) shows healthy cultures exhibit 3-5x larger spectral peaks. Chlorophyll peaks differ dramatically, providing a possible health assessment.

[0512] Comparison to healthy algae at similar concentration prevents misinterpretation of low- spectral signals. Tracking expected spectra can flag these events.

[0513] The devices, systems, and methods of the appended claims are not limited in scope by the specific compositions, articles, and methods described herein, which are intended as illustrations of a few aspects of the claims. Any devices, systems, and methods that are functionally equivalent are intended to fail within the scope of the claims. Various modifications of the devices, systems, and methods in addition to those shown and described herein are intended to fall within the scope of the appended claims. Further, while only- certain representative compositions, articles, devices, systems, components, and method steps disclosed herein are specifically described, other compositions, articles, devices, systems, components, and method steps also are intended to fall within the scope of the appended claims, even if not specifically recited. Thus, a combination of steps, elements, components, or constituents may be explicitly mentioned herein, however, other combinations of steps, elements, components, and constituents are included, even though not explicitly stated. The term “comprising” and variations thereof as used herein is used synonymously with the term “including” and variations thereof and are open, non-limiting terms. Although the terms “comprising” and “including” have been used herein to describe various embodiments, the terms “consisting essentially of’ and “consisting of’ can be used in place of “comprising” and “including” to provide for more specific embodiments of the invention and are also disclosed. Other than where noted, all numbers expressing geometries, dimensions, and so forth used in the specification and claims are to be understood at the very Attorney Docket No. 10975-081W01

[0514] least, and not as an attempt to limit the application of the doctrine of equivalents to the scope of the claims, to be construed in light of the number of significant digits and ordinary rounding approaches.

[0515] Unless defined otherwise, all technical and scientific terms used herein have the same meanings as commonly understood by one of skill in the art to which the disclosed invention belongs. Publications cited herein and the materials for which they are cited are specifically incorporated by reference.

Claims

Attorney Docket No. 10975-081W01WHAT IS CLAIMED IS:

1. A device for monitoring an algal population in water, the device comprising:(a) a housing;(b) an optical measurement system coupled to the housing and comprising:(i) a light source configured to emit light at one or more wavelengths;(ii) optical windows defining a measurement path from the light source and passing through a sample of the water comprising the algal population when the device is deployed in the water; and(iii) a sensor configured to detect light transmitted from the light source along the measurement path; and(c) a controller operatively connected to the light source and the sensor, the controller configured to control light emission from the light source and acquire measurement data from the sensor.

2. The device of claim 1, wherein the housing comprises a buoyant housing configured to float on a surface of the water.

3. The device of any one of claims 1-2, wherein the housing comprises a body portion with a first measurement leg and a second measurement leg extending downward from the housing.

4. The device of claim 3, wherein the body portion comprises a cylindrical or tapered cylindrical shape.

5. The device of any of claims 3-4, wherein the first measurement leg and the second measurement leg house the optical measurement system.

6. The device of any one of claims 3-5, wherein the controller is disposed within the body portion.

7. The device of any one of claims 3-6, wherein the first measurement leg and the second measurement leg are adjustable to accommodate deployment depths of the opticalAttorney Docket No. 10975-081W01measurement system of from 1 inch to 24 inches below the surface of the water, such as from 1 inch to 12 inches below the surface of the water or from 1 inch to 6 inches below the surface of the water.

8. The device of any one of claims 1-7, wherein the optical windows comprise a material that provides for optical transmission across ultraviolet, visible, and near-infrared wavelengths.

9. The device of claim 8, wherein the optical windows comprise quartz.

10. The device of any one of claims 1-9, wherein the measurement path has a length of from 0.5 cm to 5 cm, such as approximately 1.0 cm.

11. The device of any one of claims 1-10, wherein the controller is further configured to calculate an absorbance value at one or more wavelengths from the measurement data from the sensor.

12. The device of any one of claims 1-11, wherein the device further comprises a communication module operatively connected to the controller and configured to transmit measurement data from the sensor, measurement data processed by the controller, or a combination thereof to a data collection platform.

13. The device of claim 12, wherein the communication module comprises a wireless transceiver configured for long-range data transmission.

14. The device of any one of claims 1-13, wherein the one or more wavelengths comprise one or more near-infrared wavelengths, one or more visible wavelengths, or a combination thereof.

1. The device of claim 14, wherein the one or more wavelengths comprise one or more near-infrared wavelengths and one or more visible wavelengths.

16. The device of claim 15, wherein the light source comprises one or more light emitting diodes (LEDs).Attorney Docket No. 10975-081W0117. The device of claim 16, wherein the light source comprises a white LED configured to emit broadband illumination across the visible spectrum and an infrared LED configured to emit light at a wavelength of from 800 nm to 950 nm.

18. The device of any one of claims 1-17, wherein the sensor comprises a multi-channel spectral sensor.

19. The device of claim 18, wherein the multi-channel spectral sensor is configured to detect light at two or more wavelengths of from 400 nm to 700 nm and one or more wavelengths of from 850 nm to 950 nm.

20. The device of claim 19, wherein the controller is configured to perform a three-phase measurement protocol comprising a background measurement phase performed with the light source disabled, a visible spectrum measurement phase with the white LED activated, and a near-infrared measurement with the infrared LED activated.

21. The device of claim 19, wherein each measurement phase includes a thermal stabilization delay and collection of a plurality of absorbance readings at one or more wavelengths.

22. The device of any one of claims 1-21, wherein the device further comprises a battery configured to power the device for at least 12 hours, such as at least 24 hours, at least 72 hours, or at least one week.

23. The device of claim 22, wherein the controller is configured to track battery status and provide low-power warnings.

24. The device of any one of claims 1-23, wherein the device further comprises a photovoltaic cell providing power to one or more component of the device.Attorney Docket No. 10975-081W0125. The device of any one of claims 1-24, wherein the controller is configured to calculate sum-normalized spectral data by dividing each wavelength measurement by total visible spectrum absorbance to enable species identification independent of biomass concentration.

26. A method for monitoring an algal population in water, the method comprising:deploying the device of any one of claim 1-25 in the water,emitting light at one or more wavelengths from the light source;detecting light transmitted from the light source along the measurement path passing through sample of the water comprising the algal population using the sensor; and analyzing the detected transmitted light to determine information regarding the algal population in the water.

27. The method of claim 26, wherein analyzing the detected transmitted light comprises determining absorbance values at one or more wavelengths from the detected transmitted light.

28. The method of claim 27, wherein determining absorbance values comprises performing background subtraction to remove ambient light and sensor dark current effects.

29. The method of claim 28, wherein performing background subtraction comprises collecting measurements with the light source disabled and subtracting averaged background values from sample measurements.

30. The method of any one of claims 26-29, wherein the one or more wavelengths comprise one or more near-infrared wavelengths;analyzing the detected transmitted light comprises determining absorbance values at one or more near-infrared wavelengths and determining an algal biomass concentration in the water from the absorbance values at one or more near-infrared wavelengths.

1. The method of any one of claims 26-30, wherein the one or more wavelengths comprise one or more visible wavelengths;analyzing the detected transmitted light comprises determining absorbance values at one or more visible wavelengths and determining spectral characteristics from the absorbance values at one or more visible wavelengths.Attorney Docket No. 10975-081W0132. The method of claim 31, further comprising performing sum normalization by dividing the absorbance value at each of the one or more visible wavelengths by total visible spectrum absorbance to allow for species identification independent of biomass concentration.

33. The method of claim 32, further comprising identifying algal species in the water based on sum-normalized spectral patterns.

34. The method of claim 31, further comprising detecting contamination events by comparing the spectral characteristics to reference spectral patterns for healthy cultures.

35. A system for monitoring an algal population in a body of water, the system comprising:one or more of the devices of any one of claims 1 -25;a data collection platform configured to receive transmissions from the one or more devices of any one of claims 1-25; anda processing system operatively connected to the data collection platform and configured to analyze received measurement data from the one or more sensors, measurement data processed by the one or more controllers, or a combination thereof.

36. The system of claim 35, wherein the one or more devices of any one of claims 1-25 are deployed in an open raceway pond.

37. The system of claim 35, wherein the one or more devices of any one of claims 1-25 are deployed in natural body of water.

38. The system of any one of claims 35-37, wherein the processing system is configured to analyze received measurement data from the one or more sensors, measurement data processed by the one or more controllers, or a combination thereof to determine an algal biomass concentration in the body of water, assess algal health in the body of water, identify an algal species in the body of water, detecting contamination in the body of water, or any combination thereof.Attorney Docket No. 10975-081W0139. The system of claim 38, further comprising an automated control system operatively connected to the processing system and configured to adjust cultivation parameters in the body of water based on the determined algal biomass concentration, the assessed algal health status, the identified algal species, the detected contamination, or a combination there.

40. The system of claim 39, wherein the automated control system is configured to control pond pump operation, harvest timing, or a combination thereof based on the determined biomass concentration.

41. The system of any one of claims 35-40, wherein the system comprises two or more devices deployed at different locations in the body of water that each transmit measurement data from the one or more sensors, measurement data processed by the one or more controllers, or a combination thereof to the data collection platform, to each other, or a combination thereof.

42. The system of claim 41, wherein the processing system is configured to analyze received measurement data and correlate the received measurement data with a location of each of the one or more sensors to obtain information regarding the algal population at different locations in the body of water.

43. The system of claim 41, wherein the processing system is configured to analyze received measurement data and analyze the received measurement data to obtain a collective understanding of the algal population in the body of water.

Images