Introduction

Freshwater ecosystems tend to have the highest proportion of species threatened with extinction (Millennium Ecosystem Assessment 2005). In agro-ecosystems, rapid and intensive urbanization along with agricultural development has led to the environmental deterioration and loss of freshwater habitats, resulting in a significant decline in many populations of freshwater species (Allan and Flecker 1993). Rice paddy ecosystems serve as important habitats for freshwater and wetland animals and plants. In Japan, rice paddy fields have provided alternative habitats for many species inhabiting wet hinterlands of rivers (Kiritani 2000; Moriyama 1997). For example, a huge number of insect species have been recorded in rice paddy fields (Kiritani 2010; Kobayashi et al. 1973; Yano 2002).

However, recent agricultural intensification has caused biodiversity loss of various taxa in paddy fields (Katayama et al. 2015), and as a result, a number of species dwelling in paddy fields are now endangered (Hidaka 1998). Agrochemicals and farmland consolidation were identified as serious factors threatening biodiversity loss in paddy fields (Katayama et al. 2015). Pesticides have reduced biodiversity of stream invertebrates in Europe and Australia (Beketov et al. 2013). Insecticides may also have major impacts on biodiversity loss or population decline in insects, particularly aquatic insects (Tanaka 2004). Recently, factors threatening the populations of a few aquatic and semi-aquatic insects were identified. The population of dragonflies, Sympetrum frequens (Selys) (Odonata: Libellulidae), has dramatically decreased and insecticides such as neonicotinoids and fipronil, applied to rice-seedling nursery boxes, were shown to be responsible (Futahashi 2012; Jinguji et al. 2009, 2013; Nakanishi et al. 2018). Insecticide susceptibilities were reported in several aquatic/semi-aquatic insect species, e.g. odonate nymphs (Ishida and Murata 1992; Jinguji et al. 2009), Kirkaldyia deyrollei (Vuillefroy) (Hemiptera: Belostomatidae) nymphs (National Institute for Environmental Studies 1995), Appasus major Esaki (Hemiptera: Belostomatidae) nymphs (Konno 2000), Gerris (Gerris) latiabdominis Miyamoto (Hemiptera: Gerridae) (Konno 1999), Aquarius paludum (Fabricius) (Hemiptera: Gerridae) and Hydaticus grammicus (Germar) (Coleoptera: Dytiscidae) (Kanaoka et al. 1994). However, insecticide susceptibilities have yet to be investigated in many other aquatic/semi-aquatic insect species. Moreover, it has been reported that insecticide susceptibilities differ among taxa or species of aquatic/semi-aquatic insects (Tanaka 2004). Hence, to understand the impact of insecticides on aquatic/semi-aquatic insects, further evidence and detailed data of their effects have to be accumulated.

The hydrometrid marsh treaders are predatory semi-aquatic hemipterans that inhabit the water edges of ponds and paddy fields. They prey upon insects including rice pests such as planthoppers and other arthropods that move on or fall onto the water surface. The hydrometrids of 110 species belonging to seven genera, three sub-families were recorded worldwide and many species of them are distributed across tropics and subtropics (Smith 1988). In Japan, four hydrometrid species, Hydrometra procera Horváth (Hemiptera: Hydrometridae), H. albolineata (Scott) (Hemiptera: Hydrometridae), H. okinawana Drake (Hemiptera: Hydrometridae), and H. annamana Hungerford et Evans (Hemiptera: Hydrometridae), were known (Kawai and Tanida 2005). Hydrometra yasumatsui Miyamoto (Hemiptera: Hydrometridae) is a synonym of H. okinawana. And recently, another species, H. gracilenta Horváth (Hemiptera: Hydrometridae), has been found in Aomori Prefecture and Hokkaido (Usui and Hayashi 2010; Usui et al. 2016). The habitats of four Hydrometra species except for H. gracilenta involve paddy fields (Kiritani 2010). Hydrometra procera is common in paddy fields and ponds throughout Japan, however, its biology was not well known until recent years. Murata et al. (2007) and Murata (2009) evaluated the predatory abilities of this species by investigating its functional response to prey densities and revealed its life cycle, seasonal population trends, and prey composition in paddy fields of Kumamoto Prefecture. In addition, a survey conducted by Murata et al. (2014) regarding the wing forms and habitats of H. procera in Kumamoto Prefecture, showed that this species flourished in the organic paddy fields but was not observed in the conventional paddy fields. They also indicated that macropterous forms were more abundant than brachypterous forms in these fields, which suggested migration between habitats. For H. albolineata, H. okinawana, and H. annamana, only a few aspects of biology have been reported so far. The population of H. okinawana is distributed from the Kanto district southwards, inhabiting ponds, paddy fields, and forest wetlands (Kawai and Tanida 2005). Hydrometra annamana is found in the open waters including paddy fields in Amami-ohshima Island and southwards (Kawai and Tanida 2005).

Hydrometra albolineata had been widely recorded from Honshu, Shikoku, Kyushu, the Tokara Islands, and Amami-ohshima Island. However, its populations have substantially decreased since the 1960s; hence, this species was designated as a vulnerable species in the Red List of the Environment Agency of Japan (Ministry of Environment 2019). In the Red Data Books or the Red Lists published by the prefectural governments in Kyushu, H. albolineata has been designated as a critically endangered species in Oita Prefecture and Kumamoto Prefecture (Oita Prefecture 2011; Kumamoto Prefecture 2019), an endangered species in Fukuoka Prefecture (Fukuoka Prefecture 2015 b, a), and a vulnerable species in Miyazaki and Kagoshima Prefecture (Kagoshima Prefecture 2014; Miyazaki Prefecture 2015)Thus, this species is now found in quite restricted habitats throughout Japan and should be conserved. Of other hydrometrids, H. okinawana has been designated as a vulnerable species in Kumamoto Prefecture (Kumamoto Prefecture 2019), and a near threatened species in Tokushima Prefecture (Tokushima Tokushima Prefecture 2013). Hydrometra procera has been assigned as a near threatened species in Tokushima and Ehime Prefecture (Ehime Prefecture 2014; Tokushima Prefecture, 2013), suggesting that this species might become endangered in future.

To conserve these hydrometrids, it is important to understand the causes that have reduced their populations. Any role played by insecticides in reducing their populations needs to be investigated. To date, there are no reports of insecticide susceptibilities of hydrometrid species. In this study, we examined the susceptibilities of three hydrometrid species, H. procera, H. albolineata, and H. okinawana, to seven insecticides that have frequently applied to paddy fields in Japan.

Materials and methods

Insects

Adults of H. procera were collected from an organic paddy field located in the campus of Tokai University, Minami-aso Village, Aso Gun, Kumamoto Prefecture in June 2014. No agrochemicals or chemical fertilizers have been used in this paddy field since 2007. Adults of H. okinawana and H. albolineata were obtained from a pond located in Amakusa City, Kumamoto Prefecture in June 2015. For insecticide susceptibility tests, we used the third instar nymphs and adults which were raised in the laboratory from the eggs deposited by the field-collected adults. The three hydrometrid species were around third instar during July and August when insecticides were applied in paddy fields in Kumamoto Prefecture (Murata 2009), accordingly, we used the third instars for experiments. We used both the third instars and adults of H. procera for this study. For H. okinawana and H. albolineata, however, we examined only the third instars because sufficient number of individuals did not develop to adults.

We reared the insects, including the field-collected adults and their offspring nymphs, at 25 °C under a photoperiod of 16:8 (L:D) h. The insects were individually housed in plastic Petri dishes (9 cm diameter by 2 cm height) in which a moist filter paper was lined at its bottom and a small cup (made from a lid of a PET bottle) filled with moist paper was placed. The insects were supplied with one prey insect, i.e. a nymph/adult of planthopper, Sogatella furcifera (Horváth) (Hemiptera: Delphacidae), or an adult dipteran (chironomid, culicid, or agromyzid) daily, collected from the organic paddy field of Tokai University.

Insecticide susceptibility tests

For insecticide susceptibility tests, we used seven insecticides, i.e. an organophosphate [diazinon 40% emulsifiable concentrate (EC)], a synthetic pyrethroid (etofenprox 20% EC), two neonicotinoids [dinotefuran 20% wettable granules (WG), and imidacloprid 10% wettable powder (WP)], a carbamate (fenobucarb 50% EC), a thiocarbamate [cartap 75% water-soluble granules (WSG)], and an insect growth regulator (IGR: buprofezin 25% WP inhibiting chitin synthesis). Each insecticide was diluted by adding distilled water to a recommended concentration of 1/2000, except for cartap, which was diluted to 1/1500. We carried out the susceptibility tests using the method described by Tanaka (1999) and Tanaka et al. (2000). Briefly, the insects were dipped in insecticide solutions using a glass test tube (3 cm diameter by 10 cm length). A circle frame was made with a piece of wire adjusted to the open end of test tube and the wire frame was covered with fine polyester mesh (Fig. 1). An end of wire was extended from the frame and was used as a handle. The test insects were individually put into the test tube and the mesh-covered wire frame was attached to the open end of test tube. This end of test tube was dipped in the insecticide solution for 5 s, then, the test tube was placed on several sheets of filter paper to remove the solution. Distilled water was treated as control. Nine individuals of test insects were used for each concentration of each insecticide. The insecticide-treated insects were reared in the same way as before the insecticide treatment and their mortality was recorded 24 and 48 h after treatment. Buprofezin (an IGR), however, inhibits chitin synthesis and consequently the molting of insects. Accordingly, the buprofezin susceptibility of the third instar nymphs was determined by observing their mortality until they completed the molting to the fourth instar. The mean duration of H. procera third instar was 3.0 days (2–4 days) at 25 °C (Murata et al. 2007). The duration of third instar nymphs of other two species was the same to that of H. procera (Murata, unpublished). The buprofezin-treated insects were observed for 120 h, at maximum, after the treatment. To determine the mortality, insects that were dead, in agony, or could not walk were counted as dead. In addition, for buprofezin, abnormal molting and death just after molting were also regarded as dead. We evaluated the effect of each insecticide on the insects according to the hazard assessment classes with four categories under laboratory conditions by the International Organization for Biological Control and Integrated Control of Noxious Animals and Plants/West Palaearctic Regional Section (IOBC/WPRS) (Hassan 1992). It was based on the mortality (m) 48 h after the insecticide treatment for insecticides except buprofezin: not harmful (m < 30%), slightly harmful (30% ≤ m < 80%), moderately harmful (80% ≤ m < 99%), and seriously harmful (99% ≤ m). For buprofezin, hazard assessment was based on the mortality until the molting to the fourth instar was completed (third instar nymphs) or at 48 h after the buprofezin treatment (H. procera adults).

Fig. 1
figure 1

Test tube and the wire frame covered with fine polyester mesh used for insecticide susceptibility tests

Full size image

To investigate the effect of insecticides at concentrations lower than the recommended value, H. procera third instar nymphs were treated with six (cartap) or five (the other insecticides) concentrations. The median lethal concentration (LC50) was calculated by probit analysis (Bliss 1935) based on the observed mortality until the molting completion (buprofezin) or the mortality after 48 h of treatment (the other insecticides).

Results

No mortality was observed in the control (dipped in distilled water) after 48 h of treatment for three hydrometrid species. The mortality rate of third instar nymphs of H. procera was 100% after 48 h of treatment for three insecticides, i.e. etofenprox, imidacloprid, and fenobucarb, at a recommended concentration of 1/2000 (Table 1). For buprofezin at a recommended concentration of 1/2000, although the acute toxicity (mortality at 24 or 48 h of treatment) was low, the mortality until the molting completion was 100% (Table 1); all the nine nymphs were dead without molting by 72 h or 96 h. Based on the hazard assessment classes by IBPC/WRPS, effect of these four insecticides were found as seriously harmful. On the other hand, the insecticides which were evaluated as seriously harmful to the H. procera nymphs at lower concentrations than the recommended value were etofenprox up to a concentration of 1/100,000 (2 ppm), imidacloprid up to 1/20,000 (5 ppm), fenobucarb up to 1/5000 (100 ppm), and buprofezin up to 1/20,000 (12.5 ppm) (Table 1). The effect on the H. procera nymphs was evaluated as moderately harmful for cartap at a recommended concentration of 1/1500 (Table 1). Dinotefuran was classified as seriously harmful at a concentration of 1/5000 (40 ppm), although it was categorized as slightly harmful at 1/2000 (100 ppm) (Table 1). Of four insecticides in which the LC50 values were obtained, the LC50 value was lowest for dinotefuran, followed by fenobucarb, cartap and diazinon. These results indicated that the toxicity of etofenprox, imidacloprid, and buprofezin to the H. procera third instars was relatively high, while the order of toxicity of other test insecticides was dinotefuran > fenobucarb > cartap > diazinon. The mortality of H. procera third instars increased from 24 to 48 h after insecticide treatment in case of dinotefuran, imidacloprid, and cartap.

Table 1 Mortality of Hydrometra procera third instar nymphs, 50% lethal concentration (LC50) and hazard assessment of insecticides treated on the nymphs
Full size table

The harmful effect to the H. procera adults was the same as that to the third instars in etofenprox (100% mortality, seriously harmful), diazinon (slightly harmful), and similar in dinotefuran (slightly harmful to males, moderately harmful to females) (Table 2). In case of imidacloprid, fenobucarb, cartap, and buprofezin, however, the mortality of adults was lower than that of the third instars (Table 2). In dinotefuran, the hazard assessment classes were evaluated as different categories between sexes, however, the mortality itself of males and females after 48 h of treatment was similar.

Table 2 Mortality of Hydrometra procera adults and hazard assessment of insecticides treated on the adults
Full size table

The mortality of the H. okinawana third instar nymphs after 48 h of treatment was 100%, i.e. seriously harmful, for etofenprox, dinotefuran, imidacloprid, and fenobucarb (Table 3). For buprofezin, seven nymphs were dead without molting by 72 h; two nymphs had molted by 72 h, however, they were dead with unhardened body and bended abdomen. Thus, the mortality in buprofezin was 100%, and buprofezin was evaluated as seriously harmful (Table 3). In contrast, the mortality was 0%, i.e. not harmful, for cartap, while diazinon was evaluated as slightly harmful (Table 3).

Table 3 Mortality of Hydrometra okinawana third instar nymphs and hazard assessment of insecticides treated on the nymphs
Full size table

The mortality of the H. albolineata third instar nymphs was 100% for diazinon, etofenprox, dinotefuran, imidacloprid, and fenobucarb (Table 4). For buprofezin, two, three, and two nymphs were dead without molting at 24, 48, and 72 h of treatment, respectively; a nymph was dead during molting and a nymph was dead just after molting with unhardened body and bended abdomen by 72 h. Accordingly, the mortality in buprofezin to H. albolineata was 100%, and buprofezin was evaluated as seriously harmful (Table 4). Thus, the largest number of insecticides was seriously harmful to H. albolineata among the three hydrometrid species. On the other hand, cartap was evaluated as not harmful (Table 4).

Table 4 Mortality of Hydrometra albolineata third instar nymphs and hazard assessment of insecticides treated on the nymphs
Full size table

Discussion

Toxicity of insecticides to three hydrometrid species

Etofenprox (a synthetic pyrethroid), imidacloprid (a neonicotinoid), and fenobucarb (a carbamate) were found as seriously harmful to the third instar nymphs of the three hydrometrid species examined in this study based on the hazard assessment classes by IBPC/WRPS. Particularly, etofenprox was seriously harmful to the H. procera third instars at the lowest concentration among the insecticides tested. In addition, 100 ppm etofenprox was seriously harmful to the H. procera adults. In general, synthetic pyrethroids have a wide range of insecticidal spectrum, which makes it effective against a wide taxa of pest insects. However, it tends to be toxic to non-target taxa including natural enemies of pest insects. Arthropods inhabiting paddy fields, for instance, the giant water bug A. major was found susceptible to etofenprox (Konno 2000), and the natural enemies of rice planthoppers [spiders, the mirid bug Cyrtorhinus lividipennis Reuter (Hemiptera: Miridae), and the dryinid wasp Haplogonatopus apicalis Perkins (Hymenoptera: Dryinidae)] were also susceptible to synthetic pyrethroids (Tanaka et al. 2000). Neonicotinoids including imidacloprid also have a wide range of insecticidal spectrum. Imidacloprid was found toxic to some paddy-dwelling non-target insects, e.g. Sympetrum frequens (Jinguji et al. 2009), G. latiabdominis (Konno 1999), C. lividipennis, and H. apicalis (Tanaka et al. 2000). Fenobucarb also has been reported toxic to paddy-dwelling non-target insects, e.g. G. latiabdominis (Konno 1999), H. apicalis (Tanaka et al. 2000), stream-dwelling larvae of mayflies, Baetis thermicus Uéno (Ephemeroptera: Baetidae), Epeorus latifolium Uéno (Ephemeroptera: Heptageniidae), and Ecdyonurus yoshidae Takahashi (Ephemeroptera: Heptageniidae) (Tada 1998).

For buprofezin (an IGR) at the recommended concentration, no nymphs of the three hydrometrid species completed the molting to the fourth instar, i.e. 100% mortality, although the acute toxicity at 24 or 48 h was low. Thus, buprofezin was evaluated as seriously harmful. The IGRs, in general, have selective toxicity to specific arthropod taxa including pest insects but tend to be nontoxic against non-target taxa. There are few reports indicating that the IGRs caused high mortality in the natural enemies of the rice insect pests. Thus, a further detailed investigation is required to study the effect of buprofezin on the hydrometrid species in laboratory and fields.

Based on our results, dinotefuran (a neonicotinoid), diazinon (an organophosphate), and cartap (a thiocarbamate) were evaluated as seriously harmful to the third instars of two, one, and no Hydrometrid species, respectively, and effect of these three insecticides may be different among the species. However, to confirm these results, we should conduct further tests using more replications.

Differences in insecticide susceptibilities between adults and nymphs

This study examined insecticide susceptibilities in adults and the third instar nymphs of H. procera as there are no previous reports on differences in insecticide susceptibilities between the hydrometrid adults and nymphs. The harmfulness of etofenprox, diazinon, and dinotefuran on H. procera adults was similar to that of the third instar nymphs. For imidacloprid, fenobucarb, cartap, and buprofezin, however, the harmful effect on the third instars was greater than that on the adults. The insecticidal activity of buprofezin may be lower against adults than larvae, because the adults molt no longer. However, Yasui et al. (1987) reported that the hatchability of eggs laid by adults of Trialeurodes vaporariorum (Westwood) (Homoptera: Aleyrodidae) treated by buprofezin for longer than 24 h was extremely low. Tiwari et al. (2012) reported that buprofezin restrained the next generation of Diaphrina citri Kuwayama (Hemiptera: Liviidae), because buprofezin effectively suppressed their adult emergence, egg production, and egg hatchability. Therefore, further investigation is needed to examine influence of buprofezin on the next generation of hydrometrids such as decrease in fertility and increase in egg mortality.

Toxicity of insecticides at low concentration

The insecticides which were evaluated as seriously harmful to the H. procera nymphs at concentrations lower than the recommended value were etofenprox up to a concentration of 1/100,000 (2 ppm), imidacloprid up to 1/20,000 (5 ppm), fenobucarb up to 1/5000 (100 ppm), and buprofezin up to 1/20,000 (12.5 ppm). Thus, some insecticides, particularly etofenprox, imidacloprid, and buprofezin, were toxic to H. procera even at very low concentrations, suggesting their detrimental effects in the fields. For dinotefuran, the mortality of H. procera nymphs was higher at a concentration of 40 ppm than that at a concentration of 100 ppm. The effect of dinotefuran at different concentrations should be further tested with more number of test insects.

Effects of insecticides causing decline in hydrometrid populations

Etofenprox, imidacloprid, fenobucarb and buprofezin were evaluated as seriously harmful to all the three hydrometrid species tested in this study. Etofenprox has been heavily applied to the rice paddy fields in Japan for controlling rice leafhoppers, planthoppers, and stinkbugs. Imidacloprid has been frequently used primarily in rice-seedling nursery boxes for controlling rice leafhoppers, planthoppers, the rice water weevil Lissorhoptus oryzophilus Kuschel (Coleoptera: Erirhinidae), and the rice leaf beetle Oulema oryzae (Kuwayama) (Coleoptera: Chrysomelidae). Fenobucarb has been used in large quantities since the 1960s for controlling rice pests such as rice leafhoppers and planthoppers, although its use has been recently decreased (Ueyama 2013).

The IGRs, in general, have selective toxicity to particular insect pests but are nontoxic to non-target taxa. However, this study showed that buprofezin caused a high mortality in the third instar nymphs of the three hydrometrid species. The nymphs were dead before, during, or just after the molting to the fourth instar, indicating that buprofezin might inhibit the molting. Buprofezin has been used in large quantities since the 1990s for controlling rice pests (Ueyama 2013). We should further investigate whether buprofezin may have detrimental effect on hydrometrid populations in the fields.

Dinotefuran use has been recently increased for controlling rice leafhoppers, planthoppers, and stinkbugs. On the other hand, the use of diazinon has decreased in the recent years, although it has been heavily used since the 1950s (Ueyama 2013). These records of insecticide use indicated that the seriously harmful insecticides, particularly to the three hydrometrid species examined in this study, have been frequently applied to paddy fields in Japan in the past or are still in use.

The populations of the endangered species H. albolineata have drastically declined since the 1960s (Ministry of the Environment 2014). The organophosphate (diazinon) and carbamate (fenobucarb) in which the amount of use increased in the 1950s and 1960s, respectively, were seriously harmful to H. albolineata. These results suggest that the heavy use of these insecticides in paddy fields is one of factors causing the decline in H. albolineata populations.

Hydrometra procera and H. okinawana have been abundant and their habitats have not been restricted, compared to H. albolineata. Therefore, these two species were not designated as endangered species in the national and prefectural Red List/Red Data Book. However, these species were assigned as vulnerable or near threatened species in a few prefectures as mentioned earlier. Moreover, H. procera inhabited only organic paddy fields in Kumamoto Prefecture (Murata et al. 2014). Considering these facts, if the insecticides harmful to the hydrometrid species continued to be heavily applied in paddy fields, the two species may become endangered like H. albolineata.

To conserve the hydrometrids, it is required to investigate the habitat of each species in details and to identify environmental factors essential for their survival. In addition, we should survey their use of temporal, semi-permanent, and permanent habitats such as paddy fields and ponds, and migration between these habitats. These researches would enable us to estimate application and inflow of the harmful insecticides into the potential habitats of the hydrometrids. Especially for conservation of the endangered species H. albolineata, immediate action is required. Insecticides were present in higher concentrations than those previously considered in aquatic areas around paddy fields including irrigation ditches, drainage canals, and ponds (Furihata et al. 2019). Thus, it is required to prevent inflow of insecticides into the present habitats of H. albolineata. In addition, we should establish long-term conservation strategies such as sustainable management of specific environments.