Aquaculture 596 (2025) 741761

Contents lists available at ScienceDirect

Aquaculture
journal homepage: www.elsevier.com/locate/aquaculture

The effect of the molting cycle on the acoustic characteristics of clicks

emitted by Litopenaeus vannamei
Silvio Peixoto * , Cecilia Craveiro , Fábio Costa Filho , Hildemário Castro-Neto , Roberta Soares
Department of Fisheries and Aquaculture, Federal Rural University of Pernambuco (UFRPE), Recife, Brazil

A R T I C L E I N F O A B S T R A C T

Keywords: In penaeid shrimp, the mandibles are responsible for emitting click sounds when they collide during food intake,
Ecdysis which has been used to assess feeding activity by passive acoustic monitoring. However, the acoustic parameters
Intermolt of the clicks could be affected by variations in the thickness of mandible cuticle during the molting cycle. Thus,
Postmolt
the aim of this study was to evaluate the effects of the molting cycle on the acoustic characteristics of clicks
Passive acoustic monitoring
Acoustic-based feeders
emitted during the feeding activity of Litopenaeus vannamei fed commercial pelleted diet. Shrimp with an average
weight of 8.75 ± 1.05 g were individually maintained in 24 tanks in a recirculating water system. A total of 12
animals had their feeding activity individually recorded soon after ecdysis (postmolt group), while another 12
were recorded in the middle of their molting cycle (intermolt group). Recordings took place in anechoic
chambers with a hydrophone connected to a digital recorder (sampling rate of 192 kHz), and the click acoustic
parameters were characterized in Raven® 1.5 Pro software. After the recordings, the shrimp had their mandibles
removed for length measurement and histological analysis of the mandibular cuticle. Body (weight and length)
and mandibular measurements did not differ between shrimp groups, but those in the postmolt stage showed
significantly lower cuticle thickness. The acoustic characteristics of the clicks were affected by shrimp molting
stages, with significantly lower maximum energy (dB) and maximum frequency (kHz) in postmolt shrimp,
probably related to their lower mandibular cuticle thickness. These results may contribute to the assessment of
shrimp feeding behavior associated with the molting cycle, as well as to the optimization of algorithms con­
trolling acoustic automatic feeders in farming systems.

1. Introduction tissue mass; and intermolt (C), when the new cuticle is fully rigid,
allowing for muscle growth and the accumulation of reserves for the
Ecdysis, or molting, is the process in which shrimp shed their old next molt (Dall et al., 1990; Lemos and Weissman, 2021). Almeida Neto
exoskeleton, replacing their rigid cuticle with a new, softer, and more and Freire (2007) analyzed the texture of the exoskeleton of Litopenaeus
flexible underlying cuticle (Dall et al., 1990). During this process, the vannamei to determine the molting period. Those authors suggested that
animals absorb a large amount of water and rapidly increase in size, the relationship between stage A and soft texture could be explained by
which is accompanied by the hardening of the new cuticle. The cyclical the semi-calcified cuticle with high permeability, while the normal
shedding of the exoskeleton occurs more frequently during larval stages (rigid) texture at stage C could be due to the complete calcification of the
and early life phases, becoming more spaced out during juvenile and exoskeleton.
adult phases. The interval between successive molts can vary from 1.5 The exoskeleton of crustaceans is composed of the following layers: a
days in post-larvae to 7 days in juveniles (5 g; Lemos and Weissman, thin outer epicuticle, a thick procuticle, and the underlying epidermis.
2021), or between 5 and 11 days for shrimp weighing 2 to 15 g, The procuticle, when developed, consists of an outer exocuticle and an
respectively (Corteel and Nauwynck, 2010). The complete molting cycle inner endocuticle. The cuticle is secreted by the epidermis, a single layer
can be divided into the following stages: premolt (D), when morpho­ of cells with intense secretory activity. The composition of the cuticle
logical and physiological preparations for ecdysis occur; molt or ecdysis consists mainly of water, proteins, a polysaccharide (chitin), minerals,
(E), the moment when the old cuticle is shed; postmolt (A-B), the stage and other substances that are distributed differently among the various
immediately after molting when the incorporated water is replaced by layers. Chitin forms a web-like structure that, filled with protein,

* Corresponding author at: Department of Fisheries and Aquaculture, Federal Rural University of Pernambuco (UFRPE), 52171-900 Recife, PE, Brazil.
E-mail address: [email protected] (S. Peixoto).

https://doi.org/10.1016/j.aquaculture.2024.741761
Received 6 August 2024; Received in revised form 10 September 2024; Accepted 12 October 2024
Available online 15 October 2024
0044-8486/© 2024 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
S. Peixoto et al. Aquaculture 596 (2025) 741761

provides strength to the exoskeleton (Corteel and Nauwynck, 2010). constantly provided through an electromagnetic air compressor (2 HP;
However, mineralization with calcium carbonate has been adopted by Ibram, SP, Brazil) connected to silicone hoses with a porous stone per
most crustaceans to reinforce their exoskeleton. Typically, the most tank. The photoperiod was natural, approximately 12 h of light and 12 h
mineralized part of the exoskeleton is the mandible, which is equipped of darkness. Shrimp were fed three times a day (0800, 1200, and 1600)
with specialized dental processes, “incisors” for biting and cutting, and using commercial feed (Density 40 CR2; 40 % protein; pellet size ~1.5
“molars” for grinding and chewing. The mandibles of crustaceans mm; Presence, RN, Brazil). Water quality parameters of temperature,
exhibit elaborate biomineralization that combines various minerals ar­ salinity, and pH were monitored once daily (0800) in the recirculation
ranged in well-defined distribution patterns, specific to each species, system using a multiparameter meter (YSI 556 MPS, YSI Inc. / Xylem
reflecting their feeding habits. Evolutionarily, shrimp and other crus­ Inc., USA). Ammonia and nitrite concentrations in the system water
taceans began to include calcium phosphate to increase the hardness were analyzed every two days (0800) using a spectrophotometer (HACH
and wear resistance of their mandibles due to the high mechanical load 3900, Hach, Germany).
in this region (Bentov et al., 2016; Zhu et al., 2024). The experiment began with daily monitoring of the 24 shrimp in the
The mandibles are also responsible for emitting click-like sounds early morning to detect the first molting in the tanks and calculate the
when they clash during food intake by penaeid shrimp Litopenaeus seti­ time (in days) of the molting cycle for each animal when they molted for
ferus (Berk, 1998), Penaeus monodon (Smith and Tabrett, 2013), and L. the second time. As shrimp started molting for the second time, a group
vannamei (Silva et al., 2019; Zhu et al., 2024), as well as the caridean of 12 shrimp was randomly selected to be recorded immediately after
Macrobrachium rosenbergii (Hamilton et al., 2021). The emission of these their second molt, equivalent to the early postmolt stage or stage A
clicks has enabled the use of passive acoustic monitoring (PAM) in (postmolt group). Another 12 shrimp were recorded halfway through
studies on shrimp feeding behavior (Peixoto et al., 2020a, 2020b; Soares their next molting cycle, i.e., approximately in the middle of their sub­
et al., 2021a, Soares et al., 2021b; Peixoto et al., 2022; Hamilton et al., sequent intermolt stage or stage C (intermolt group) (Fig. 1).
2023; Tabbara et al., 2024), as well as the development of feeding sys­
tems based on passive acoustics (Napaumpaipom et al., 2013; Jescovitch 2.2. Acoustic recordings of feeding activity
et al., 2018; Ullman et al., 2019a, 2019b; Reis et al., 2020, 2021; Walsh
et al., 2022; Reis et al., 2023; Strebel et al., 2023). In aquatic environ­ Recordings for characterization of clicks emitted during feeding ac­
ments, PAM studies commonly utilize a hydrophone and a digital tivity were conducted in anechoic chambers composed of rectangular
recorder (Lindseth and Lobel, 2018). Hydrophones are omnidirectional polyethylene tanks (48x28x30 cm) lined with polyurethane foam
sensors that convert sound propagated through water into an electrical (thickness: 30 mm; density: 28 kg/m3), to minimize sound reflection and
signal. Digital recorders then capture the intensity (energy) of these resonance (Peixoto et al., 2020a, 2020b). These chambers had a useful
sounds, converting the analog signals into digital form and enabling volume of 25 L and were filled with the same water from the shrimp’s
time-based reconstruction through signal processing. The recorded data origin tanks, but without water recirculation and aeration to avoid noise
can be either stored for later analysis or processed in real-time using in the recordings. When the second molt was detected in the tanks
computational algorithms (Lindseth and Lobel, 2018; De Tailly et al., (postmolt group) or reaching the estimated time relative to the halfway
2021; Reis et al., 2022). point of the intermolt stage (intermolt group), shrimp were individually
It has been observed that mandibular size is strongly correlated with transferred in the early morning (8 h) and acclimated for 1 h in the
the acoustic parameters of clicks produced by L. vannamei ranging from anechoic chambers before acoustic recordings. A total of three chambers
0.01 to 12 g, with lower acoustic energy (dB) produced by shrimp under were available in case multiple recordings were needed on the same day.
1.16 g (Hamilton et al., 2024). Thus, we could hypothesize that the The recording equipment consisted of omnidirectional hydrophones
acoustic parameters of clicks would also be affected by variations in the AS-1 (Aquarian Audio, USA) with a linear frequency response from 1 Hz
hardness of the mandibular teeth, related to their cuticular thickness to 100 kHz (sensitivity of -208 dB re 1 V/μPa), coupled with pre­
during the molting cycle. However, no PAM studies have been con­ amplifiers PA4 (Aquarian Audio, USA) with a gain of 26 dB and con­
ducted to clarify this, nor is there information on its influence on the nected to a digital recorder F6 (Zoom Inc. USA). The recorder’s highest
detection of click energy emitted during the shrimp molting cycle by sampling rate was used (192 kHz; 16 bits), allowing capturing sounds
automated feeder algorithms. In this context, the aim of this study was to with maximum frequencies of up to 96 kHz. Acoustic recordings of
evaluate the effects of the molting cycle on the acoustic characteristics of feeding activity lasted for 30 min, counted after offering the same feed
clicks emitted during the feeding activity of L. vannamei. Acoustic pa­ (ad libitum) used during shrimp maintenance in the recirculation
rameters and cuticular thickness of the mandible were compared in system.
shrimp at intermolt (rigid cuticle) and postmolt (soft cuticle) stages. The
results may contribute to the interpretation and detection of acoustic 2.3. Acoustic characterization of clicks
signals associated with molting events and the improvement of auto­
mated feeding management based on PAM. The audio files (.wav) of feeding activity of each shrimp in the
intermolt and postmolt stages were transferred from the recorder to a
2. Material and methods high-performance computer (Intel® Core™ i5–12,400, 32 GB DDR4, 1
TB SSD 500w), where they were analyzed using Raven® 1.5 Pro soft­
2.1. Experimental design ware (Cornell University, USA). Spectrograms (Type: hann; Overlap: 50
%; Resolution: 512 samples) of the audio files were generated for
The experiment was conducted indoors at the Laboratory of Aqua­ manual identification of sound pulses (clicks), according to general
culture Technology, Federal Rural University of Pernambuco, PE, Brazil. acoustic parameters described by Silva et al. (2019) and Hamilton et al.
A group of L. vannamei with an average weight of 8.75 ± 1.05 g, total (2024) for L. vannamei. After identification, all clicks in the recording
length of 10.31 ± 0.39 cm, and carapace length of 2.31 ± 0.08 cm was were selected for their individual acoustic characterization. The selected
randomly distributed and individually kept in 24 rectangular poly­ variables to characterize the clicks were: minimum and maximum fre­
ethylene tanks (48x28x24 cm; 27 L). These tanks were connected in a quency (kHz), peak frequency (kHz), maximum energy (dB), and
water recirculation system (salinity 10), composed of a 500 L reservoir duration (ms).
tank (sump), screen filter (Preamar Aquicultura, SP, Brazil), biological
filter (Aquatic Experts, NC, USA) and skimmer (Jebo, China). Two 2.4. Analysis of acoustic energy level during feeding activity
centrifugal pumps (1 HP; Jebo, China) maintained a water recirculation
flow rate of 0.06 m3 per hour in the system tanks. Aeration was The audio files were also evaluated for the acoustic energy level (dB)

2
S. Peixoto et al. Aquaculture 596 (2025) 741761

Fig. 1. Schematic drawing of recordings of feeding activity in L. vannamei during intermolt and postmolt stages for obtaining acoustic parameters. Green circles
indicate molting events (M), and red arrows indicate the moments when recordings were made for the postmolt and intermolt groups. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)

generated during shrimp feeding activity in intermolt and postmolt were photomicrographed and digitized using Leica LAS EZ 3.4 software
stages. In this analysis, the spectrograms of the audio files were first (Leica Microsystems, Germany) with objectives ranging from 4× to
analyzed in Raven 1.5 Pro software to manually remove any additional 100× magnification. The thickness of the cuticular tissue was measured
noise during the recordings (30 min), such as water impact from the (μm) using ImageJ software for Windows (Schneider et al., 2012). The
initial feed offering or accidental impacts on the hydrophone by shrimp. measurements were taken for the total thickness of the mandibular
After this step, automatic selections with a fixed duration of 15 s were cuticle (TMC, μm) and the thickness of the endocuticular layer (TME,
made in the audio files, and the maximum energy values (maximum μm), as exemplified in Fig. 2B.
power; dB) were calculated for each time interval. These values were
used for graphical representation and comparison of maximum energy 2.6. Statistical analyses
in 15 s intervals over the 30 min of recordings for shrimp in intermolt
and postmolt stages. Normality of sample distribution and homogeneity of variances were
tested before each analysis using the Ryan-Joiner and Levene tests,
2.5. Morphometric and histological analysis of mandibles respectively. The lengths of all left and right mandibles of the shrimp did
not differ significantly when compared by the Student’s t-test, so pooled
After the recordings, shrimp in postmolt and intermolt stages were measurements of left and right mandibles were compared again between
immobilized with ice and euthanized for the removal of the pair of the intermolt and postmolt shrimp groups. The body weight and length
mandibles. The mandibles (right and left) were dissected and measured measurements of shrimp, as well as thickness of the mandibular cuticle,
under a Leica EZ4 stereomicroscope (Leica Microsystems, Germany) were also compared between the intermolt and postmolt groups using
with an integrated digital camera (0.01 mm), where measurements of the t-test. The data of the acoustic parameters of clicks and maximum
total mandible length (ML, mm) were obtained, as shown in Fig. 2A. The energy in 15-s intervals in the recordings did not show a normal dis­
mandibles were then immersed in Davidson’s saline solution (12h) for tribution and were compared between the intermolt and postmolt
demineralization and fixation, and preserved in 70 % alcohol for 24 h. shrimp groups using the non-parametric Mann-Whitney test. All ana­
Subsequently, the right mandibles were subjected to the standard his­ lyses were performed using Minitab version 18.1 software (Minitab,
tological process, being dehydrated in increasing ethanol series, cleared Inc), with a significance level of p < 0.05.
in xylene, and embedded in paraffin at 58 ◦ C.
Histological sections of the right mandibles were made using a Leica 3. Results
RM2145 rotary microtome (Leica Microsystems, Germany), sectioned at
5 μm in the transverse position, and stained with hematoxylin and eosin- The water quality parameters in the recirculation system showed the
floxin (H/E–F) for the identification of acidic and basic substances, following values: temperature 29.7 ± 1.5 ◦ C; salinity 10.2 ± 1.1; dis­
respectively. Slides were observed under an optical microscope (Leica solved oxygen 4.5 ± 0.4 mg/L; pH 8.9 ± 0.2; alkalinity 209 ± 58 mg/L;
DM500, Leica Microsystems, Germany) equipped with a digital camera ammonia 0.3 ± 0.2 mg/L, and nitrite 1.5 ± 0.5 mg/L.
(Leica ICC50HD, Leica Microsystems, Germany). Histological samples The average molting interval of the shrimp, observed during their

Fig. 2. Photomicrograph indicating measurements related to the mandibles of L. vannamei. (A) Total mandible length (ML) measured macroscopically. (B) Length
associated with the thickness of the mandibular cuticle (TMC) and endocuticle layer (TME) measured microscopically through histological sections.

3
S. Peixoto et al. Aquaculture 596 (2025) 741761

maintenance in the recirculation system, was 11 ± 4 days. The mean 4. Discussion

weight of shrimp groups in the intermolt and postmolt stages, corre­
sponding to 8.85 ± 1.22 g and 8.66 ± 0.86 g, respectively, did not show The molting cycle of shrimp occurs at intervals of days to weeks,
significant differences between these groups, nor did their measure­ unlike other decapod crustaceans (e.g., crabs, lobsters) which tend to
ments of total length and carapace length (Table 1). have longer intervals between successive molts (Dall et al., 1990).
The total length of the mandibles did not differ significantly between Intraspecific variations in the molting cycle occur based on individual
the intermolt (2.26 ± 0.12 mm) and postmolt stages (2.18 ± 0.16 mm) size and age, as well as environmental factors such as temperature,
(Table 1). The morphological characteristics of the mandibles were also photoperiod, and available space (Chang, 1995; Dall et al., 1990; Lemos
similar between these two groups of shrimp, showing a pair of flattened and Weissman, 2021). The molting cycle duration (11 days) observed
incisor teeth and well-developed denticles at the ends. during shrimp maintenance in this study falls within the characteristic
The histological analysis of the mandibles for shrimp in the intermolt range for the life phase and environmental conditions of L. vannamei
and postmolt stages showed that both presented the mandibular cuticle rearing. The molting cycle duration of L. vannamei, with an approximate
composed of three layers: the outermost layer of epicuticle, the inter­ weight of 8 g, was estimated to be between 10 and 13 days in a com­
mediate layer of exocuticle, and a thicker inner layer called endocuticle mercial farm with water temperature ranging from 28 to 31 ◦ C (Almeida
(Fig. 3). Despite the similarity in the distribution of cuticular layers, Neto and Freire, 2007). A similar molt cycle duration (10–11 days) was
significantly higher thickness values were observed in both the reported for Farfantepenaeus duorarum (4–8 g) maintained individually
mandibular cuticle and the endocuticular layer in intermolt shrimp in tanks at 28 ◦ C (Bonilla-Gómez et al., 2013). However, when L. van­
(Table 1). namei of similar size to this study (11.5 to 13 cm total length) were kept
Twenty-four recordings of feeding activity were conducted, corre­ individually in tanks with lower temperatures (20–22 ◦ C), the molt cycle
sponding to 12 shrimp in each stage of postmolt and intermolt. In these duration extended to 24 to 40 days (Chan et al., 1988). It is also worth
recordings, shrimp were acoustically active and emitted “click” sounds noting that maintaining shrimp individually in small tanks, as done in
regardless of whether they were in the intermolt or postmolt stage. this study, can also influence the molt cycle duration of crustaceans
However, a lower click emission rate was observed in the postmolt stage (Chang, 1995).
probably due to an expected decrease in feed intake during this molting Throughout the various stages comprising the shrimp molt cycle,
stage, which resulted in unbalanced sample size (i.e. number of clicks) namely premolt (D), molt or ecdysis (E), postmolt (A-B), and intermolt
between shrimp groups in the intermolt and postmolt stages. Therefore, (C), the animals undergo drastic changes in their body structure,
for the characterization of the acoustic parameters, a total of 197 clicks metabolism, and behavior (Promwikorn et al., 2004; Lemos and
in the recordings were analyzed, with 131 from intermolt and 66 from Weissman, 2021). The premolt stage, characterized by the biological
postmolt shrimp groups. Nevertheless, this unbalanced number of clicks preparation of organs and tissues required for cuticle shedding, can be
was accounted in the data analysis to ensure sufficient statistical power divided into four sub-stages (D1, D2, D3, and D4) depending on the
and accuracy, which were improved by the homogeneous variances and degree of retraction of the epidermal tissue from the cuticle. After sub-
large sample sizes (131 and 66) between shrimp groups. The acoustic stage D4, shrimp take only a few seconds to shed their old cuticle dur­
parameters of these clicks are presented in Table 2. ing the ecdysis process (stage E). Subsequently, when animals enter the
The mean values of minimum frequency were similar in clicks postmolt stage (A-B), synthesis and hardening of the new cuticle occur
emitted by shrimp in intermolt and postmolt stages, but the maximum immediately after ecdysis (A) and continue during the mineralization
frequency was significantly higher in the intermolt group (Table 2). The process of the new cuticle (B). During the intermolt stage (C) and its
peak frequency, where the maximum energy of the click is concentrated, respective sub-stages (C1, C2, C3, and C4), shrimp complete the syn­
did not show significant differences between the intermolt and postmolt thesis of the procuticle and achieve full rigidity of their exoskeleton
groups, but the energy emitted by clicks of shrimp in intermolt was (Promwikorn et al., 2004; Lemos and Weissman, 2021). Feeding
significantly higher than in the postmolt stage. The duration of the click behavior related to the molt cycle has been the main focus of research,
did not show significant differences between shrimp in intermolt and mainly due to decreases in feed intake during shrimp ecdysis. This
postmolt stages (Table 2). Fig. 4 presents examples of clicks from shrimp phenomenon is associated with the exoskeleton shedding process, which
in intermolt and postmolt stages, highlighting the differences in affects the functionality of the buccal structure, esophagus, and anterior
maximum frequency and maximum energy in their spectrograms. part of the stomach (Panakorn, 2018; Lemos and Weissman, 2021).
The mean maximum acoustic energy recorded in fixed 15-s intervals There seems to be a consensus that feeding starts to decrease in the sub-
over the 30 min of feeding activity recordings in shrimp at the intermolt stages preceding molt (D2-D3), ceasing during sub-stage D4 and stage E,
and postmolt stages can be seen in Fig. 5. The mean maximum energy and resuming from stages A and B (Dall et al., 1990; Lemos and
values in these intervals were 42.21 ± 7.73 dB for shrimp in intermolt Weissman, 2021). Similarly, feeding activity of L. vannamei (11.5 to 13
and 39.81 ± 4.95 dB in postmolt, differing significantly between these cm total length) was interrupted and resumed in these same sub-stages
two groups (p < 0.001). The 15-s intervals showed a minimum value of and stages of the molt cycle (Chan et al., 1988). Bonilla-Gómez et al.
36.3 dB (i.e., background acoustic landscape energy) in the recordings of (2013) reported that feed consumption in F. duorarum (4–8 g) declined
both shrimp groups, while the maximum energy values recorded were starting 20 h before ecdysis, intensified four hours before it, resumed
88.2 dB and 69.8 dB for shrimp in the intermolt and postmolt stages, eight hours after ecdysis, and returned to normal after about 40 h. The
respectively. authors emphasize that although feed intake decreased during molt, it
did not completely cease in both wild-caught and captive-bred in­
Table 1 dividuals. The results of this study support these observations, as shrimp
Mean values (± SD) of morphometric measurements of L. vannamei in intermolt in the postmolt stage (A-B) also showed acoustic activity through click
and postmolt stages. emissions during feeding, but at a lower clicking rate (total of 66 clicks)
Morphometric Measurements Intermolt Postmolt p-Value
when compared to intermolt group (131 clicks) probably due to a
a a
decrease in feed intake. Although this is the first record of acoustic ac­
Weight (g) 8.85 ± 1.22 8.66 ± 0.86 0.688
tivity in the postmolt stage of L. vannamei, the experimental design did
Total length (cm) 10.25 ± 0.51 a 10.36 ± 0.36 a 0.656
Carapace length (cm) 2.30 ± 0.10 a 2.33 ± 0.07 a 0.305 not allow to determine the exact time elapsed between ecdysis and
Mandible length (mm) 2.26 ± 0.12 a 2.18 ± 0.16 a 0.165 recording of shrimp feeding activity, which could have occurred up to
Mandibular cuticle thickness (μm) 36.22 ± 0.96 a 30.08 ± 1.13 b < 0.001 approximately 16 h after ecdysis (stage E). Thus, more detailed moni­
Endocuticle thickness (μm) 22.59 ± 0.38 a 19.92 ± 0.71 b < 0.001 toring of the sub-stages and stages of the L. vannamei molt cycle using
Different letters on the same row indicate significant differences (p < 0.05). PAM could provide further insights into the respective acoustic

4
S. Peixoto et al. Aquaculture 596 (2025) 741761

Fig. 3. Photomicrograph showing hematoxylin and eosin-floxine stained histological sections of the mandibular cuticle of L. vannamei in the intermolt stage (A) and
postmolt stage (B). The cuticular layers are indicated by numbering for epicuticle (1), exocuticle (2), and endocuticle (3).

Table 2
Mean values (± SD) of acoustic parameters of clicks emitted by L. vannamei
during feeding activity in intermolt and postmolt stages.
Acoustic Parameters Intermolt Postmolt p-value

Minimum frequency (kHz) 9.97 ± 8.29 a 9.99 ± 7.26 a 0.526

Maximum frequency (kHz) 81.26 ± 16.78 a 72.05 ± 20.22 b 0.002
Peak frequency (kHz) 30.78 ± 21.06 a 32.37 ± 23.08 a 0.888
Maximum energy (dB) 46.56 ± 9.87 a 42.41 ± 6.42 b 0.004
Duration (ms) 4.79 ± 1.94 a 5.48 ± 2.39 a 0.062

Different letters on the same row indicate significant differences (p < 0.05).

parameters of clicks and feeding behavior.

Acoustic feeding activity was detected in both intermolt and post­
molt stage shrimp, but the lower level of sound energy (dB) observed in
recordings of the latter may be associated with reduced voracity and
consequently fewer click emissions during this stage (Chan et al., 1988;
Dall et al., 1990; Bonilla-Gómez et al., 2013; Lemos and Weissman,
2021). Bardera et al. (2019) used video analysis to study the effects of
molting cycle (premolt, intermolt, and postmolt) on the feeding
behavior of L. vannamei (10.5 g) in the laboratory. The authors reported
that shrimp in the postmolt stage spent most of their time engaging in
passive behaviors (e.g., inactivity and food detection) and fed for less
time compared to other stages. The lower acoustic energy generated by
shrimp in the postmolt stage is probably related with variations in
mandible thickness (and possibly hardness) due to the molting process,
which may also have affected some click sound parameters. In this
context, the significantly lower values of maximum frequency (kHz) and
maximum energy (dB) of isolated clicks in postmolt stage shrimp,
associated with their reduced voracity, would help explain the lower Fig. 4. Example of a frequency spectrogram over time, showing clicks of L.
maximum energy in 15-s intervals in recordings of feeding activity in L. vannamei in the intermolt (A) and postmolt stages (B). Yellow arrows indicate
vannamei at this stage compared to shrimp in the intermolt stage. the maximum frequency of clicks, while red arrows indicate the peak fre­
The mandible is the most mineralized structure of penaeid shrimp quency, where the maximum energy of the signal is located (dB; colour scale
exoskeletons, containing specialized incisive dental processes for biting bar on the right). (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)
and cutting (Bentov et al., 2021). Although calcium carbonate miner­
alization is used for overall exoskeleton reinforcement, shrimp, among
other crustaceans, have evolved to deposit calcium phosphate in specific (Smith and Tabrett, 2013; Silva et al., 2019; Zhu et al., 2024) and has
locations of the mandible to increase mechanical strength (Bentov et al., been used to assess L. vannamei feeding activity in laboratory studies
2016; Bentov et al., 2021; Zhu et al., 2024). This calcification is more using Passive Acoustic Monitoring (PAM) (Peixoto et al., 2020a, 2020b;
pronounced at the tips of the incisive denticles, where rigidity is highest, Soares et al., 2021a, 2021b; Peixoto et al., 2022; Hamilton et al., 2023;
and decreases toward the base of the mandible (Bentov et al., 2016; Zhu Tabbara et al., 2024; Zhu et al., 2024) and aquaculture ponds
et al., 2024). The inner part of the mandible consists of muscular tissue (Jescovitch et al., 2018; Ullman et al., 2019a, 2019b; Reis et al., 2020,
responsible for opening and closing the structure for food breaking 2021; Walsh et al., 2022; Strebel et al., 2023; Reis et al., 2023). Recently,
through contraction and relaxation (Dall et al., 1990). Sound emission mandible size has been associated with acoustic parameter character­
during mandibular occlusion in penaeid shrimp is well-documented istics of clicks produced by L. vannamei from 0.01 to 12 g, which showed

5
S. Peixoto et al. Aquaculture 596 (2025) 741761

Fig. 5. Graphical representation of the mean maximum energy values (dB) recorded at 15-s intervals over 30 min in recordings of feeding activity in L. vannamei in
the intermolt and postmolt stages.

lower acoustic energy (dB) produced by juveniles weighing less than 5. Conclusion
1.16 g (Hamilton et al., 2024). However, the effects of the molting
process and consequent changes in mandibular cuticle thickness on the This is the first report that the acoustic characteristics of the clicks
acoustic characteristics of clicks emitted by L. vannamei were evaluated emitted by L. vannamei during feeding are affected by the molting stages.
for the first time in this study. Among the layers forming crustacean The lower maximum energy and maximum frequency of clicks in post­
cuticles, the endocuticle is primarily responsible for exoskeleton hard­ molt shrimp are probably related to their lower mandibular cuticle
ness due to its greater thickness and calcification (Roer and Dillaman, thickness. Although variations in the maximum acoustic energy of the
1984). Thus, the histology results of the mandibles support the hy­ clicks were observed between the postmolt and intermolt stages, it is
pothesis that the lower energy of clicks emitted by postmolt stage shrimp believed that changes in the acoustic landscape energy may be related to
is probably related to the thinner cuticle and endocuticle, and possibly a decreased rate of click emissions, likely due to reduced feeding activity
reduced hardness, of mandibular incisive teeth. in postmolt shrimp. These findings enhance the evaluation of feeding
The acoustic characteristics of clicks in this study fell within the behavior throughout the molting cycle and help optimize algorithms for
parameter limits observed for L. vannamei in different studies using controlling acoustic automatic feeders in large-scale shrimp farming
passive acoustic monitoring (PAM) as reviewed by Reis et al. (2022). systems.
Hamilton et al. (2024) conducted an acoustic characterization of clicks
emitted by different size classes of L. vannamei, observing mean CRediT authorship contribution statement
maximum energy values ranging from 40.8 to 44.8 dB for shrimp with
similar mandible weight and length as in the present study. These Silvio Peixoto: Writing – review & editing, Writing – original draft,
maximum click energy values were similar to those recorded in this Supervision, Resources, Project administration, Methodology, Investi­
study for shrimp in postmolt and intermolt stages (42.41 and 46.56 dB, gation, Funding acquisition, Formal analysis, Data curation, Conceptu­
respectively), which although significantly different under controlled alization. Cecilia Craveiro: Writing – review & editing, Methodology,
laboratory conditions, would likely still be equally detected in auto­ Formal analysis. Fábio Costa Filho: Writing – review & editing, Meth­
mated acoustic feeding systems. In contrast, it was suggested that limi­ odology, Formal analysis. Hildemário Castro-Neto: Methodology,
tations for using acoustic feeders in early stages of L. vannamei culture Formal analysis. Roberta Soares: Writing – review & editing, Writing –
were related to the lower maximum click energy (~23 dB) emitted by original draft, Methodology, Investigation, Funding acquisition,
juveniles weighing less than 1.16 g (Hamilton et al., 2024). However, Conceptualization.
there is no industry-provided information on such click energy detection
processes or decision-making by algorithms related to shrimp molt cy­
Declaration of competing interest
cles in aquaculture systems using this technology. Our results suggest
that while the mandibular thickness in the postmolt stage may affect
The authors declare that they have no known competing financial
click acoustic parameters, possible variations in acoustic landscape en­
interests or personal relationships that could have appeared to influence
ergy due to the molting process are more likely to be related to re­
the work reported in this paper.
ductions in click emission rates associated with decreased feeding
activity. The information from this study may contribute to optimizing
Data availability
algorithms controlling automated acoustic feeders, but it is essential to
validate the generated data in studies using PAM in large-scale shrimp
Data will be made available on request.
farming systems.

Acknowledgements

The authors would like to thank the Foundation for Science and

6
S. Peixoto et al. Aquaculture 596 (2025) 741761

Technology Development of the State of Pernambuco (FACEPE) for the Panakorn, S., 2018. Managing shrimp moulting. Aquacult Asia Pac 14, 18–21.
Peixoto, S., Soares, R., Silva, J.F., Hamilton, S., Davis, D.A., 2020a. Acoustic activity of
research grant (APQ-1561-5.06/22) and providing the postdoctoral
Litopenaeus vannamei fed pelleted and extruded diets. Aquaculture 525, 735307.
scholarship for Cecília Craveiro (BCT-0739-5.06/22). Roberta Soares Peixoto, S., Soares Davis, R.P., Davis, D.A., 2020b. An acoustic based approach to
and Silvio Peixoto are fellow productivity researcher of the National evaluate the effect of different diet lengths on feeding behavior of Litopenaeus
Council for Scientific and Technological Development (CNPq). The au­ vannamei. Aquac. Eng. 91, 102114.
Peixoto, S., Strebel, L., Soares, R., Davis, D.A., 2022. Acoustic feeding responses using
thors also express their gratitude to the students and staff who collab­ marine chemoattractants in plant-based diets for naive and non-naive Litopenaeus
orate with this research project. vannamei. Appl. Anim. Behav. Sci. 257, 105792.
Promwikorn, W., Kirirat, P., Thaweethamsewee, P., 2004. Index of molt staging in the
black tiger shrimp (Penaeus monodon). Songklanakarin J. Sci. Technol. 26, 765–772.
References Reis, J., Hussain, A.S., Weldon, A., Walsh, S., Stites, W., Rhodes, M., Davis, D.A., 2023.
Passive acoustic feeders as a tool to assess feed response and growth in shrimp pond
Almeida Neto, M.E., Freire, A.G., 2007. Avaliação de consumo alimentar e textura do production. Aquac. Int. 31, 1643–1657.
exoesqueleto do camarão marinho Litopenaeus vannamei (Crustacea: Penaeidae) em Reis, J., Novriadi, R., Swanepoel, A., Jingping, G., Rhodes, M., Davis, D.A., 2020.
cultivo comercial, durante o ciclo de muda. B Inst Pesca 33, 147–156. Optimizing feed automation: improving timer-feeders and on demand systems in
Bardera, G., Owen, M.A.G., Pountney, D., Alexander, M.E., Sloman, K.A., 2019. The semi-intensive pond culture of shrimp Litopenaeus vannamei. Aquaculture 519,
effect of short-term feed-deprivation and moult status on feeding behaviour of the 734759.
Pacific white shrimp (Litopenaeus vannamei). Aquaculture 511, 734222. Reis, J., Weldon, A., Ito, P., Stites, W., Rhodes, M., Davis, D.A., 2021. Automated feeding
Bentov, S., Aflalo, E.D., Tynyakov, J., Glazer, L., Sagi, A., 2016. Calcium phosphate systems for shrimp: effects of feeding schedules and passive feedback feeding
mineralization is widely applied in crustacean mandibles. Sci. Rep. 6, 22118. systems. Aquaculture 541, 736800.
Bentov, S., Palmer, B.A., Bar-On, B., Shelef, Y., Aflalo, E.D., Sagi, A., 2021. Reis, J., Peixoto, S., Soares, R., Rhodes, M., Ching, C., Davis, D.A., 2022. Passive acoustic
Reinforcement of bio-apatite by zinc substitution in the incisor tooth of a prawn. monitoring as a tool to assess feed response and growth of shrimp in ponds and
Acta Biomater. 120, 116–123. research systems. Aquaculture 546, 737326.
Berk, I.M., 1998. Sound production by the white shrimp (Penaeus setiferus), analysis of Roer, R., Dillaman, R., 1984. The structure and calcification of the crustacean cuticle.
another crustacean-like sound from the Gulf of Mexico, and applications for passive Amer Zool 24, 893–909.
sonar in the shrimping industry. J. Shellfish. Res. 17, 1497–1500. Schneider, C., Rasband, W., Eliceiri, K., 2012. NIH image to ImageJ: 25 years of image
Bonilla-Gómez, J.L., Chiappa-Carrara, X., Galindo, C., Cuzón, G., Gaxiola, G., 2013. analysis. Nat. Methods 9, 671–675.
Effects of adaptation to laboratory conditions on growth, molting, and food Silva, J.F., Hamilton, S., Rocha, J.V., Borie, A., Travassos, P., Soares, R., Peixoto, S.,
consumption of juvenile Farfantepenaeus duorarum (Decapoda: Penaeidae). 2019. Acoustic characterization of feeding activity of Litopenaeus vannamei in
J. Crustac. Biol. 33, 191–197. captivity. Aquaculture 501, 76–81.
Chan, S.M., Rankin, S.M., Keeley, L.L., 1988. Characterization of the molt stages in Smith, D.V., Tabrett, S., 2013. The use of passive acoustics to measure feed consumption
Penaeus vannamei: Setogenesis and hemolymph levels of total protein, ecdysteroids, by Penaeus monodon (giant tiger prawn) in cultured systems. Aquac. Eng. 57, 38–47.
and glucose. Biol. Bull. 175, 185–192. Soares, R., Peixoto, S., Galkanda-Arachchige, H.S.C., Davis, D.A., 2021a. Growth
Chang, E.S., 1995. Physiological and biochemical changes during the molt cycle in performance and acoustic feeding behavior of two size classes of Litopenaeus
decapod crustaceans: an overview. J Exp Mar Bio Ecol 193, 1–14. vannamei fed pelleted and extruded diets. Aquac Int 29, 399–415.
Corteel, M., Nauwynck, H.J., 2010. The integument of shrimp: Cuticle and its moult Soares, R., Peixoto, S., Davis, R.P., Davis, D.A., 2021b. Feeding behavior and growth of
cycle. In: Alday-Sanz, V. (Ed.), The Shrimp Book. Nottingham University Press, Litopenaeus vannamei fed soybean-based diets with added feeding effectors.
London, pp. 73–88. Aquaculture 536, 736487.
Dall, W., Hill, B.J., Rothlisberg, P.C., Staples, D.J., 1990. The biology of the Penaeidae. Strebel, L.M., Nguyen, K., Araujo, A., Corby, T., Rhodes, M., Beck, B.H., Roy, L.A.,
In: Advances in Marine Biology. Academic Press, London. Davis, D.A., 2023. On demand feeding and the response of Pacific white shrimp
De Tailly, J.B.D., Keitel, J., Owen, M.A.G., Alcaraz-Calero, J.M., Alexander, M.E., (Litopenaeus vannamei) to varying dietary protein levels in semi-intensive pond
Sloman, K.A., 2021. Monitoring methods of feeding behavior to answer key production. Aquaculture 574, 739698.
questions in penaeid shrimp feeding. Rev Aquac 13, 1828–1843. Tabbara, M., Strebel, L., Peixoto, S., Soares, R., Morais, S., Davis, D.A., 2024. Use of
Hamilton, S., Silva, J.F., Costa Filho, F., Peixoto, S., 2021. Sound emission of passive acoustic monitoring to evaluate the effects of a feed effector on feeding
Macrobrachium rosenbergii during feeding activity. Aquacult. Res. 52, 4841–4848. behavior, growth performance, and salinity stress tolerance of Litopenaeus vannamei.
Hamilton, S., Costa Filho, F., Silva, J.F., Duarte-Neto, P.J., Soares, R., Peixoto, S., 2023. Aquaculture 582, 740499.
The loud crowd: interactions between stocking density and acoustic feeding activity Ullman, C., Rhodes, M., Cline, D., Hanson, T.R., Davis, D.A., 2019a. Effects of four
of different size classes of Litopenaeus vannamei. Aquaculture 563, 738904. different feeding techniques on the pond culture of pacific white shrimp Litopenaeus
Hamilton, S., Moura, G.J.B., Costa, F., Craveiro, C., Castro-Neto, H., Almeida, M., Duarte- vannamei. J World Aquac Soc 50, 54–64.
Neto, P.J., Soares, R., Peixoto, S., 2024. Size matters: variability in the acoustic Ullman, C., Rhodes, M.A., Davis, D.A., 2019b. Feed management and the use of
parameters during feeding activity of Penaeus vannamei in different size classes. automatic feeders in the pond production of pacific white shrimp Litopenaeus
Aquaculture 587, 740843. vannamei. Aquaculture 498, 44–49.
Jescovitch, L.N., Ullman, C., Rhodes, M., Davis, D.A., 2018. Effects of different feed Walsh, S., Nguyen, K., Strebel, L., Rhodes, M., Davis, A., 2022. Utilising feed effectors
management treatments on water quality for Pacific white shrimp Litopenaeus and automated feeders for semi-intensive pacific white shrimp (Litopenaeus
vannamei. Aquacult. Res. 49, 526–531. vannamei) production. Aquac Fish Fish 2, 540–551.
Lemos, D., Weissman, D., 2021. Moulting in the grow-out of farmed shrimp: a review. Zhu, B., Wang, Z., Li, Y., Shan, H., Liu, D., Dong, S., Han, X., Wang, F., 2024.
Rev. Aquac. 13, 5–17. Morphological and structural analysis of Penaeus vannamei mandibles and an attempt
Lindseth, A.V., Lobel, P.S., 2018. Underwater soundscape monitoring and fish at real-time cannibalism monitoring based on passive acoustics. Aquaculture Reports
bioacoustics: a review. Fishes 36, 2–15. 37, 102199.
Napaumpaipom, T., Chuchird, N., Taparhudee, W., 2013. Study on the efficiency of three
different feeding techniques in the culture of pacific white shrimp (Litopenaeus
vannamei). J Fish Environ 37, 8–16.