Ocean, stream or rain: choosing water sounds for winding down

Three textures, not three versions of one sound

Rain is made of many small, independent impacts. Each drop contributes a brief click and, on water, a short ringing from the bubble trapped in the splash. Research on underwater rain measurement describes these two mechanisms clearly: the impact itself, then bubble sound that is often the louder of the two. Drop size changes the result. Small drops of roughly 0.8 to 1.2 mm produce bubbles with nearly every splash and a strong signal in the 13 to 25 kHz band, medium drops of about 1.2 to 2.0 mm often trap no bubble and are surprisingly quiet, and large drops make bigger bubbles that ring as low as about 1 kHz.

Those figures describe sound recorded under the ocean surface, not rain on your bedroom window, so do not read them as a spectrum for a rain recording. The useful part is the principle: rain is not one sound, it is a statistical pile-up of small events, and the size of the drops and the surface they hit decide whether you hear a fine hiss, a soft patter, or a heavy clatter with body underneath it.

A stream is closer to continuous. Turbulent flow over rocks produces a near-constant broadband rush, with gurgles and pulses layered on top where water folds over an obstacle. The result usually sits mid-to-high in the spectrum, with more moment-to-moment variation than steady rain but no long cycle.

Ocean surf has the longest cycle of the three. A wave builds, breaks with a bright, broadband crash, then drains away over several seconds before the next one. That means the sound level rises and falls on a timescale you can consciously follow, and there are quiet gaps between waves where room noise becomes audible again. This is the single biggest practical difference between surf and the other two.

What listening studies actually found

Most careful research on water sounds comes from urban soundscape work rather than sleep science, but it is still the best evidence we have on why one water sound feels better than another.

In a 2015 study in the Journal of the Acoustical Society of America, Rådsten Ekman, Lundén and Nilsson recorded 32 fountains in Stockholm public spaces and had 57 listeners rate them for similarity and pleasantness. The sounds split into two perceptual groups: soft, variable sounds and loud, steady-state ones. Pleasantness was negatively related to overall level and positively related to temporal variability. In a follow-up where every sound was set to the same level, pleasantness was negatively related to spectral centroid, meaning brighter sounds scored worse when loudness no longer dominated the ratings. Several loud steady sounds stayed unpleasant even at equal level.

Galbrun and Ali, writing in the same journal in 2013, generated a wide range of water sounds in the laboratory and assessed them for peacefulness and relaxation against road traffic noise. On average, stream sounds were preferred to fountain sounds, which were preferred to waterfall sounds, and low sharpness with large temporal variation was preferred. The authors also reported that no acoustical or psychoacoustical parameter correlated well with individual sound preferences. That caveat matters more than the average ranking.

Jeon and colleagues, in a 2012 study of water features in urban open spaces, found sharpness — a measure related to high-frequency content — to be a dominant factor in soundscape preference, with ratings tied to adjectives like freshness and calmness.

Put together, the pattern is consistent: quieter, less bright, more varied water sounds tend to be rated as more pleasant, and thin trickles and roaring falls both tend to lose. But these were daytime listening tests in urban settings, with eyes open and attention engaged. None of them tested people trying to fall asleep, and none of them produce a ranking you should trust over your own ears.

The evidence, and where it stops

The most cited general claim about water sounds comes from a 2021 systematic review and meta-analysis by Buxton and colleagues in PNAS. Across 36 identified publications, with 18 entering the meta-analyses, natural sounds were associated with decreased stress and annoyance (g = −0.60, 95% CI −0.97 to −0.23) and improved health and positive affect (g = 1.63, 95% CI 0.09 to 3.16). Water sounds had the largest mean effect for health and positive affect (2.01, 95% CI 0.35 to 3.67).

Read the intervals rather than the headlines. The health and positive affect estimate runs from nearly zero to very large, the authors reported significant heterogeneity across all models, and only 19 percent of the reviewed studies measured traditional physiological outcomes such as heart rate, blood pressure or cortisol. This is reasonable evidence that natural sound, including water, tends to be experienced as more pleasant than noise in short exposures. It is not evidence that a rain track improves your sleep, and the review was not about sleep.

So the honest position is narrow. Choosing a water sound you like may make a room feel better to be in while you wind down. Nobody has shown that ocean beats rain for sleep, or that any of these is a treatment for insomnia, anxiety, tinnitus or ADHD. If a sound annoys you, it is not doing anything useful, whatever category it belongs to.

A short listening test you can run tonight

Comparing water sounds while reading about them rarely works. Preference shows up in the first minute of actually not paying attention, so set the test up to mimic the real situation.

Play each candidate for 60 to 90 seconds at one fixed volume, on the device and speaker you actually use at bedtime, with the lights off and your eyes closed. Do not adjust the volume between sounds. Level had a strong influence on pleasantness in the fountain study, so if you turn one up you are testing loudness rather than character.

Notice one thing only: what your attention does. Does it drift and stay drifted, or does it keep getting pulled back to a particular event — a bright crash, a gurgle, a drip that sounds almost like a voice? A sound that repeatedly recaptures attention is the wrong sound, even if you like it in the abstract.

In soothescape the water sounds sit among the 32 ambient elements, alongside the four synthesized noise colours. You get five hours of actual listening to explore everything, which is far more than a comparison like this needs, and it is an app feature rather than an Apple subscription trial. After that, six solo sounds stay free, and a lifetime unlock keeps the full set. Everything plays offline, so the test works in a bedroom with the phone on airplane mode.

Layering water with something steadier

Surf's quiet gaps are the most common reason people abandon ocean sounds at night. A layer underneath fixes this. You can run up to four layers, so try ocean with a low noise colour set quietly enough that you notice it only when the wave recedes. The waves keep their shape, the gaps stop being holes, and the overall level can stay lower than surf alone would need to be.

Rain pairs the same way, for a different reason. A steady low layer gives light rain some body and lets you keep the rain itself quieter, which helps if the recording's brightness starts to feel fine and papery after ten minutes.

Stream is usually the one that needs the least help, because it is already close to continuous. If you add to it, add something that does not compete in the same frequency region.

Stacking two bright water sources — rain over a stream, say — is tempting and often disappointing. You add level and high-frequency energy without adding much calm, which is the combination the fountain study associated with lower pleasantness. If you find a pairing that works, save it as a mix and set the sleep timer so it does not run all night.

Who each one may not suit

Ocean may not suit you if you are a light sleeper who tracks changes in sound. The slow cycle is followable by design, and some people find themselves waiting for the next wave rather than losing interest in it.

Rain may not suit you if bright, high-frequency textures feel tiring. It may also not suit shared bedrooms, since the higher-frequency content that makes rain feel airy is easier for a partner to notice than a low rumble.

Stream may not suit you if the gurgles read as something else. Intermittent bursts in flowing water can sound uncannily like speech or like a dripping tap, and once you have heard that, it is hard to unhear.

None of the three is a good answer to loud traffic. Galbrun and Ali found that for water sound to mask dense road traffic noise, it needed to be at a similar level or no more than about 3 dB below it, and that most water sounds are poorly matched to traffic's low-frequency content anyway. Playing water loudly enough to bury a busy street is not a comfortable bedroom setting. Better window seals or earplugs do that job.

On volume generally, the NIDCD notes that sounds at or below 70 A-weighted decibels are unlikely to cause hearing loss even after long exposure, while long or repeated exposure at or above 85 dBA can. Background sound for sleep should sit well under that, quiet enough to talk over comfortably.

A simple way to decide

If you want the least eventful option, start with steady rain or a stream. Both are close to continuous, which means fewer moments that can catch your attention on the way down.

If you find continuous sound flat or clinical, start with ocean and accept that you will probably want a quiet layer underneath it.

If your room is bright and reflective, favour the darker, less sharp option of whatever you are comparing. If your room is already soft and quiet, a lighter rain can work at a lower level than you would expect.

Then stop optimising. The research is consistent about average preferences and equally consistent that individual preference is not well predicted by any acoustic measure. Once you have found something you stop noticing, you have found the right one. Save it, set the timer, and leave it alone.

The soothescape App Store listing is coming soon.