Manual Vs Auto White Balance Modes

Purpose

The purpose of this guide is to help give users an understanding of Auto White Balance (AWB) and Manual White Balance (MWB) systems and their respective limitations. The goal is that users can take from these techniques to use in the field to better their use of the different modes and lighting to achieve the best results possible. 

Background

Light from the sun, and other sources such as LEDs,  is emitted across a broad spectrum of wavelengths. In a perfect vacuum, all of these wavelengths would travel without loss and remain visible to anything within line of sight of the sun. In reality, particles between the light source and the observer absorb and scatter portions of this spectrum, reducing the strength of some wavelengths or preventing them from reaching us at all. This process is called attenuation.

We see this effect in our atmosphere every day. The atmosphere filters and weakens parts of the solar spectrum, such as ultraviolet light, which helps protect us from more severe sun exposure. Heavy cloud cover causes additional attenuation: sunlight can still pass through the suspended water droplets, but much of the spectrum is scattered or absorbed, leaving the muted grey light typical of an overcast day.

The same phenomenon occurs underwater. As sunlight enters the ocean or a lake, water molecules, dissolved substances, and suspended particles absorb and scatter light. The overall intensity of light decreases with depth, and different wavelengths are removed at different rates. Longer wavelengths such as red, orange, and yellow are absorbed first, while shorter wavelengths like blue and green penetrate much deeper into the water column.

This progressive loss of colour means that the deeper a camera goes underwater, the less of the original sunlight spectrum remains and the more blue and green images appear. Understanding how light attenuates with depth is the key reason cameras rely on white balance adjustments to restore natural colour in underwater images.

Manual White Balance

There is some relatively simple image processing that can help alter the strong blue or green colour cast common in underwater images. In practice, we reduce the influence of blue and increase the influence of red in the camera’s colour processing. We are not actually removing blue light from the environment or adding red light; instead, we are independently adjusting the sensitivity of the pixels in the image sensor collecting blue and red light.

This is essentially what happens when we use Manual White Balance (MWB). In this mode, the operator visually evaluates the image and adjusts the camera’s colour response to achieve the preferred look. Some users prefer a more natural green or blue image at depth, while others prefer a balanced image that appears closer to what we would see near the surface, where a broader spectrum of light is present.

One limitation of manual white balance appears when lighting conditions change. For example, if a user dives to depth and adjusts the white balance manually, the image may look correct at that depth. However, if the ROV then returns to the surface without readjusting the settings, the image will often appear strongly pink or red. This happens because the camera was tuned to be less sensitive to blue light and more sensitive to red light. At depth, this produces a balanced image because very little red light is present compared to blue. Near the surface, where red light is abundant again, the increased red sensitivity causes the image to appear overly red or pink.

A similar effect occurs when artificial lighting is introduced at depth. Most standard LEDs produce a broad spectrum of light. If the camera has been balanced for the limited spectrum available at depth (while ROV lights are off) and high-intensity LED lighting is introduced, red wavelengths are suddenly added back into the scene. Because the camera is still highly sensitive to red, the resulting image can again appear pink or red.

Auto White Balance

Because underwater lighting conditions can change quickly with depth, movement, and the use of artificial lighting, manually adjusting white balance is not always practical. In these situations, Auto White Balance (AWB) can be a useful tool.

Auto white balance continuously analyzes the colours present in the image and automatically adjusts the camera’s colour sensitivity to maintain a more neutral result. Instead of relying on the operator to manually tune the camera at each depth or lighting condition, the system dynamically compensates for the changing balance of red, green, and blue light. This allows the camera to adapt as the ROV moves between depths or when artificial lighting is introduced, helping maintain a more consistent and usable image without constant manual adjustment.

In our software, AWB is automatically triggered whenever the LEDs are commanded on or off. This ensures that the camera adapts to the sudden addition or removal of broad-spectrum light, maintaining a more accurate overall colour balance.

Modes

Auto White Balance (AWB) can operate with different response speeds, implemented as Air and Underwater profiles. Each is optimized for different lighting conditions and operational needs.

• Air Profile – This profile averages colour over a large number of frames, making it very slow to adjust. It produces smooth, gradual changes in colour, but is limited in its ability to correct underwater scenes. Because it makes only small adjustments, Air Profile often retains a strong green/blue cast in most underwater conditions and may lag behind when lighting changes quickly. This profile is most suitable for shallow or surface scenes where subtle, slow adjustments are sufficient.
  
  

• Underwater Profile – This profile averages colour over fewer frames and allows for a much larger range of adjustments. It reacts quickly to sudden changes in lighting, such as moving between depths or when artificial LEDs are turned on or off. Water Profile is capable of neutralizing the colour balance of the scene, producing more natural-looking images in dynamic underwater environments. Because it is faster and more responsive than Air Profile, it is generally preferred for professional ROV operations. However, rapid adjustments can sometimes produce visible colour “pumping” in highly dynamic scenes.

While Underwater Auto White Balance is convenient, it has some important limitations underwater. Because AWB calculates colour balance based on the average of the full frame, small objects that are brightly lit may not appear correctly. For example, if the ROV’s LEDs are only illuminating a small target while the surrounding environment remains dominated by blue or green light, AWB will prioritize balancing the overall frame. The result can be that the small, well-lit target appears pink or red, even though it is illuminated correctly.

Operators should be aware of this effect when using localized lighting or working with small, high-contrast subjects. In these cases, manual white balance or careful framing may produce the desired colour representation.

Closing Thoughts

Understanding how light behaves underwater and how your camera responds is key to capturing clear, natural-looking images. Manual White Balance gives you full control to fine-tune colours at depth, while Auto White Balance provides a convenient way to maintain consistent colour across changing conditions. Both systems have limitations—small targets, dynamic lighting, and depth changes can all affect results—but knowing how and when to use each, along with the AWB speed profiles and LED rechecks, will help you achieve the best possible imagery.

With practice, operators can balance control and convenience, ensuring that every dive produces visually accurate and usable footage.

References

1. NASA PACE Ocean Color – Ocean Light and Underwater Attenuation. Available at: https://pace.oceansciences.org/ocean_light.cgi

2. Kirk, J.T.O. (2011). Light and Photosynthesis in Aquatic Ecosystems. Cambridge University Press.

3. Mobley, C.D. (1994). Light and Water: Radiative Transfer in Natural Waters. Academic Press.

4. Lalli, C.M. & Parsons, T.R. (1997). Biological Oceanography: An Introduction. Butterworth-Heinemann.