The number of fireflies in Japan has decreased with the development of cities and industries. As a first step toward the regeneration of an environment for fireflies, we report an attempt to breed aquatic firefly larvae. Focusing on the operational stability of equipment used, we use a safety evaluation method like that employed in petrochemical plants. The death of larvae is a major risk in the operational safety of firefly larvae breeding. The larval death scenario is analyzed and clarified using fault tree analysis, demonstrating that reducing operational mistakes in monitoring water flow and temperature conditions contribute most to the reduction of the probability of larval death. An experiment simulating water supply stoppage to estimate the permissible time for a response once the water temperature rose is reported. A firefly larval breeding system equipped with an alarm function for prompting emergency responses is constructed. Fireflies, including mature larvae, are successfully bred and the effectiveness of this system is demonstrated by comparing it with a conventional system. The superiority of the breeding system of this study is indicated by a twofold increase in the yield of mature larvae and the lower expenses related to this strategy.
The developed system has superior operational stability, economics and enables obtaining larvae in high yields.
The death of the larvae is identified a significant risk by the safety evaluation method.
Error reduction in monitoring water flow and temperature can have a great impact on larval death probability.
Similar content being viewed by others
The flickering light of fireflies in their natural habitat has been cherished by young and old people for over thousand years . Two species popular in Japan—Luciola cruciata (Coleoptera: Lampyridae) and L. lateralis (Coleoptera: Lampyridae)—are rare types of fireflies whose larvae live in water . Most of the existing studies on aquatic flashing fireflies, such as L. cruciata and L. lateralis, were conducted by Japanese researchers. These studies are categorized into ecology, genetics, flashing, education, and conservation activities (for instance, the effect of rainfall on population , synchronized luminescence , effect of prey density , DNA of the Genji firefly , luciferase of fireflies , phenomena involved in firefly blinks , and education ).
Fireflies have been a very familiar presence in Japan since a long time, and the flashing fireflies flying around, in particular, have been a part of Japan’s rural landscape [1, 2]. Due to habitat loss, their populations have decreased significantly over the last 30 years. Ohba, and Suzuki  reported that L. cruciata were abundant before 1988 but have gradually decreased, and investigations into the cause of the decline and plans for conservation and breeding have been pursued. As fireflies are becoming less familiar, Ueda and Kusamitsu  reported that the decrease in the number of fireflies due to urbanization has gradually diminished interest in the fireflies by reducing the chances of seeing them. They reported that the decrease in the number of fireflies follows the trend of increase the people who do not see fireflies despite the presence of fireflies in their surroundings. In their paper, they suggested the need for measures to increase human interest in living organisms. Firefly conservation activities, education, and events for watching and observing fireflies can be considered as measures to raise interest in fireflies [12,13,14]. As for the conservation activities, efforts for restoring the firefly population and protecting the existing fireflies are increasing. For instance, Yajima  reported the progress of a project in which L. cruciata and L. lateralis were released into the grounds of the Imperial Palace in hopes of them settling there. Forty years since its inception, the project reached a stage where the population of fireflies around the forest of the Imperial Palace evidently increased. Kashio  reported collaborations between public and private sectors, such as artificial breeding of L. cruciata and cultivation of black snail in Moriyama, Shiga Prefecture. In a report on education, Nakamura  stated that regaining interest in rivers is important for regenerating rivers in cities and to receive cooperation from elementary schools for regenerating firefly canals where children can play safely. Studies on firefly observation events include observational skills , creating a firefly map and providing firefly information, improving tourism resources, strengthening ties with the community through waterside conservation activities, and citizens’ awareness of water improvement .
In this study, we focused on the conservation of fireflies including measures to raise human interest in these insects. Therefore, the objective of this research is to increase the population of large flashing L. cruciata, which is native to Japan.
This study contributes to “developing observational skills,” “increasing citizen awareness of needs for water quality improvement,” and “conservation of fireflies as a measure to increase interest in fireflies and other wonders of nature that have decreased in prevalence.”
In the next section, we present the methods used in this study. A method for constructing an artificial rearing device for larvae using risk management techniques is described. Section 3 shows the result of this approach. First, a risk assessment and risk treatment are presented. Subsequently, the construction and operation results of the larval artificial breeding apparatus are described. In Sect. 4, we present the discussion. The results are summarized, and future issues are indicated. The final section presents our conclusions.
The goal is to regenerate an environment in which fireflies fly naturally. The first step is promoting enlightenment activities by releasing breeding fireflies, the second step is to improve the environment where prey for fireflies can live, and the third step is to improve the environment where fireflies can live, eventually achieving the natural regeneration of the firefly population. This study aimed at the successful artificialization of egg hatching and larvae breeding, i.e., establishing the artificial breeding and release of firefly larvae. Construction and stabilization of the artificial breeding system of the larvae was one of the challenges in achieving the first step. Regarding the studies on breeding device, Yoshioka  developed a container for stable breeding of L. cruciata immediately after hatching, using plankton net as a material. Water exchange is facilitated because the larvae cannot pass through the plankton net (mesh size 0.1 mm). Minami  tried breeding in a constant temperature aquarium as a way to grow fireflies in an artificial garden that resembles the natural world. Since both are water-stop systems, it is troublesome to change water regularly and control the water temperature. Further, it is not suitable for mass breeding. Therefore, at first a device that could breed large numbers of larvae is built. However, if the larvae are washed away during the breeding or if they die as a result of the worsening of the breeding conditions, the step to natural regeneration is interrupted and difficult to restart. Based on the understanding that risk assessment and risk treatment during the breeding are important, this study attempted to apply the safety evaluation system of petrochemical plants—a complex technological system consisting of humans, operations, and facilities whose accident scale is relatively large—to the firefly larval breeding system. The details of the method are shown below.
2.1 Water quality analysis
Before the breeding experiments, the water quality for larval breeding was confirmed. The analysis methods are as follows (Japanese Industrial Standards, JIS):
Ions were analyzed using infrared analysis (JIS K 0101 25.2), ion chromatograph method (JIS K 0101 32.5, 42.4, 37.2.5), ICP emission spectroscopy (JIS K 0101 49.3, 50.3), iodine titration method (JIS K0102 32.1). pH was analyzed through the glass electrode method (JIS K0101 11.1). Dissolved oxygen was analyzed using the iodine titration method (JIS K0102 32.1).
2.2 Methods summary of the natural processes to be accommodated
Similar to other holometabola, L. cruciata become adults through three periods: eggs, larvae, and pupae. Because the eggs are laid next to a riverbank and hatched larvae immediately move into the river, it is necessary to devise complex rearing procedures that replicate the larval development stage, subsequent pupation, and adult emergence without significant mortality. Below, we first summarize natural growth processes (1–4), then the artificial rearing methods developed to maximize the numbers of adults produced.
Female L. cruciata lay their eggs on moss that grows on slopes within approximately 50 cm of the riverbank. Spawning takes place at night and the number of eggs laid per insect is approximately 300–500. The eggs are ellipsoidal, with a minor axis of approximately 0.50 mm and a major axis of approximately 0.55 mm, and are pale yellow in color, becoming grayish and dark brown over time. The eggs hatch in early and mid-July, approximately one month after spawning. Hatching begins at midnight and ends around 5 am.
The hatched larvae fall from the moss into the river and immediately commence underwater life. The size of each larva after hatching is approximately 1.5 mm in length and 0.3 mm in width. They molt six times in water and mature, during which period their body size become larger than that of the adult. Females are 20–30 mm long and 5.5–8.0 mm wide, and males are 16–26 mm long and 3.5–5.6 mm wide. Their baits are snails, which live in the river.
The mature larvae emerge from the river and reaches land around 7 pm on rainy nights between late March and early April. The landed larvae search for the suitable sites for pupation; upon finding a suitable site, they dig through the soil. Normally, pupation takes place 4–5 cm underground.
The pupae reach their state between the end of May and the end of June. Generally, in warmer regions, the adult stage is reached earlier, and vice versa. The flight speed is usually approximately 30 cm/s, and flights reach a height of around 3 m. Adults usually fly up to three times a night, at around 9 pm, 12 pm, and 3 am, of which the first is the most frequent. During mating, the male seeks and follows the light of the female. Copulation takes place on leaves and stems in a stationary state and lasts 3–24 h. The adult has a lifespan of 1–2 weeks and consumes only water during its lifetime.
2.3 Artificial rearing methods
This study aimed to successfully artificialize stages (1) and (2) described in Sect. 2.2. More specifically, this study is related to the construction and operation of a new aquatic firefly larval breeding device. The research proceeded in the following three processes:
2.3.1 Process (1) Risk assessment: Reduction of risks that inhibit the breeding of firefly larvae
The undesirable phenomena, which affect the breeding of L. cruciata larvae, are failure to produce flying adult fireflies in a given year and halting the cycle, i.e., from egg laying to hatching, to larval stage, to pupal stage, and to the flight of adults. The stable operation of the devices is necessary for the success of larval breeding. To identify, analyze, and evaluate the risks that prohibit the stable operation, it was decided that the evaluation method used at petrochemical plants, where strict adherence to risk management is required, could be applied to this study.
What-If analysis and fault tree analysis (FTA) adapted from Risk Management literature were used to identify the most significant mortality risks. Appropriate references, e.g., [25, 26] were taken into consideration. Considering the fact that the firefly breeding device is not a complex technical system unlike the petrochemical plant, which consists of various types of equipment, we selected What-If analysis, which can be used for the risk identification easily. In conducting the analysis, the What-If analysis items were categorized, and focal points were systematized to comprehensively identify the risks. These categories included material hazards, equipment characteristics, control systems, and management of manipulations and operations. The analysis team comprised several experts, including petrochemical plant process technicians, those who have experience in operating manufacturing facilities, and technicians from the engineering department.
To implement risk reduction measures, it is necessary to break down the route leading to accidents and disasters as follows:
Clarify the series of occurrences causing serious events.
Identify and clarify fatal events in the system, elimination of which leads to improvements in system operation.
Evaluate the importance of each basic event.
Calculate the probability of occurrence of a serious event (top event) from the probability of occurrence of basic events.
FTA was selected as the analysis method to obtain the above results via risk assessment. FTA is a quantitative method that can estimate scenario from individual basic events to serious events and calculate the probability of occurrence of serious event. The significant risks among the identified risks were quantitatively analyzed using the fault tree (FT) diagram.
2.3.2 Process (2) Experiments for risk evaluation: Experiments to obtain heat transfer coefficient
Breeding of the firefly larvae requires an environment with a temperature of 28 °C or below , and it is possible to maintain the appropriate temperature by pumping groundwater with a temperature of 16.5 °C ± 0.5 °C year-round. However, there is a concern that if the pump stopped due to some kind of trouble such as blackout, the liquid would stagnate, making it impossible to maintain the constant liquid temperature inside the breeding tray regardless of the outside air temperature, causing the temperature to rise to a level that would hinder larval growth. To understand the risks involved in the breeding, a preliminary study regarding the liquid-temperature rise caused by the water supply stoppage was conducted.
2.3.3 Principle of the simulation model for change of liquid temperature inside the breeding tray
Here θ is the time (h) beginning at initial time θ0 = 0, T is liquid temperature (°C), Ta is the outside air temperature (°C), T0 is the initial liquid temperature (°C), U is the heat transfer coefficient (kcal/m2 h °C), A is the heat transfer area (m2), w is the water holding capacity (kg), and Cp is the specific heat of water (kcal/kg °C).
It is important to accurately obtain the value of heat transfer coefficient U to calculate the temperature rise.
Preliminary experiment (determine of the heat transfer coefficient)
Heat transfer coefficient was obtained through a preliminary experiment when the temperature of the liquid inside the tray rises due to the heat input from the outside air. Generally, there is a difference between the heat transfer coefficient of natural convection heat transfer on a horizontal plane and on a vertical plane. Thus, two containers with different shapes, namely one with a dominant horizontal plane (tray type) and the other with a dominant vertical plane (cylindrical type), were used for the preliminary experiment. The specifications of tray type and cylindrical type are shown in Table 1(a).
Temperature was measured using mercury rod thermometers.
Heat transfer experiment with the firefly larval breeding device
An experiment using the firefly larval breeding tray was conducted to determine the validity of the heat transfer coefficient obtained through the preliminary experiment. The specification of tray is shown in Table 1(b).
In the calculation for volume, the heat capacity of coral stone was assumed to be the same as that of water. Therefore, the whole amount was considered to be water.
Temperature was measured using mercury rod thermometers.
2.3.4 Process (3) Building the breeding device and operation
We built a device that solves the problems of frequent water exchange, water temperature control, and small number breeding, which were the problems of conventional breeding equipment. In addition, Process 3 reflects the risk assessment and risk treatment discussed in the previous processes for the aquatic firefly larval breeding device.
3.1 Water quality
Spring water was obtained from the premises of the chemical factory. Analysis confirmed that the concentration of harmful substances in the spring water was not detected. For the purpose of understanding and categorizing the characteristics of groundwater quality, trilinear diagrams or hexadiagrams are often used [27, 28]. The ion balance of the spring water used in this study represented in a hexadiagram is shown in Fig. 1. The plot used the data from September 2018 to June 2020. Since each ion was well-balanced, and since there were no temporal changes, one can see that the water quality was stable.
3.2 Process (1) Risk assessment
In the What-If analysis, we focused on the individual elements and devices that constitute the system and repeated questions “What if it is this?” and “What if we do this?” that assumed the occurrence of abnormal phenomena such as breakdown or maloperation of the device to identify the potential risks in the subject facility and operational risks. Various risks, including maloperation of pumps and valves and measures against nematodes (Metagonimus yokogawai) infesting Semisulcospira libertina , on which the larvae feed, were considered, and the flow rate of water supplied to the device was identified as the significant risk. Table 2 shows the important risks identified through the What-If analysis.
The result of the What-If analysis determined that a flowmeter should be installed to detect the rate of flow of water to the breeding device. The risk of damage caused by high flow rates can be reduced by implementing the measures proposed in Table 2. Calculating the risk of water supply stoppage, which could increase temperature and the duration of immersion of larvae, causing death of larvae is difficult. The occurrence of some deaths may also cause fatal damage to surviving larvae.
A FT diagram shown in Fig. 2 was prepared with firefly larval death as the top event. Top event is an unwanted system failure event. It includes the final form of failure and the operating state of the system, among others. In FTA, the cause is deductively estimated from the top event and the route to each basic event is drawn. Basic event is an event that cannot be further developed theoretically or structurally, that is, it is an event that triggers an accident. From the FT diagram, the larval death scenario was clarified, and accidents other than the larval death can also be understood.
The probability of occurrence F of each phenomenon can be calculated as the product of failure rate λ and time t.
The failure rate of individual devices can be quoted from databases of chemical plants and nuclear power stations. For instance, the former includes the Guideline for Process Equipment Reliability Data with Data Tables by the Center for Chemical Process Safety, and the latter includes the Component Reliability Data Sheet 2010 by the U.S. Nuclear Regulatory Commission and data from Japanese nuclear power stations .
Tasks with similar contents to human errors (HE) involved in the usage of firefly larval breeding device were selected from the literature on HE [31, 32] with the consideration of the tasks involved, and their human error probabilities (HEPs, the probability of occurrence F in Eq. 3) were quoted. For this purpose, it was deemed that humans always have the same reliability.
Table 3 shows the failure rates and HEPs used for estimating the probability of occurrence of each phenomenon in the FT diagram.
The probabilities of the basic events were calculated as follows, and their results were substituted into the parts indicated with black hatching in the FT diagram (Fig. 2). The operation duration was set to be 6 months between July and December (4320 h = 6 × 30 × 24), and the values from Table 3 were used for the failure rate (1/h) to calculate the probability of occurrence of the basic event using Eq. 3. HEP also used the values from Table 3.
The calculations of “AND gate” and “OR gate” in the FT diagram (Fig. 2) were conducted using Eqs. (4) and (5), respectively, with the unreliability of the system (probability of occurrence) as F(t) and the unreliability of the phenomenon i (from i = 1 to n, where n = 2 or 3 in this study) as Fi(t).
From the diagram, the probability of occurrence of the top event in 6 months was calculated as 2.1 × 10−2events/6 months.
To understand the impact of the probability of occurrence of each basic event on the probability of occurrence of the top event, the probability of occurrence of the top event was calculated by setting each basic event as either not occurring individually (λ = 0) or occurring individually (λ = 1). The results are shown in Table 4. It was suggested that the reduction of maloperations (monitoring errors) is most effective in reducing the probability of occurrence of top event occurrence (indicated with hatching in Table 4). Moreover, when each basic event occurred individually, the probability of occurrence of the top event was almost the same regardless of the occurrence of the basic event.
3.3 Process (2) Experiments for risk evaluation: calculating the heat transfer coefficient
3.3.1 Preliminary experiment (determination of the heat transfer coefficient)
The heat transfer coefficient was obtained through a preliminary experiment when the temperature of the liquid inside the tray rises due to the heat input from the outside air.
Figure 3 shows the experiment results for calculation of heat transfer coefficient. It is inferred that the heat transfer coefficient was ~ 8–10 kcal/m2 h °C (9–12 W/m2 K) at both horizontal and vertical plane dominant cases. The averages of the dry-bulb and wet-bulb temperatures were used as the outdoor air temperature for this calculation.
3.3.2 Heat transfer experiment with the firefly larval breeding device
An experiment using the firefly larval breeding device was conducted to determine the validity of the heat transfer coefficient obtained through the preliminary experiment. Its result is shown in Fig. 4. The result confirmed that the heat transfer coefficient, 8–10 kcal/m2 h °C (9–12 W/m2 K), was mostly valid. We expected the heat transfer coefficient obtained from this experiment to be 6% smaller than that from preliminary experiment due to the difference in materials (stainless steel and PP), but the difference between the two experiments was found to be negligible.
3.3.3 Simulation of pump stoppage in summer
Figure 5 shows the simulation results of the liquid-temperature rise inside the breeding tray when the liquid supply is stopped in summer, for instance, due to power outages. As a result of performing calculation under two cases with outside air temperatures of 40 °C and 35 °C and relative humidity of 80%, it was found that the liquid temperature inside the breeding tray, which was initially at 15 °C, exceeded 30 °C in ~ 5–10 h.
3.4 Process (3) Building and operating the breeding device
3.4.1 Risk treatment
It was demonstrated that the management of supply water flow rate to the breeding device is important for the prevention of the death of firefly larvae caused by water temperature rise. The possible responses to this risk are:
Using two water pumps
Using one pump in operation and the other pump as a backup, reduction of water supply stoppage risk due to equipment malfunction was attempted.
It accurately measures the flow rate of supply water.
Water flow abnormality detection system
When the amount of water flowing into the system detected by the abovementioned flowmeter is either below or above the reference range, the issue is notified to the mobile phones of several staff, including the device operator.
3.4.2 Firefly larval breeding device
Initially, a device that the water is continuously supplied and the breeding trays are multitiered was built. This was a system in which pumps, tanks, and breeding trays were connected in series. Next, the results of the risk treatment analysis were incorporated into the new system.
Figure 6 shows the larval breeding device used in this study. The spring water issued naturally from the well was first collected in the pit, pressurized by the pump, and then monitored by the flow meter as it passed to the tank. The flow rate was controlled by the valve. The flow rate is accurately measured by the ultrasonic flowmeter. If the water flow rate is abnormal, the water flow detection system was able to immediately detect it and issue an alarm. The overflowing water from the tank was supplied to the tray (W 600 mm × L 900 mm × H 200 mm), and the water flowed down due to the difference in elevation. To make the water in the breeding tray mildly alkaline, 2–5 cm-long corallites were laid at the bottom of the tray up to a depth of ~ 30 mm. As the spring water has low dissolved oxygen, air was supplied to all the tanks and breeding trays to increase oxygen. The device consisted of two units (Unit A and Unit B), and each of them was configured with one tank and six breeding trays. With multiple breeding trays, local accumulation of the larvae was prevented. In Fig. 6, FI and TI represent the flowmeter (ultrasonic flowmeter, Maker: KEYENCE CORPORATION, Type: FD-Q50C) and the thermometer (resistance temperature detector, Maker: NETVOX TECHNOLOGY CO., LTD., Type: R718B2), respectively. Thermometers were installed to record the water temperature and detect anomalies. Temperature data was stored in the cloud via the Internet and was able to be referenced at any time.
3.4.3 Firefly larval breeding, egg laying, and hatching eggs
Dozens of male and female L. cruciata collected around the stream in the forest near chemical factories were placed in the egg-laying boxes lined with sphagnum moss to allow them to lay eggs. The hatched larvae dropped into the tank under the box. A total of 3600 hatched larvae were obtained. The number of eggs laid by one female was 0–500 and with an average of ~ 200 eggs.
Device operation with water
For ~ 3 weeks, 15 L/min/unit of water was supplied to the device to test the operation of each device and to clean the device. During this time, it was confirmed that there were no abnormalities in the calibration of the flowmeter, the indications of the thermometer, and other instrumentation.
Black snails (S. libertina), which is a thin and long snail living in freshwater in East Asia, are prey for the firefly larvae. Black snails living in the same ecological area as the fireflies were used. It is advised that the larvae need to be given black snails according to their size. Therefore, Black snails are finely crushed and fed as the prey. In this study, we tried to breed black snails by putting 500 g of black snails with a total length of 10–15 mm into each tray (500 g × 12 trays) and exposing them to sunlight. After 2–3 weeks, a significant number of juvenile snails were found.
Freshly hatched larvae are small (length 1.5 mm, width 0.3 mm) and they live in water with temperatures ranging from 10 to 20 °C . Hence, the water flow rate was determined to prevent flowing away of larvae and to keep tray water temperature less than 20 °C. The water flow rate was set at 5.5 L/min/unit. Approximately 400 hatched L. cruciata larvae were placed on each breeding tray in mid-July.
To reduce the risk of larval loss, it was important to control the following: (1) water temperature and (2) flow rate of the supply water, (3) dissolved oxygen in the water, and (4) pH of the water. Figure 7 shows the data from the breeding period (July–December 2020).
The temporal change of the water temperature is shown in Fig. 7a. The temperature in the water tank during the operation was between 16.0 and 17.5 °C, which was almost the same as that of the spring water from the well (not shown in Fig. 7). The installation of heat insulator on the piping was effective in suppressing the effects of heat conduction and convective heat transfer from the outside atmosphere on the water piping between the well and the water tanks and on the water tanks, limiting temperature rise. Figure 7a also shows the average temperature of the measurements at three locations each in Units A and B. The reason why there was a slight difference between them and the water tank temperature is the effect of heat transfer. There was a moment in late July when the temperature suddenly increased. This was caused by a trouble in the electricity supply system, which in turn caused an instantaneous drop in voltage, tripping the water supply pump. As a result, the temperature in the breeding trays was increasing as per outside air temperature (33 °C). In this instance, the flow rate abnormality notification system discussed in risk treatment was triggered. Due to operator intervention, the rise in water temperature could be suppressed below the upper limit temperature. It is difficult to determine the upper limit temperature uniformly. Here, we consider that it is rare for the temperature to exceed 30 °C in a natural running water system, and therefore, set 30 °C as the upper limit temperature. Aside from this occasion, the flow rate of the supply water was stable, as shown in Fig. 7b, and there was no problem with the flow control. The water flow rate was increased to 7 L/min/unit after late August. This is to promote gill respiration of the larvae. At this time, it was confirmed that the larvae did not flow away. In this study, the water flow rate was determined as 5.5 and 7 L/min/unit, as described above, but there seems to be opportunities for optimization.
As the water was spring water from underground, the dissolved oxygen (DO) in the well was low (70–75% of saturated concentration). However, it was eventually increased up to the saturated concentration by supplying air (see Fig. 7c). Regarding pH, it was neutral to slightly alkaline (between 7 and 8) (see Fig. 7d). The actual values under the aforementioned conditions were within the range of firefly breeding conditions, which was presented by Furukawa .
500 g/tray of black snails (6–18 mm long) were added in August and October. Since larvae may have invaded inside the vacant shells of black snails, the empty black snails had not been removed.
Release of larvae into the forest near chemical factories
By December, the larvae grew up to the size between 25 and 35 mm (Fig. 8). Approximately 1500 larvae were collected according to their weight. All of them were released into the forest near the chemical factories. Furthermore, ~ 3600 hatched larvae had been charged into the breeding device, so the yield was 42% (= 1500/3600). This is equivalent to 125/300 on average in one tray (twelve trays were used). Here, the yield is defined as follows: (the number of mature larvae) / (the number of hatched larvae) × 100.
Luciola cruciata is an aquatic firefly native to Japan, but it has been decreasing in prevalence owing to environmental changes. As a result, the custom by which Japanese people used to watch fireflies in early summer is being lost. Activities to bring back the lost fireflies have begun. This study concerned the stability of the operation of devices used for the breeding of aquatic firefly larvae. First, using the safety evaluation method used in petrochemical plants and other venues, we identified that the risk of unstable operation of devices must not be materialized, as it can lead to the death of the larvae. Next in FTA, analysis was conducted with the larval death as the top event to clarify the larval death scenario and calculate the probability of occurrence of the death. Through FTA, it was found that eliminating the mistakes in condition monitoring such as flow rate and temperature by the operators contributes to the reduction of the occurrence probability of larvae’s death. It was suggested that the prevention of human error (HE; errors in condition monitoring by the operator) was more effective than improvements in other basic events.
According to the categorization done by Reason , HE takes a limited number of forms. The following is the summary of the HE categories according to Reason. “Slip”: The plan is appropriate but fails to execute the action according to the plan. It is a visible action failure. “Lapse”: More internal phenomenon, which is often equivalent to a memory failure. “Example”: The status of your work is forgotten because someone talks to you while you are working. In the next category, the action is executed perfectly according to the plan, but the plan itself is inappropriate to achieve the intended objective. In this case, the failure occurs at a higher-level thought process (evaluation of given information, planning, formulation of intention, and the assessment of the effect expected from the action). These errors are defined as mistakes and can be further divided into two sub-groups, namely, “rule-based mistakes” and “knowledge-based mistakes.” Rule-based mistakes include the wrong application of normally correct rules, the application of wrong rules, and failure to apply good rules (violation). Knowledge-based mistakes occur when one runs out of a series of prepared solutions and has to instantly come up with the solution to the problem. It is the process when errors most likely occur.
It is thought that the HEs in condition monitoring by the operator examined in this study occur as a result of slip, lapse, or rule-based mistake. A series of habitual actions are controlled without paying attention. When a specific series of actions are executed frequently, action commands from humans can gradually be simplified. By relying on a schema for control in such a way can make the task more efficient and reduce the mental exhaustion caused by the task, but there are also disadvantages. To change the actions or thought process that became a routine, an intervention from a conscious control mode and/or direction from someone else is necessary. It is the lack of intervention and direction that are the main causes of the slips and the lapses. In addition, the preparation of an operation manual, training using the manual, and the monitoring of the status of education and execution situation are useful for preventing rule-based mistakes.
The results of this study provide a firefly larval breeding system that can reduce the death risk by reducing the occurrence of the aforementioned HE, constructing main equipment such as the pump redundant, and controlling the facility. As shown in simulation for pump stoppage, there is little time to spare when the water supply is stopped. Thus, a mechanism to constantly monitor the flow that starts an alarm remotely when a problem occurs was installed to reduce the larval death risk.
The following are some considerations regarding the ecology and breeding of fireflies:
The number of eggs laid by one female in this study was 0–500 and with an average of ~200 eggs. According to previous reports, the number of eggs laid by one female was 300–500 , 400–500 , and 500–1000 . The values in this study were almost the same as those in previous reports; some females did not lay eggs, probably because they were in the after-spawning stage.
Upper limit of water temperature during larval breeding
For the reasons described in the results, the upper limit of the water temperature was set to 30°C. Several studies have addressed the upper limit of water temperature. For example, according to Ohba , larvae can easily live at a water temperature of ~20 °C. In the breeding environment, even at water temperatures of ~30 °C, larvae can survive as long as air is blown into the breeding water. Yokohama City Pollution Research Institute and Yokohama Hotaru no Kai  stated that larvae do not die immediately even if the temperature exceeds 25 °C, but it is necessary to keep the temperature at a maximum of ~25 °C. The upper limit of the water temperature is not clear. Knowing the relationship between water temperature and mortality will be an important consideration in future research.
Number of mature larvae recovered
In natural rivers, there larval density is approximately one larva per 100 cm2 . The tray area (5400 cm2) of the device used in this study therefore corresponds to 54 larvae. Indeed, 125 larvae were obtained, almost double the expected amount. In addition, we compared the results obtained from the device in this study with those obtained from conventional devices; these are summarized in Table 5. Data from conventional devices were derived from previous breeding studies. The breeding method using the conventional device was as follows: aquarium containers (52 L, 3–5 sets), tray (14 L, 2–6 sets), with air blowing and water exchange once a day.
Comparisons between the method used herein and the conventional method indicate that an improvement was seen in the yield using the present method. Ohba  noted that it is possible to raise 150–200 larvae in a 60.5 cm × 30.3 cm × 45.0 cm (0.0825 m3) aquarium. In this study, 125 larvae were obtained in a tray of 60 cm × 90 cm × 15 cm (0.081 m3), suggesting that quantitative improvement is possible. Regarding the growth period of larvae, the present method was implemented from July to December whereas the conventional method was implemented from July to March, enabling growth over a short period of time. The present method also exhibits excellent workload and operational stability. On the contrary, it is difficult to observe larval growth because the larvae are hidden under coral stones; in the conventional method, it is easy to observe them. An economic evaluation was also conducted. The capital investment cost of the present method was 1.5 M yen. The depreciation cost was calculated from the initial investment cost with the depreciation year set to ten years. In addition, labor costs, such as monitoring and water changes were calculated as follows:
Device in this study: Data monitoring on PC and checking for abnormalities in the device, 15 min/day for 6 months at a labor cost unit price of 1000 yen/h (15/60 h/day × 30 days × 6 months × 1000 yen/h).
Conventional device: Water change and cleaning work, 1 h/day for 9 months at a labor cost unit price of 1000 yen/h (1 h/day × 30 days × 9 months × 1000 yen/h).
As a result, the present method is found to be advantageous in terms of cost.
The possible future challenges related to this study are:
Multiplication of the water supply anomaly detection
The breeding device used in this study adopted the system that detects abnormality in water supply with a flowmeter and sounds an alarm. When the flowmeter malfunctions, two issues emerge. One is the issue of “failed activation,” which is the failure to activate the alarm even when decrease in the water flow occurs. The other is the issue of “malfunction,” which detects a decrease and activates the alarm even though the flow is normal. The failed activation especially can cause a problem as it delays the discovery of problems. The possible measures to counter these problems include the redundancy of the sensor for anomaly detection to improve the reliability of the system. For instance, a thermometer can be installed in a tray to the aforementioned water supply flowmeter to make it 1 out of 2, or duplicate the thermometer to make it 2 out of 3 using three thermometers.
Improvement in mature larval yield
Whether the optimum conditions for water-stopping systems, which have been the subject of studies, can be applied to running water systems, such as this method, is unclear. Research on the optimization of the quality of water used for larval breeding (water temperature, DO, pH, etc.) would contribute to an increase in mature larval yield.
In this study, the water temperature was stable and there was less seasonal variation than in nature. It is necessary to identify the effect of this on yield and the growth of adults.
Larvae immediately after hatching are small, and they could flow out from the breeding tray. This issue needs improvement by modifying the breeding device.
Improvement of the forest toward natural breeding of fireflies
Vegetation and water environment improvements at the forest for creating and maintaining the environment where fireflies can naturally breed are important. For example, Hosoya  reported that environmental improvements are useful for the return of fireflies, prevention of genetic diversity decrease and genetic disturbance. We consider that public awareness and participation regarding the efforts to protect the environment are necessary for the conservation of fireflies. In other words, voluntary activities that perceive environmental deterioration and work toward environmental improvement contribute to environmental conservation.
Regarding the operation safety of the device used for breeding the aquatic firefly larvae, the safety evaluation method used in petrochemical plants was first used to identify that the death of the larvae is a significant risk. Following this, FTA was used to clarify the larval death scenario using larval death as the top event for the analysis. Through scenario analysis, it was demonstrated that the reduction of mistakes in monitoring water flow and temperature conditions by the operator would have the greatest contribution to the reduction of the larval death probability. Moreover, the time allowed for the response was estimated from the water temperature increase by conducting an experiment that simulates the water supply flow. From these results, a firefly larval breeding system equipped with an alarm function for notifying the necessity for emergency measures was constructed. In addition, mature larvae were obtained through the firefly larval breeding using this system. The superiority of the breeding system of this study was shown from the viewpoint of the operational stability and economic efficiency.
Data will be made available on reasonable request.
Lewis S (2016) Silent sparks. Princeton University Press, Princeton
Yuma M (1993) Water of fireflies and water of human, Shinpyoron, Tokyo
Yuma M (2007) Effect of rainfall on the long-term population dynamics of the aquatic firefly Luciola cruciata. Entomol Sci 10:237–244. https://doi.org/10.1111/j.1479-8298.2007.00219.x
Kawano K (2017) Analysis of synchronized luminescence of Genji firefly in field habitat. Bull Firefly Museum Toyota Town 9:81–99
Fujiso T, Kawaguchi Y, Takegawa Y, Yabuhara Y, Yamashiro A (2017) Analysis of density determinate environmental factors of Semisulcospira libertina, and their correlation to observed number of Genji firefly (Luciola cruciate) in Kawata river, Misato, Yoshinogawa city, in Tokushima Prefecture. J Jpn Soc Civ Eng 73:II_373-II_377. https://doi.org/10.2208/jscejer.73.II_373
Suzuki H, Sato Y, Ohba N (2002) Gene diversity and geographic differentiation in mitochondrial DNA of the Genji firefly, Luciola cruciata (Coleoptera: Lampyridae). Mol Phylogenet Evol 22:193–205. https://doi.org/10.1006/mpev.2001.1046
Oba Y (2020) Luciferase. Enzyme Eng News 83:34–38
Yoshida A, Sakurazawa S (2014) Examining for relaxation effect of entrainment phenomena in blinks of firefly. IPSJ SIG Technical Report, 2014-HCI-158(3):1–4
Koya Y, Miyake T (2012) Actual condition of activity for release of aquatic life by educational institutions in Gifu Prefecture. Sci Rep Fac Educ Gifu Univ 36:25–30
Ohba N, Suzuki H (2019) Flashing behavior and mitochondrial DNA analyses of the firefly, Luciola cruciata in the Institute for Nature Study. Rept Inst Nat Stu 50:1–12
Ueda T, Kusamitsu N (2020) The impact of urbanization on perceptions of fireflies for people. Bull Ishikawa Prefect Univ 3:23–33
Ohba N (2015) Fireflies that inhabit paddy fields and their surrounding environment. Agric Hortic 10:985–994
Mayahara T (2009) Firefly monitoring in the area around Nihon University College of Bioresource Sciences. Nihon Univ Coll Mus Bioresour Sci 19:151–155
Nagai K, Kawashima I, Oizumi F, Horiuchi Y, Yagishita Y, Murayama S (2014) The report of occurrence individuals number of Luciola cruciata Motschulsky (Coleoptera: Lampyridae) at Ikuta-Ryokuchi Park in 2014. Bull Kawasaki Munic Sci Mus Youth 25:9–11
Yajima M (2015) Forty-year efforts to establish fire fly populations at the Imperial Palace: polymorphic patterns in larval development and life cycle. Jpn J Entomol (NS) 18:106–117
Kashio T (2013) Restoration of the local environment as a crystallization of local memories: a case from Moriyama, 10, Shiga Prefecture, Japan. Shiga University Research Center for Sustainability and Environment Annual Research Report, pp 13–25. https://www.shiga-u.ac.jp/wp-content/uploads/2020/04/kashio_nenpo10-2013.pdf
Nakamura S (2019) Collaborative activities for “SATOGAWA” restoration in urban area in case of the Zenpukuji River, Tokyo. J Jpn Soc Water Environ 42:99–102
Yokosuka Environmental Policy Department (2019) Approaching fireflies in the Miura Peninsula. Park Green Space 80:29
Shinohara S (2014) Ueda City’s “Firefly Map” project to “visualize” the waterside. J Sewerage Mon 37:18–22
Yoshioka E (2017) The development of the container for the rearing of the first instar firefly (Luciora cruciata) larvae. Kobe Yamate Univ Bull 19:105–114
Minami K (1983) A study of fireflies. Scientist Inc., Tokyo, Japan. (U.S. Army translation no 1–9415 (1966). I.D. no. 2204017766)
Kanda S (1981) Reprint. Hotaru, Scientist Inc., Tokyo
Furukawa Y (2021) Firefly encyclopedia. http://www.tokyo-hotaru.com/jiten/hotaru.html. Accessed 28 Feb 2021
Ohba N (1993) Firefly breeding and observation. Heart Publishing, Tokyo
OSHA (2000) OSHA 3132 process safety management. https://www.osha.gov/sites/default/files/publications/osha3132.pdf. Accessed 20 May 2021
Sumida H (2005) Safety management of petrochemical plants. Petrotech 28:284–288
Nadiri AA, Moghaddam AA, Tsai FTC, Fijani E (2013) Hydrogeochemical analysis for Tasuj plain aquifer, Iran. J Earth Syst Sci 122:1091–1105. https://doi.org/10.1007/s12040-013-0329-4
Li P, Qian H, Wu J, Zhang Y, Zhang H (2013) Major ion chemistry of shallow groundwater in the Dongsheng coalfield, Ordos basin, China. Mine Water Environ 32:195–206. https://doi.org/10.1007/s10230-013-0234-8
Kimura D, Uga S (2003) Epidemiology of Centrocestus (Trematoda: Heterophyidae) in the Chikusa River Basin, Hyogo Prefecture 1. Infection in the first intermediate host, the freshwater Snail Semisulcospira libertine. Jpn J Environ Entomol Zool 14:97–103
JANSI CFR (2016) Estimation of domestic general equipment failure rate considering uncertainty of failure number (1982–2010, 29 years 56 plants data). www.genanshin.jp/archive/failure_rate/data/JANSI-CFR-02.pdf. Accessed 13 May 2019.
Gertman DI, Blackman HS (1994) Human reliability and safety analysis data handbook. Wiley, New York
Williams JC (1989) Human reliability data, the state of the art and the possibility. In: Proc. reliability ’89, 1, UK, June 14/16
Reason J (1997) Managing the risks of organizational accidents. Ashgate Publishing Limited, New York
Yokohama city pollution Research Institute and Yokohama Hotaru no Kai (1986) Creating the firefly breath environment, Technical manual draft. Yokohama City Pollution Research Institute materials 71
Hosoya K, Maehata M (1994) Circumstances in endangered Japanese freshwater fishes with note on the protection. Bull Natl Res Inst Aquac 23:17–25
We express our deep gratitude to Dr. Tomoki Maekawa from the Corporate Research & Development of Asahi Kasei Corporation who offered us a great amount of support when installing the firefly breeding device. We also express our gratitude to Mr. Yoji Tominaga from the Fuji Engineering Section, Corporate Production Technology of Asahi Kasei Corporation, who supported us during the installation of the facility. Further, we acknowledge the anonymous reviewers for constructive comments that improved the manuscript. We thank Crimson Interactive Pvt. Ltd. (Ulatus)—www.ulatus.jp for their assistance in manuscript translation and editing.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Conflict of interest
All the authors declare that they have no conflict of interest.
Human and animals rights
This article does not contain any studies with human participants or animals performed by any of the authors.
Informed consent was obtained from all individual participants included in the study.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Sano, K., Fukuzono, T., Yokoyama, T. et al. Breeding and conservation of Luciola cruciata Motschulsky, 1854 (Coleoptera: Lampyridae) larvae in forests near chemical factories. SN Appl. Sci. 4, 307 (2022). https://doi.org/10.1007/s42452-022-05196-5