THE NEAR-SURFACE LAYER OF THE OCEAN
convenient integral parameter for characterizing the bubble and related
sound generation (Gemmrich and Farmer, 1999).
field consists of identifying the source of the sound, and then quantifying it.
Nystuen and Selsor (1997) identify the following four ocean surface features
producing distinctive features in the sound spectrum from 1-50 kHz: wind,
drizzle, heavy rain and ambient bubbles present (Figure 7-13). Once the
acoustic classification is done, several algorithms are available to quantify
wind speed and precipitation.
a) Acoustical wind speed measurements
Vagle et al. (1990) developed an empirical algorithm for the acoustic
quantification of wind speed using 8 kHz data from the North Atlantic
Ocean. The sound level is empirically related to 10 m height wind speed by
the following formula:
/ 20
10
10
104.3 / 53.91
x
SL
U
,
(7.13)
where 8
SL is the sound level at 8 kHz, and U 10 is the 10 m wind speed. This
method is valid for wind speeds U 10 > 2.2 m s
-1 (no wind-wave breaking and
practically no acoustic signature of the wind field is available in the ocean
under very low wind speed conditions). The algorithm, however, is not
capable of providing accurate wind speed data under rainy conditions
b) Acoustic rain rate measurements
Several types of acoustic rainfall rate algorithms are available in the
literature. An acoustic estimate of rainfall rate can be made from an
empirical relationship between the sound level within a particular frequency
band and the rainfall rate. For instance, Nystuen (2001) derived the
following relationships:
10
2 5
log
51.5 /17
P SPL
(7.14)
10
4 10
log
50 /17
P SPL
(7.15)
that were tuned to match the total rainfall accumulation from a disdrometer.
Here SPL 2-5 and SPL 4-10 are the sound levels in the frequency bands 2–5 kHz
and 4–10 kHz in decibels relative to 1 PPa
2 Hz
-1 , respectively. While these
474
As stated in Section 7.4.6, inversion of the underwater ambient sound
convenient integral parameter for characterizing the bubble and related
sound generation (Gemmrich and Farmer, 1999).
field consists of identifying the source of the sound, and then quantifying it.
Nystuen and Selsor (1997) identify the following four ocean surface features
producing distinctive features in the sound spectrum from 1-50 kHz: wind,
drizzle, heavy rain and ambient bubbles present (Figure 7-13). Once the
acoustic classification is done, several algorithms are available to quantify
wind speed and precipitation.
a) Acoustical wind speed measurements
Vagle et al. (1990) developed an empirical algorithm for the acoustic
quantification of wind speed using 8 kHz data from the North Atlantic
Ocean. The sound level is empirically related to 10 m height wind speed by
the following formula:
/ 20
10
10
104.3 / 53.91
x
SL
U
,
(7.13)
where 8
SL is the sound level at 8 kHz, and U 10 is the 10 m wind speed. This
method is valid for wind speeds U 10 > 2.2 m s
-1 (no wind-wave breaking and
practically no acoustic signature of the wind field is available in the ocean
under very low wind speed conditions). The algorithm, however, is not
capable of providing accurate wind speed data under rainy conditions
b) Acoustic rain rate measurements
Several types of acoustic rainfall rate algorithms are available in the
literature. An acoustic estimate of rainfall rate can be made from an
empirical relationship between the sound level within a particular frequency
band and the rainfall rate. For instance, Nystuen (2001) derived the
following relationships:
10
2 5
log
51.5 /17
P SPL
(7.14)
10
4 10
log
50 /17
P SPL
(7.15)
that were tuned to match the total rainfall accumulation from a disdrometer.
Here SPL 2-5 and SPL 4-10 are the sound levels in the frequency bands 2–5 kHz
and 4–10 kHz in decibels relative to 1 PPa
2 Hz
-1 , respectively. While these
474
As stated in Section 7.4.6, inversion of the underwater ambient sound