How much louder is sound under
water? That is what our experiment is supposed to test. We are
testing this statement by bringing a sound sensor under water
to record data on how many volts the sound is at different lengths
in the St. Matthew's pool. Unfortunately the sound sensor encasing
filled up with water.
An interesting fact about this experiment was that the sound
sensor picked up over 127 times more sound underwater than above
water, just looking at the first peaks of the two graphs. These
results would be perfect, except, the group that did the same
type of experiment last year picked up an average multiplier
of 3324, this number was used to calculate the difference between
voltage on land to voltage in the water. Though their tests were
inconclusive, their results were not wrong, and this is the part
of the results from which we got 3324 . This means that one of
our two very different projects ( when looking at details) was
One problem that greatly affected our results was that the "watertight"
encasement that our sound sensor was in (to keep wate roff of
it) filled with around two and one half cups of water. This was
about one third of the whole encasement's area, and the sound
sensor was officially "dead" and not able to
get a reading after the first preliminary practice test was finished.
Thus, our problem: we only had one set of results so that no
averages could be recorded. The group that did the same experiment
last year got results around 26 times greater than ours. Our
results drastically changed after our first test because our
sensor died after being soaked in water and rusting.
When looking at the averages of the figures that we took from
the land and the underwater tests, we found that the underwater
average was only 7.513 times greater than the land average. This
is very distressing because as stated earlier, we got a multiplier
of just over 127 when just looking at the first two peaks of
the land and underwater graphs. This means that the peaks were
flawed. The sound sensor could have been picking up the sounds
of bubbles going to the surface and people talking above water,
so there is an ambient sound level that is a little above zero,
but there is a drastic drop in voltage from 2 to 3 feet. This
is what is throwing off our data so much. The average was being
raised greatly because the first two feet being 3.676 volts greater
than the next closest point. 3.676 is a very huge drop when the
highest point is only 5.620 volts. The drop was 65 percent of
the highest point that was exactly what was throwing off our
figures so much.
When looking at the land results, they are slightly flawed, because
the same sound level was picked up for different tests, but there
are no drastic drops or rises. When looking at our underwater
results, the graph drops to 35 percent of its original size,
this is a drastic change. When the average deviation is 86 percent
error, something is wrong. But, when we take the two first points
away we get a 57 percent error (which is not great, but it is
a lot better than 86 percent).
When you look at our results that were taken after the encasement
with the sensor in it was flooded, they are unreadable. Not only
do the results not make any sense, but the ambient sound level
was not the same every time. This was a big indicator that the
sensor was broken. The ambient level was negative sometimes and
sometimes it was much higher than any readings from any other
test. If the ambient level is not the same, it means that the
sound level in or out of the water was drastically changing.
Since we were totally quiet when we were performing the test,
the sensor was definitely broken. There are always some sounds
in or out of the water which can affect the ambiant noise level,
although there is never enough sound to effect the noise level
Since fish have evolved with ears that suit them in the water,
if they were brought above water and exposed to the exact same
sounds that they were exposed to underwater, their ears would
be very ineffective. This is becasue sounds are all magnified
underwater, so the fish's ears are used to picking up sounds
that are louder than sounds above water
(Humans, on the other hand, would have very effective hearing).
If they grew gills and lived underwater, or simply, when they
scuba dive. This is becasue their ears evolved to pick sounds
up on land, where sound is not magnified, like underwater. So
when they go underwater they have very effective hearing.
This is shown in our results because the sensor picked up much
more sound when underwater than above water, even though the
same sound was being made in both cases.
When an object makes a noise, it sends out vibrations that the
ear picks up. If sound is magnified underwater, this means that
the vibrations are magnified underwater too. This can be related
to the fact that electricity and heat also travel faster through
water compared to above water. From this, we can gather that
waves (sound, electricity, heat) all travel faster through water
than through air. We have learned that sound travels five times
faster underwater. If the sound waves travel faster through water,
then that means that sound reaches the person earlier, and not
so much of the sound is lost traveling to the person's bones.
We say bones instead of ears becasue we have two ways to hear,
through air conductivity, and through bone conductivity, air
conductivity is ineffectual underwater because there is no air
for the sound waves to travel through. Bone conductivity is the
way that we hear underwater, through vibrations in your skull
Sounds that are higher pitched on the surface are not as easily
heard underwater. The same in reverse is also true: lower pitched
sounds are more easily heard underwater than above water.
This is shown more clearly by the experiment that the group last
year did. Their foghorn was a pitch much lower than our higher
pitch quacker, so they got a multiplier that was much higher
than ours. This is becasue their low-pitched foghorn sound waves
carried so much better underwater than above water, so that the
difference between sound picked underwater than above water was
so much greater.
One thing that made our project different from last year's was
that our results were greatly affected by the fact that our sensor
died. We couldn't find an average on any our results because
we only were able to do one test before our sensor died. If we
had done more than one test we could have seen any error in the
results. The last year's groop had done multiple tests and collected
averages on both above water and under water tests.
Another thing that made our results so different from the other
group's was that we had a constant sound that we could make,
but the other group had a foghorn, that actually ran out of air.
This means that some of the tests that they did were not valid
because the same sound was not being made on a constant basis.
There are many little things that could have affected our results.
For example, bubbles, banging the tape measure on the sensor
housing, small variations in the sound coming from the quacker
and various sounds coming from in and around the pool. All of
these factors could have made the results unreliable, even though
our sensor was dead for all the tests after the practice run.