June 2025

June 2025 | Oceanography

67

Open-source/DIY fluorometers exist in the ocean sciences

space, with Chl-a fluorometers and “fluorometry-like” turbid­

ity (Matos et al., 2020) and backscattering sensors (Downing,

2006) being popular. Costs are low in most instances, though

two trends are noticeable. Either fluorometers tend to exhibit

detection limits of 0.1 μg L–1 (or 0.1 ppb) or higher (Leeuw

et al., 2013; Attivissimo et al., 2015; Park et al., 2023), which is

at least one order of magnitude worse than industrial sensors

(Park et al., 2023), or higher performing devices are configured

as benchtop units (Truter, 2015) and have not made the sacri­

fices necessary to package the technology into a form capable

of in situ deployment. The task of maintaining optical and elec­

trical performance in a small, water-tight, pressure-safe hous­

ing is not trivial, and making concessions on size/mass rules out

some of the most attractive use-cases of low-cost in situ fluo­

rometers (Dever et al., 2020; Park et al., 2023). Thus, there is a

gap in extant sensors between the advantages provided by open-

source in situ sensors and the performance provided by indus­

trial in situ fluorometers.

With this gap in mind, we introduce the PIXIE, a low-cost,

open-source, four-channel in situ fluorometer. In lab testing,

the PIXIE performs fluorometry with precision and accuracy

comparable to the sensors available on the market. The default-​

configuration PIXIE can be assembled for $1,392.75 USD with

one equipped channel. Each addition channel costs $525.25 USD,

for an average of $742.13 USD per channel when the instrument

is fully equipped.

For our work, a PIXIE unit was calibrated to demonstrate a

limit of detection (Arar and Collins, 1997; Sieben et al., 2010)

of 0.01 ppb RWT over a range 0 to ~60 ppb. The same unit was

calibrated to demonstrate a limit of detection of 0.02 ppb Chl-a

over a range of 0 to ~80 ppb. Deployed as part of a dye tracer

experiment in Halifax Harbor, Canada, (see Figure 1) to study

the near-field dispersion of RWT added to the cooling outfall of

the Tufts Cove Power Generation Station, the PIXIE was config­

ured to capture both RWT and Chl-a profiles, demonstrating its

multichannel functionality. The in situ data were checked against

discrete water samples collected in conjunction with the profiling

to assure quality, demonstrating how this low-cost, open-source

technology could assist in solving complex oceanographic tasks.

MATERIALS AND METHODS

Open-Source Fluorometer

The materials needed to assemble a PIXIE are available on

GitHub (https://github.com/KylePark0/PIXIE/tree/main), and

fall into one of three categories: documentation, firmware, or

hardware. The documentation includes a comprehensive user

guide that details the design, assembly, calibration, and opera­

tion of the device. Bills of materials (BOMs) are provided for the

mechanical and optical hardware, including vendors, and the

electrical BOM comes pre-packaged to fabricate with PCBWay

(PCBWay, Hangzhou, China). The listed optics include the com­

ponents needed to assemble any of five presets: PC, phycoeri­

thrin (PE), RWT, Chl-a, and crude oil. CAD models for every

component, including machined and 3D-printed parts, are

included. A rendering of the PIXIE with some dimensions (see

Figure 2) is provided, in both normal and exploded views.

The PIXIE can be powered using a range of 5–20 V. It com­

municates with an external terminal or datalogger via RS-232,

while drawing an average of 45 mA during active measurement

(225 mW). Using a dedicated 12 V lithium-ion cell with a nom­

inal capacity of two ampere-hours, the PIXIE can be expected to

measure for 40 hours in 4°C waters. Its off-the-shelf components

are rated for depths of at least 500 m. The housing is composed

of anodized aluminum and borosilicate glass, allowing it to with­

stand a range of solvents used in laboratory calibration of fluoro­

meters. Acetone is used to prepare standards of Chl-a (Arar and

Collins, 1997), a nearly-neutral phosphate buffer solution (PBS)

FIGURE 1. Drone photograph

of the August 2023 Halifax

Harbor tracer release experi­

ment conducted from the div­

ing vessel Eastcom. Insets: The

PIXIE open-source fluorometer

is shown mounted to the side of

a Niskin bottle (top) and during

rhodamine water tracer (RWT)

calibration (bottom) in the lab.