Introduction

 

Light fishing is one of the most effective and advanced fishing methods to catch commercial small pelagic species in both small-scale and large-scale fisheries. In the common fishing practices, fishers use either fixed or mobile fishing gear (Ben-Yami 1976; Wang et al. 2010; Yamashita et al. 2012; Ortiz et al. 2016; Solomon and Ahmed 2016; Nguyen et al. 2017; Nguyen and Winger 2019). However, artificial light consumes a significant amount of energy due to the use of numerous high-powered lamps. One of the prominent small-scale light fishing practices in Indonesia is lift net fishing, which uses a fluorescent lamp as the typical light sources. Fish production of lift net fishing in 2017 reached 48% from total production of small-scale light fishing in Indonesia (Ministry of Marine Affairs and Fisheries Republic of Indonesia 2018). The main targets of lift net fishing are anchovy, scad, sardinella and squid, which proportion of scad ranged between 10–45% from the total catch (Guntur and Munataha 2015; Rudin et al. 2017). The application of fluorescent lamp in lift net fishing has several problems, including short lifetime, high fuel consumption and low effectiveness to control the fish behaviour during fishing operation. Fishing operation using fluorescent lamps in fixed lift net consumed 5.20 to 7.00 L/night (mean 6.33 ± 0.54 SD) while light emitting diode (LED) lamps consumed 3.30 to 5.30 L/night (mean 4.11 ± 0.61 SD). However, it is argued that differences in fluorescent lamp quantities and wattages significantly affect the fishers’ income (Susanto et al. 2017a).

Light emitting diode as the latest efficient light source technology, has the potential to be applied as an artificial light source for fishing with light (An et al. 2017; Susanto et al. 2017b; Nguyen and Winger 2019). This lamp provides the maximum illumination power combined with lower energy consumption, longer lifetime, higher efficiency, better chromatic performance, and lower environmental impact compared to traditional lighting technology (Matsushita et al. 2012; Matsushita and Yamashita 2012; Yamashita et al. 2012; Breen and Lerner 2013; Hua and Xing 2013; Yeh et al. 2014; An et al. 2017; Nguyen and Winger 2019). Furthermore, the light distribution of LEDs, colour and intensity considerably affect fish behavioural and retinular response. Therefore, understanding the behaviour of target species in response to LEDs is an important step to develop efficient LEDs for lift net fisheries.

We illuminate scad (Selairoides leptolepis) by three different low powered LED light sources including green, orange and white lamps, as well as dark conditions to investigate the behavioural and retinular response. Furthermore, in order to determine the LEDs performance for lift net fishing, we focused to construct the basic evidence regarding the behavioural and the retinular response of scad through fishing trial. This information is important to develop efficient and effective LEDs for lift net fishing in Indonesia.

 

Materials and Methods

 

Fish and tank experiment

 

For the laboratory experiment, the behaviour monitoring was conducted in a black fiberglass of rectangular prims open tank (150 W x 200 L x 50 H cm) and the water depth was maintained at 30 cm. The tank was placed in the controlled dark room at State College of Fisheries Serang City, to secure no natural light existing during experiment. The tank was divided into six zones and marked at the bottom in 10 cm intervals as the calibration scale (Fig. 1). Being closer to the light source zone 3 and 4 are light zones (bright zones), while zone 1,2,5, and 6 are the dark zones. The LEDs were assembled approximately 20 cm from sea water level at experiment tank. The experimental setup was installed by following the researches of Marchesan et al. (2005), Pignatelli et al. (2011), Cha et al. (2012) and Utne-Palm et al. (2018).

The running water systems were installed to ensure the fish remain alive in optimum water quality during observation. Scad (12.45 cm in average total length (TL), N=60) were collected using a guiding barrier trap at Banten Bay and transferred to the laboratory for adaptation and acclimatization period for seven days. The scad were exposed to normal daily light-dark cycle (12 L: 12 D, sun light was used in daytime). The water salinity in all tanks was maintained at 30–33‰, the temperature was at 29–31°C and dissolved oxygen ranged from 5.9–6.1 mg/L. Before and after the experiments, fish were fed two times per day with Arthemia spp. Therefore, the behaviour experiment was conducted on static water circulation. All of the experiments were performed during the night to minimize the influence of light from any endogenous circadian effects on the fish behavioural and retinular responses.

 

Light source and behavioural methods

 

Lamps were assembled using four dual inline package (DIP) LEDs (Shenzhen Yuliang Optoelectronics Technology Co. Ltd) mounted on the metal housing (11L x 5W x 7H cm), powered by 4 V DC supply. The experiments were conducted in three colors of LEDs i.e., green [approximate peak wavelength = 565 nm], orange [approximate peak wavelength = 600 nm], and white [approximate peak wavelength = 450 nm and 545 nm]. Therefore, each of which consists of three illumination levels i.e., 20, 35, 50 lux. Light intensities were measured using ILT 5000 research radiometer at 15 cm distance bellowed sea water level of the tank. The intensity of green LEDs at 20, 35 and 50 lux is 24, 54 and 90 μW.cm-², respectively. In the same order of illumination level, the intensity for orange LEDs was 20, 21 and 24 μW.cm-² while for the white LEDs was 15, 76 and 94 μW. cm-².

The order of the experimental procedures is as follows. Firstly, before being tested 30 scad were left at the experimental tank at least three days to acclimate and were subjected to a 12 L:12 D photoperiod (light from 06:00 to 18:00; dark from 18:00 to 06:00). Ambient illumination at 15 cm deep below seawater-level was about 16 nW.cm-² in light conditions. All experimental sessions started at 19:00 and ended at 22:00. Secondly, before the experiments were started and between each experiment, fish were kept in the dark state for 30 min to ensure their retina in a scotopic condition. Subsequently, each lamp was turned on for 30 min to allow fish to respond and adapt to the light. Visual observation and video recording using the infrared camera were conducted during the dark and lighted conditions. Three replications were conducted for each experiment (Marchesan et al. 2005). A total of 540 min of video recording was analysed for each color to define fish proportion, scad swimming speed around illumination zone, mean of nearest neighbour distance (MNND), and swimming pattern of the fish schooling.

 

Retinular response to irradiance changes

 

We used eighteen fish for the retinal adaptation experiment using the following procedure. Firstly, two fish were taken from the storage tank and placed in a cylinder tank with diameter 50 cm and height 43 cm. After 30 min set in the dark state, each LED lamp was illuminated for approximately 30 min and at the end of each treatment, eye specimens from both fish were collected. Subsequently, each specimen was fixed in Bouin’s solution and infiltrated with paraffin. Tissue samples were cut in cross-sections of 4–6 μm thickness and were stained with haematoxylin and eosin for examination under the microscope. This histological process followed the procedures from Arimoto et al. (2010) and Jeong et al. (2013).

 

Fishing experiment

 

The result of laboratory experiment was applied to define the suitable LEDs color for the fishing trial which was conducted using a fixed lift net in Banten Bay during peak fishing season on August 2018. The fishing trial was conducted at 10 m water depth as common fishing ground in Banten Bay. According to the laboratory experiment, scad are especially responsive to white and green LEDs. Therefore, both lamps were compared to a fluorescent lamp, the light source used in the existing lift net fishing. We conducted the fishing experiment using 50 W of a fluorescent lamp (typical lamp for lift net fishing) and 1.4 W of LEDs due to the similar light output in the seawater column (12-15 μW.cm-²; measured using ILT 5000 at 1 m below sea water). Ten fish were collected from each lamp making it 30 fish in total. Subsequently, all fish’s eyes went through the histological procedures, followed by the examination of retinular adaptation under the microscope.

Data analysis

 

In order to determine the degree of fish preference on each light stimulus, the proportion of fish at each zone was analysed by counting the number of gathered fish in each zone per minute observation (Kim and Mandrak 2017). Social behaviour was determined by the mean nearest neighbour distance (MNND), which is the average of the planar distances between each fish (head) and its closest neighbour (Parrish et al. 2002). It was used to define the

 

 

Fig. 1: The open tank experiment for the behavioural response. The side view of experimental tank set up (a), top view (b), and infrared camera view (c)

 

 

Fig. 2: Illustration of NND method from the image recording. N1 to N5 represent the number of fish. R1 to R7 represent the planar distances between each fish (head) and its closest neighbour

 

 

Fig. 3: Photomicrographs show various states of retinal light adaptation of Selaroides spp. A thickness from the limiting membrane to the surface of the retina and B thickness of cone cell migration. a. Dark adapted, b. transitional stage, and c. light adapted. CS: cone cell. Scale bar = 10 μm

effect of different colors and intensity of schooling characteristics of fish (Torisawa et al. 2007; Jolles et al. 2017). The MNND was analysed using images that converted from the movie at the beginning (< 10 min), intermediate (11–20 min), and the end of observation (21–30 min). The distance was analysed with Kinovea 0.8.15 at the center head of fish (Fig. 2).

The degree of retinal light adaptation was represented in the adaptation ratio (%) (Arimoto et al. 2010), which was calculated by (B/A) × 100 (%), where A (μm) represents the distance between the limiting membrane and the surface of the retina, and B (μm) represents the migration of cone cells when it was stimulated by light. A and B were measured using photomicrographs (Fig. 3). Swimming patterns at both LEDs colors and irradiance levels were analysed by using video tracking and trajectory software (Kinovea 0.8.15). Videotapes were preliminary observed at 4X speed, to obtain the first qualitative swimming pattern and the characteristic of each stage. In all experiments at every repetition, three dominant swimming patterns from each LED’s stimulus were chosen at the beginning, intermediate and the end of the observation period. Parameters used to analyse the swimming speed of each pattern in total length per second (TL/s) are (1) time-lapse in each pattern; (2) number of frames; (3) distance

 

 

Fig. 4: The proportion of fish related to the color and light intensity. The proportion of scad observed during replicated color treatments (green-circle; orange-square; white-triangle;) and dark condition (filled circle)

 

 

Fig. 5: The light adaptation of scad retina cells to different color and light intensity in laboratory experiment (a) and the light adaptation of the retina cells collected from the fishing experiment (b). The cone index of fluorescent-circle, white LED-square, and green LED-triangle

of each pattern. One-way ANOVA was applied to analyse the effect of different light stimuli (as explanatory variable) to proportion, MNND, and swimming speed during observation (as response variable). Post-hoc comparisons, wherever significance was found, were conducted using Tukey HSD test with the significance level was set at P < 0.05.

 

Results

 

Fish preference to light stimuli

 

The scad responded to the different light colors. At all illumination levels of each color, fish showed high aggregation levels to the light zone. It was indicated by higher fish proportions in zone 3 and 4 than dark zone as presented in Fig. 4. However, there was not a significant difference in fish proportions at the dark condition for all zones (zone 1 to 6). The fish proportion at the dark condition between the range 14 and 19%. The proportion tended to increase related to the aggregating of light intensity, especially at the light zone. At the low intensity, the proportion of fish at zone 3 and zone 4 was ranged between 22–32%. However, the proportion was increased to 34–44% at high intensity level. There was also evidence to suggest that light intensity (in the range 15–94 μW.cm-²) influences the behaviour of the fish by modifying swimming aggregations and preference in all LEDs colors.

Relationship between the retinular response and irradiance change

 

The light adaptation of scad was influenced by LEDs color and intensity. The degrees of adaptation of scad at scotopic adaptation are in the range between 26 and 34%. Adaptation ratio increased with increasing light intensity at each LEDs color (Fig. 5a). The green LEDs generated the highest adaptation ratio with a slight increase, range 83% (low intensity) to 93% (high intensity). However, the degree of light adaptation with orange and white LEDs produced various tendencies. The adaptation ratio of orange LEDs at low intensities is 32% and increase to 67% for the high light level. Furthermore, white LEDs generated higher adaptation ratio than orange LEDs with the ratio between 73% (low intensity) to 92% (high intensity).

 

Swimming behaviour

 

The swimming behaviour of scad was strongly affected by light intensities at each LED color. There was a significant decrease of MNND in all treatment with increasing intensity. The farthest individual distance generated with orange LED was 14.8 cm, while the closest distance was found at a white LED of 8.2 cm. In all experiments, the decline of MNND has a relationship with increasing swimming speed. Scad has the fastest swimming speed at green LED approximately 3.0 TL.s-1, while the lowest speed was initiated at low intensity of white LED approximately 1.4 TL.s-1 (Fig. 6).

The different LED treatments also influenced the pattern of swimming behaviour. In all dark conditions, fish swam randomly around the experiment tank. There are inconsistent and irregular swimming patterns of scad during the dark observation period. The light exposure caused changes in swimming behaviour. In low intensity, the different color of LEDs did not affect the behavioural patterns. Moreover, the increasing light intensity induces the transformed swimming pattern in all colors. The fish swam randomly and inconsistently at a high intensity of the white LED. However, the scad showed a consistent response to orange and green LED with different radii in swimming patterns. Its radius with green LED is closer to the main light zone than orange LEDs, as presented in Fig. 7. Moreover, there were constant and stable swimming patterns related to the time elapsed at the green LEDs, approximately after 20 min of observation. The specifics of fish swimming behaviour are presented in Table 1.

 

Fishing trials

 

 

Fig. 6: The swimming behaviour of scad to different LED colors and intensities. The mean nearest neighbor distances (a) and swimming speed (b)

There are 1,082 scads were caught during fishing experiment. The light adaptation of the retina collected during the fishing trial is shown in Fig. 5b. The cone index was found to be between the range of 78 and 91%. The green LED generated a higher mean adaptation ratio (84%) than white LED (81%) and a fluorescent lamp (83%). However, there were no significant differences in cone index from the sea experiment (P > 0.05).

 

Discussion

 

The highest number of fish proportion in the bright observation zone indicated the scad as a phototaxis fish that attracted by light. The proportion was superior to the bright zones. However, these zones have a smaller area than shadow zones. There were significant differences in the fish proportion between colors and intensities, whereas the proportion in the bright zone at each color gradually increased with rising intensity. The brightness level influences the level of fish activity. Thus, it would have been relevant to increase the fish proportion in all bright zones (Marchesan et al. 2005; Utne-Palm et al. 2018). This condition is an adaptation response to maintaining the formed characteristic of scad schooling behaviour, related to the exposure of light intensity in their environment (Woodhead 1966; Martin and Perez 2006).

There were significant differences in light adaptation ratio between dark and light state (P < 0.05). The degree of light adaptation ratio has a positive association with an increase of light intensity (Susanto et al. 2017c). With all colors used in the experiment, the adaptation stage of cone cells increases with expanding intensity. The exposure of high-intensity LEDs induces the cone cells into photopic adaptation rapidly. Thus, the light adaptation ratio was increased (Tamura and Niwa 1967; Nakano et al. 2006; Migaud et al. 2007). The green LEDs generated the highest adaptation ratio of scad in all intensity levels. Thus, it would have been relevant to conclude that the maximum sensitivity of the Carangidae fish family, including scad, has peak sensitivity between 494–500 nm (Munz and McFarland 1973).

Light intensity has prevailing influence on the visual ability of fish. However, scad’s ability to use vision to maintain the schooling characteristic during light level increases necessitates phototaxis. Moreover, there were significant differences in swimming behaviour, including swimming patterns and MNND in the different light conditions, whereas the fish activity increased with rising intensity. The MNND has a negative relationship with light intensity, whereas the MNND decreased with increasing intensity levels. However, the swimming speeds of scad showed different tendencies with MNND during expanding light levels. The fish swam faster at the high intensity at all LED colors. The high intensity induced fish easier to maintain the direction and orientation of their schooling, due to an increase in their swimming speed at all treatments (Miyazaki et al. 2000). Similar tendencies were found at the swimming speed of Atlantic salmon Salmo salar. Its speed was increased from Table 1: Fish behaviour related to the time elapsed observation

 

Time elapsed (minute)

Fish Behavioural Response

0–5

There was no schooling and swimming behaviour pattern at the beginning treatment. Fish swam in all directions due to orientation and adaptation period related to the light color and intensity at the experimental tank.

6–10

The fish started to school with several swimming patterns. However, there were an unstable direction and swimming speed.

11–20

There was a stable and consistent swimming pattern. The swimming speed was increased related to time-lapse.

21–30

The swimming patterns were stable and consistent with steady swimming speed. The radius of swimming was stable and relatively closest to the center of the light zone.

 

 

Fig. 7: The individual swimming pattern of scad in different color and LEDs intensity. The swimming pattern in dark condition (a), green LED (b), orange LED (c) white LED (d). Different fish are color-coded

0.2 BL/s to 0.5 BL/s related to the increase of the light level at sea cage observation (Hansen et al. 2017).

The swimming patterns of scad in the green and orange LEDs have similar tendencies. However, the swimming radius at the green LED was closer to the center of the light zone than an orange LED. It was related to the visual adaptation level and the spectral sensitivity of scad. The scad have more reactive to the green light because it is suitable with peak sensitivity level. The fish have a proper response to green light due to an increase in visual ability and significant influence on fish capability to maintain their schooling characteristics. Exposure of green light in different light intensities was induced the stable and consistence swimming pattern. In one example of schooling during increased light intensity, increasing visual ability influenced each individual, enabling them to maintain their distance and formation relative to the rest of the school during swimming (Glass et al. 1986; McMahon and Holanov 1995; Miyazaki and Nakamura 1990).

Even though the orange light has longer wavelength than green light, it has lower of photon energy. It causes the scads to have less reactive and induce wider swimming radius than green light. However, the swimming pattern of scad in orange light was relatively stable and consistence during observation. In other example, orange light has similar influenced with green light to the schooling characteristic of Mugil cephalus, Sparus auratus, and Lithognathus mormyrus (Marchesan et al. 2005).

The light adaptation of scad from the fishing experiment has similar tendencies with laboratory observations. The green LED generated a higher degree of adaptation than the white LED and fluorescent lamp at the same intensity. The information on the retinular response and adaptation to light source was utilized in studying the relationship between a light fishing procedure, light color, and light intensity to develop an efficient use of the LED fishing lamp (Jeong et al. 2013). In this research, we compare the characteristics of scad swimming behaviour, and light adaptation between experimental tanks and fishing trials to determine the suitable low powered LED color as a light source when fishing. From these results, the green LED, which induces good light adaptation and stable swimming patterns, is suitable enough to substitute fluorescent lamps currently used in lift net fishing.

The combination in both green and white LED can be a more effective light attractant to scad fishing at lift net fisheries. In lift net fisheries, white LED is useful when gathering scads and other target species at the initial fishing operation. However, the swimming pattern of fish school at white LED was random and unstable, due to the light source having to change with green LED when focusing the fish at a catchable area. Green light would be more suitable to keep fish close to the light source. Moreover, this light not induces stress behaviour for long exposure time (Shin et al. 2013; Stien et al. 2014; Sierra-Flores et al. 2015). The application of green light in focusing of fish can reduce uncaught fish during hauling process and improve the fishing efficiency and effectiveness in lift net fishing.

The LED innovation as artificial light has several advantages. The LED provides maximum illumination power, combined with minimum energy consumption, long lifespan, high efficiency, better chromatic performance, and reduced environmental impact compared to traditional lighting technology (Matsushita et al. 2012; Matsushita and Yamashita 2012; Yamashita et al. 2012; Breen and Lerner 2013; Hua and Xing 2013; Yeh et al. 2014; Nguyen and Tran 2015; An et al. 2017). However, further fishing trials are recommended to validate the effectiveness of both white and green LED as a light stimulus for gathering and focusing scad at lift net fisheries.

 

Conclusion

 

Present investigation showed that the white LED is a suitable enough to attract fish to catchable area while the green LED is an effective color for focusing and control behaviour of scad in catchable area. We conclude that combination of white and green LED can be a more effective light attractant to scad fishing at lift net fisheries.

 

Acknowledgments

 

This research is supported by the Ministry of Research, Technology and Higher Education, The Republic of Indonesia through graduate team research grant number 1549/IT3.11/PN/2018.

References

 

An YI, P He, T Arimoto, UJ Jang (2017). Catch performance and fuel consumption of LED fishing lamps in the Korea hairtail angling fishery. Fish Sci 83:343‒352

Arimoto T, CW Glass, X Zhang (2010). Fish vision and its role in fish capture. In: Behavior of Marine Fishes: Capture Processes and Conservation Challenges. He P (Ed.). Blackwell Scientific, USA

Ben-Yami M (1976). Fishing with light. Fishing News Books, Oxford, USA

Breen M, A Lerner (2013). An introduction to light and its measurement when investigating fish behaviour. In: Symposium on the Light session and the Topic Group Lights: ICES-FAO Working Group on Fishing Technology and Fish Behaviour. May 6‒10, Bangkok, Thailand

Cha BJ, BS Bae, SK Chob, JK Oh (2012). A simple method to quantify fish behaviour by forming time-lapse images. Aquac Eng 51:15‒20

Glass CW, CS Vardle, WR Mojsiewicz (1986). A light intensity threshold for schooling in the Atlantic Mackerel Scomber scombrus. J Fish Biol 29:71‒81

Guntur F, A Munataha (2015). Effect of underwater lamp intensity on the lift net’s fishing catches. J Mar Fish 6:195‒202

Hansen TJ, PG Fjelldal, O Folkedal, T Vågseth, F Oppedal (2017). Effects of light source and intensity on sexual maturation, growth and swimming behaviour of Atlantic Salmon in sea cages. Aquac Environ Interact 9:193‒204

Hua LT, J Xing (2013). Research on LED fishing light. Res J Appl Sci Eng Technol 5:4138‒4141

Jeong H, S Yoo, J Lee, YI An (2013). The retinular responses of common squid Todarodes pacificus for energy efficient fishing lamp using LED. Renew Ener 54:101‒104

Jolles JW, NJ Boogert, VH Sridhar, LD Couzin, A Manica (2017). Consistent individual differences drive collective behaviour and group functioning of schooling fish. Curr Biol 27:1‒7

Kim J, NE Mandrak (2017). Effects of strobe lights on the behaviour of fresh water fishes. Environ Biol Fish 100:1427‒1434

Marchesan M, M Spoto, L Verginellab, EA Ferreroa (2005). Behavioural effects of artificial light on fish species of commercial interest. Fish Res 73:171‒185

Martin RS, JAA Perez (2006). Cephalopods and fish attracted by night lights in coastal shallow-waters, off Southern Brazil, with the description of squid and fish behaviour. Rev Etol 8:27‒34

Matsushita Y, Y Yamashita (2012). Effect of a stepwise lighting method termed “stage reduced lighting” using LED and metal halide fishing lamps in the japanese common squid jigging fishery. Fish Sci 78:977‒983

Matsushita Y, T Azuno, Y Yamashita (2012). Fuel reduction in coastal squid jigging boats equipped with various combinations of conventional metal halide lamps and low-energy LED panels. Fish Res 125–126:14‒19

McMahon TE, SH Holanov (1995). Foraging success of largemouth bass at different light intensities: Implications for time and depth of feeding. J Fish Biol 46:759‒767

Migaud H, M Cowan, J Taylor, HW Ferguson (2007). The effect of spectral composition and light intensity on melatonin, stress and retinal damage in post-smolt Atlantic salmon, Salmo salar. Aquaculture 270:390‒404

Ministry of Marine Affairs and Fisheries Republic of Indonesia (2018). Statistics of Fish Production based on Fishing Gear in Indonesia 2017, Jakarta, Indonesia

Miyazaki T, Y Nakamura (1990). Single line acuity of 0-yearold Japanese parrotfish determined by the conditioned reflex method. Nipp Suis Gakk 56:887‒892

Miyazaki T, S Shiozawa, T Kogane, R Masuda, K Maruyama, K Tsukamoto (2000). Developmental changes of the light intensity threshold for school formation in the striped jack Pseudocaranx dentex. Mar Ecol Progr Ser 192:267‒275

Munz FW, WN McFarland (1973). The significance of spectral position in the rhodopsins of tropical marine fishes. Vis Res 13:1829‒1874

Nakano N, R Kawabe, N Yamashita, T Hiraishi, K Yamamoto, K Nashimoto (2006). Color vision, spectral sensitivity, accommodation, and visual acuity in juvenile masu salmon Oncorhynchus masou masou. Fish Sci 72:239‒249

Nguyen KQ, PD Winger (2019). Artificial light in commercial industrialized fishing applications: A review. Rev Fish Sci Aquac 27:106‒126

Nguyen KQ, PD Winger, C Morris, SM Grant (2017). Artificial lights improve the catchability of snow crab (Chionoecetes opilio) traps. Aquac Fish 2:124‒133

Nguyen QK, DP Tran (2015). Benefits of using LED light for purse seine fisheries: A case study in Ninh Thuan Province, Viet Nam. Fish Peop 13:30‒36

Ortiz N, JC Mangel, J Wang, J Alfaro-Shigueto, S Pingo, A Jimenez, T Suarez, Y Swimmer, F Carvalho, BJ Godley (2016). Reducing green turtle bycatch in small-scale fisheries using illuminated gillnets: The cost of saving a sea turtle. Mar Ecol Progr Ser 545:251‒259

Parrish JK, SV Viscido, D Grunbaum (2002). Self-organized fish schools: an examination of emergent properties. Biol Bull 202:296‒305

Pignatelli V, SE Temple, TH Chiou, NW Roberts, SP Collin, NJ Marshall (2011). Behavioural relevance of polarization sensitivity as a target detection mechanism in cephalopods and fishes. Phil Trans Roy Soc B 366:734‒741

Rudin MJ, R Irnawati, A Rahmawati (2017). Differences of fixed lift net catch result by using CFL lamps and underwater LED in Banten Bay water. Mar Fish J 7:167‒180

Shin SH, NN Kima, YJ Choi, HR Habibi, JW Kim, CY Choi (2013). Light-emitting diode spectral sensitivity relationship with reproductive parameters and ovarian maturation in Yellowtail damselfish, Chrysiptera parasema. J Photochem Photobiol B Biol 127:108‒113

Sierra-Flores R, A Davie, B Grant, S Carboni, T Atack, H Migaud (2015). Effects of light spectrum and tank background colour on Atlantic cod (Gadus morhua) and turbot (Scophthalmus maximus) larvae performances. Aquaculture 450:6‒13

Solomon OO, OO Ahmed (2016). Fishing with light: Ecological consequences for coastal habitats. Intl J Fish Aquac Stud 4:474‒483

Stien LH, JE Fosseidengen, ME Malm, H Sveier, T Torgersen, DW Wright, F Oppedal (2014). Low intensity light of different colours modifies Atlantic salmon depth use. Aquac Eng 62:42‒48

Susanto A, R Irnawati, Mustahal, MA Syabana (2017a). Fishing efficiency of LED lamps for fixed lift net fisheries in Banten Bay Indonesia. Turk J Fish Aquat Sci 17:283‒291

Susanto A, MS Baskoro, SH Wisudo, M Riyanto, F Purwangka (2017b). Seawater battery with Al-Cu, Zn-Cu, Gal-Cu electrodes for fishing lamp. Intl J Renew Ener Res 7:1857‒1868

Susanto A, ADP Fitri, Y Putra, H Sutanto, T Alawiyah (2017c). Response and adaptation of Anchovy (Stolephorus spp.) to light emitting diode (LED) lamp. J Mar Fish 8:39‒49

Tamura T, H Niwa (1967). Spectral sensitivity and color vision of fish as indicated by S-potential. Compar Biochem Physiol 22:745‒754

Torisawa S, T Takagi, H Fukuda, Y Ishibhasi, Y Sawada, T Okada, S Miyashita, K Suzuki, T Yamane (2007). Schooling behaviour and retinomotor response of juvenile Pacific bluefin tuna Thunnus orientalis under different light intensities. J Fish Biol 71:411‒420

Utne-Palm AC, M Breen, S Løkkeborg, OB Humborstad (2018). Behavioural responses of krill and cod to artificial light in laboratory experiments. PLoS One 13; Article e0190918

Wang JH, S Fisler, Y Swimmer (2010). Developing visual deterrents to reduce sea turtle bycatch in gill net fisheries. Mar Ecol Progr Ser 408:241‒250

Woodhead PMJ (1966). The behaviour of fish in relation to light in the sea. Oceanogr Mar Biol Annu Rev 4:337‒403

Yamashita Y, Y Matsushita, T Azuno (2012). Catch performance of coastal squid jigging boats using LED Panels in combination with metal halide lamps. Fish Res 113:182‒189

Yeh N, P Yeh, N Shih, O Byadgi, TC Cheng (2014). Applications of light-emitting diodes in researches conducted in aquatic environment. Renew Sustain Ener Rev 32:611‒618