Agriculture Reference
In-Depth Information
Table 14.2. Avian species identified by listening to digital recordings in KBS LTER
Main Cropping System Experiment (MCSE) locations. a
Site b
Dominant Species (call
density)
Species
Richness
Number of
Vocalizations
Shannon-
Wiener Index
Frequency
with Maximum
Acoustic Power
(kHz)
A1
Song sparrow (0.68)
9
34
1.64
5
A2
Song sparrow (0.58)
13
59
2.10
3
P1
Indigo bunting (0.65)
11
27
2.09
5
P2
Song sparrow (0.69)
11
38
1.97
3
C1
Red-winged blackbird (1.00)
16
80
2.31
3
C2
Tufted titmouse (0.42)
14
46
2.46
3
C3
Northern cardinal (0.53)
9
30
1.69
3
S1
Song sparrow (0.59)
25
146
2.64
5
S2
Brown thrasher (0.41)
17
53
2.48
5
S3
Northern cardinal (0.45)
17
77
2.47
3
D1
Scarlet tanager (0.6)
12
73
2.09
3
D2
Baltimore oriole (0.44)
21
120
2.65
3
D3
Eastern wood-pewee (0.31)
25
98
2.91
3
a The dominant species was determined based on call density (i.e., the number of vocalizations for that species divided
by the total number of recordings). The normalized acoustic power density was generated in each frequency range by
slicing every 1000 Hz from 0 to 11,000 Hz. The right column is the most powerful frequency recorded at each site.
b A = Alfalfa, P = Poplar, C = Coniferous Forest, S = Mid-successional, and D = Deciduous Forest systems of the
MCSE; numbers refer to replicate locations.
intervals, (B) a wireless router to send acoustic recordings around tall vegetation
(e.g., poplar trees), (C) a local server to receive the recordings via wireless com-
munication and store the sound recordings locally, and (D) a regional server where
the sound recordings are received via the Internet from the local server and ana-
lyzed. The recordings, results from the analysis of them (normalized soundscape
power by frequency interval), and computed soundscape indices from these values
are then placed in a sound library (E) that can be accessed simultaneously by users.
The characteristics of this digital acoustic library are described further in Kasten
et al. (2012).
Early efforts using autonomous acoustic recorders in the field identified
power as the factor that limited recordings to short periods until the advent of
low-power processors. Wireless technology allowed the deployment of distrib-
uted acoustic sensors, powered by a 12-V battery charged by solar panels, that
collect sound recordings frequently (e.g., 30-minute intervals for 30-second
durations) and transmit the recorded sounds to a local server for subsequent
transfer to a remote server via the Internet. The acoustic sensor platform was
designed and developed based on the Crossbow Stargate processor (Crossbow
2006). This processor operated using Linux and required relatively low power
(~3 watt). The hardware components of the sensor platform (Fig. 14.5) included
 
Search WWH ::




Custom Search