Photo © 2013 S. Beyer


Generally there are two ways of experiencing underwater fluorescence, either with invisible (ultraviolet) or with visible (usually blue) excitation light.

In the case of ultraviolet excitation light (with wavelengths <400 nm), no filters are needed for your mask and camera. You may need a "Wood's Glass" type of excitation filter on your torch, though, in case you use a torch with white light (or a strobe), which should have lots of output in the ultraviolet range of the spectrum, such as HID (high intensity discharge) lights.

The same can be done with a white light (or strobe) and a blue excitation filter in order to obtain a blue excitation light, by the way.

However, using a white light in combination with an excitation filter that will throw away most of the torch's light output is not as efficient as using a torch which has the desired range of wavelengths in the first place - see also further below.

On the other hand, such a setup may have the advantage to be able to switch between white light and excitation light for fluorescence very quickly, during the dive, obviating the need to carry and to handle another torch with white light.

Note however that this can also easily be solved the other way round, by using detachable phosphor filters which transform blue light into white light; see phosphor filters.


The photo at the top right shows red fluorescence (of a cup sponge) stimulated with green excitation light and captured with a red camera filter. Note that most red fluorescence in nature stems from chlorophyll, the molecule which performs photosynthesis in plants (e.g. algae) and turns them green under white light conditions.

Ultraviolet versus blue excitation light:

Even though using invisible excitation light has the big advantage that usually no filters are needed at all, it has been discovered that visible light (most prominently blue light, usually with wavelengths around 450-470 nm) is MUCH more effective in stimulating fluorescence of Green Fluorescent Protein (GFP) than ultraviolet light (by about a factor of four, with light of the same energy).

Another aspect to consider is the fact that ultraviolet light emitting diodes (LEDs) are MUCH more expensive than blue LEDs (by about a factor of four, for the same nominal electrical power, in Watt), and MUCH less efficient, in terms of light output or "radiant flux" (also by roughly a factor of four, measured in Watt or milli-Watt).

This means that an ultraviolet LED is about 16 times less efficient to excite fluorescence in GFP than a blue LED of equal nominal electrical power, for about 4 times the price.

In other words, it would cost you about 64 times more to obtain the same results with UV LEDs as those obtained with blue LEDs.

Even worse, the beam of ultraviolett light is quickly transformed into visible light, as shown in the photo below, due to the fluorescence of omnipresent organic matter dissolved in the water, mostly fulvic and humic acids from decaying organic matter, runoff from land, etc. (it should be noted by the way that silt will generally not fluoresce to any significant degree).

This means that the beam of UV light quickly becomes useless after a short distance (as can be seen in the photo below, the beam fades away completely), which means that one has to get extremely close (less than a meter/yard) to any fluorescent subjects in order to obtain any significant fluorescence effect.

[UV Beam]
Beam of UV light under water, visible due to dissolved organic matter

Note that depending on the characteristics of the fluorescent pigments to be excited, such as pigments which fluoresce in red, still longer wavelengths of light, such as in the green range of the spectrum (with wavelengths around 530-550 nm), may be needed in order to excite their fluorescence most effectively. But then you will only see red fluorescence, and no other fluorescent colours.

Blue light is therefore the best solution in order to see the most fluorescent colours with the highest efficiency.

But other wavelengths for excitation lights are possible, there is no single "correct" wavelength. The results to be seen will differ, though. For photographers and filmmakers for instance the use of other wavelengths opens a wide field for experimentation.

When using blue light (e.g. 450-470 nm), you will definitely need (yellow) filters for your mask and camera in order to separate excitation from fluorescent light (also known as "emission light"). Otherwise you (and your camera) would only see blue, because fluorescence is a comparatively weak effect, and the exciting blue light would simply outshine the weak fluorescent light.

For more information, see also the explanation about filters below.

Beware however that shining powerful UV radiation into your eyes (or your buddy's) may cause severe harm to your or your buddy's eyesight, especially because this radiation is invisible and therefore the corneal reflex (blink reflex) and the pupillary reflex do not work.

So powerful ultraviolet lights might not be something you would want to give to inexperienced divers to handle, after all.

Strong blue light (as any strong light) can also cause damages, by the way, but since you will immediately feel a strong discomfort this is somewhat less likely to occur.

Research into the effects of UV or blue light on marine life is still outstanding. It is known that light at night in general may disrupt the circadian rhythm of sleeping animals, or may even disrupt their reproductive cycles. Parrot fish for instance produce a "sleeping bag" made of mucus. If woken up and caused to flee, they will not be able to produce another one and will spend the rest of the night unprotected from predators and parasites. Direct damage from the strong light of the UV or blue light torches is a suspected possibility, but the energies involved are much lower than those of the ultraviolet and blue radiation from the sun during the day, at least at shallow depths, especially when considering the usually very short times of exposure of the organisms to the UV or blue light torches of a few seconds, or at most a few minutes. Moreover, the torches used for fluo diving are no match by far for the high-power lighting systems used by reef tank owners which are much brighter, used at much closer range, and burn for many hours, every day. This is common practice since 1959. If this was harmful to their marine organisms, reef tank owners would have noticed so a long time ago.

Finally, to see a comparison between the fluorescence offered by a UV source (with 365 nm wavelength) and by blue light (with the appropriate filters, of course), see also the following videos:

UV (365 nm)
UV (365 nm)
Blue light

There is no video yet to demonstrate this, but there are actually corals which do not fluoresce under UV light at all, while they do under blue light. This is presumably due to the fact that underwater organisms have evolved to adapt to the properties of water, which is more transparent to blue light than it is to UV (or to any other colour than blue, for that matter).

As a conclusion, we always recommend to use blue light, unless the use of filters is out of the question for some very important reason, or if you want to experiment, e.g. to take a different kind of fluorescence pictures (see also images 1, 2, 3, 4 from our photo gallery).

See also the article Why NightSea uses blue light for underwater fluorescence by Dr. Charles Mazel for some charts which illustrate how much more efficient blue light is than UV to stimulate fluorescence in GFP.


Two types of filters are needed for experiencing fluorescence with visible excitation light (usually blue): excitation filters and barrier filters.

Excitation filters are needed to discard all light from your torch above a certain wavelength, usually with a cutoff wavelength around 500 nm for blue excitation light (which typically has 450-470 nm). All of our fluorescence torches come equipped with such a filter.

This filter is usually a "dichroic" filter, i.e., a transparent substrate (usually glass) onto which several layers with different refractive indexes of predetermined thicknesses have been deposited in a vacuum chamber. These layers cause a certain range of wavelengths to pass the filter, whereas all other wavelengths are reflected. This is a major difference and advantage compared to conventional filters, which absorb the unwanted wavelengths of light. This can cause conventional filters to heat up considerably, or even to melt. Dichroic filters on the other hand do not suffer from this drawback. The most important advantage of dichroic filters however is that their spectral properties can be controlled much more precisely than those of conventional filters, allowing unprecedented fine tuning.

It may seem unnecessary to use a blue-pass dichroic filter on a torch with blue LEDs, but the filter is necessary for "trimming" the spectrum of the torch's light output in order to match and complement the properties of the yellow barrier filter (see below for more on these) very precisely. Otherwise, very faint fluorescence would be masked or outshined. The main advantage is that overall contrast is enhanced and red fluorescence is much more vivid with the filter than without. Without the filter, images look pale and lack brightness.

See also the special page about excitation filters for several pictures which will demonstrate the difference between blue light with and without excitation filter.

[Dichroic Filters]
Dichroic filters used as excitation filters

Barrier filters are needed to discard all light below a certain wavelength, usually also with a cutoff wavelength around 500 nm, in the case of blue excitation light (which typically has 450-470 nm).

Barrier filters are used in front of your dive mask and/or in front of your camera in order to filter out any light coming from your torch, either directly or through reflection. At the same time the barrier filter needs to let pass any fluorescent light coming from any fluorescent objects. Coincidentally, fluorescent light always has a longer wavelength than the excitation light used to excite the fluorescence, so this all works out very nicely to enable you to view fluorescence without being blinded by the excitation light.

In a certain way of speaking these filters make the blue excitation light invisible to your eyes, similar to invisible ultraviolet light, which can be used to excite fluorescence without using any filters, because the human eye is insensitive to it. The filters make your eyes (or the camera's sensor) insensitive to blue light as well.

Barrier filters are usually conventional filters of a predetermined colour (usually yellow for blue excitation light) usually made from plexiglas (also known as acrylic glass), or from glass.

The advantage of plexiglas is that it is easier to cut into arbitrary shapes, and that there is less risk of injury from breaking, whereas the disadvantage is that this material is more prone to scratching. However, this latter problem is alleviated by water, which usually fills the scratches under water and renders them imperceptible.

Barrier filters made from glass are therefore usually used for camera filters only.

See also the special page about barrier filters for several pictures which will demonstrate the importance of choosing the right barrier filter.

[Camera Filters]
Barrier filters for cameras

Influence and importance of the camera:

As you can see on the pages excitation filters and barrier filters, we have optimized torches, excitation filters and barrier filters for best possible results.

Our conclusion from these efforts is that the torch used, the excitation filter and the barrier filter all together form an intimately interconnected system whose components have to be carefully tuned to match and complement each other's optical properties, in order to achieve optimal results. For this reason, mixing equipment from different manufacturers is likely to give very disappointing results; see also Mixing equipment from different vendors for some illustrative images.

We therefore wondered about the importance and influence of the other remaining, very important component of the system, the camera.

We wondered, everything else kept the same, how big is the influence of the camera on the end result when trying to capture underwater fluorescence?

To answer this question we filmed fluorescent corals in Bonaire (September 2013) with three cameras at the same time, synchronously.

The cameras were an Intova SP1, a GoPro HD Hero3 Black Edition and a Nikon CoolPix P300, which used identical yellow camera filters (barrier filters).

Here are our results:

Comparison Intova SP1 (top left), GoPro HD Hero3 Black Edition (top right), Nikon CoolPix P300 (bottom)

As it turns out, the camera has about as much influence on the results as the filters and torches used!

White light with filter versus dedicated blue light:

Many people believe that a white light torch donned with a dichroic filter is sufficient to observe underwater fluorescence. While it is certainly true that you will probably be able to see SOMETHING with such a setup, there is more to fluorescence diving than meets the casual eye.

First of all, you throw away about 80% of the light output of your white light torch (see also this video for illustration). And since fluorescence is a relatively weak effect, you want to have as much power to excite the fluorescence as you possibly can. This is why throwing away about 80% of your light output is such a disaster, and something that might well make the difference between a nice crispy photo or a completely motion blurred one.

Secondly, and even more so with very strong white light torches or strobes, the images obtained with such a setup are inferior to those obtained with a dedicated blue light torch, as the following two example photos and two example videos below demonstrate:

[White light]
White light torch with dichroic filter
[Blue light]
Dedicated blue light torch (notice red fluorescence)

TillyTec LED 4500 torch head (white light) with a single dichroic excitation filter

Torch head "ProBlue" (blue light) with a single dichroic excitation filter
(Note that you can play both videos at the same time for easier comparison by clicking on the two "play" icons in rapid succession)

Note the greyish-whitish background produced by the white light torch (left), as opposed to the blue background produced by the blue light torch (right). Note also that the white light torch (left) is visibly weaker than the blue light torch (right), although both torches have about the same power and both use three high-power LEDs.

The following spectrographs of a few (white) dive lights without (left) and with (right) a dichroic filter further illustrate what was said above:

UK C8 Cannon HID:

[UK C8 Cannon HID]
UK C8 Cannon HID
[UK C8 Cannon HID with filter]
UK C8 Cannon HID with dichroic filter
As one can see, this is a rather "noisy" torch. It's a great dive light but it is marginal as a filtered fluo dive light.

UK Xenon:

[UK Xenon]
UK Xenon
[UK Xenon with filter]
UK Xenon with dichroic filter
Like the UK HID, this is an exceptional white light dive torch but as a filtered fluo dive unit, it is completely unacceptable.


[UK C4 eLED]
[UK C4 eLED with filter]
UK C4 eLED with dichroic filter
The peak blue component is towards the 465-470 nm portion of the spectrum. This might work, but notice that even the blue content is attenuated by the dichroic filter.

ScubaPro Fuego LED Light:

[ScubaPro Fuego]
ScubaPro Fuego LED Light
[ScubaPro Fuego with filter]
ScubaPro Fuego LED Light with dichroic filter
The Fuego unit has less output power than the above units, but it has more spectral content in the blue range around 430-440 nm.

Tekna 6:

[Tekna 6]
Tekna 6
[Tekna 6 with filter]
Tekna 6 with dichroic filter
This torch has the most spectral content in the blue range. Nevertheless, beware of the effect demonstrated by the two example photos/videos above.

More information:

See also the more general explanations about the science behind underwater fluorescence.

Our publications:

See our publications for more detailed information.



Excitation Filters

Barrier Filters

[Flag Counter]

© 2013-2016 FluoMedia.org