Poisoned trap trees: a positive or negative tool to control Ips cembrae?

Abstract

Ips cembrae is a destructive forest pest attacking European larch. In 2021 and 2022, effectiveness of poisoned trap trees, trap logs and slot traps baited with the Cembräwit in trapping I. cembrae was tested. The negative impact on the predators of genus Thanasimuswas evaluated. We found that poisoned trap trees and logs are more effective in trapping I. cembrae than slot traps. The aggregation of bark beetles on the trap log edge was significantly lower than in the central part, where the pheromone lure was placed. The application of poisoned trap trees is possible in cases of extreme outbreaks.

Keywords:

• alpha-cypermethrin
• Cembräwit
• forest protection
• Ips cembrae
• Thanasimus formicarius
• trap trees

Introduction

The large larch bark beetle (Ips cembrae (Heer)) is an important pest of European larch (Larix decidua (Mill.)), attacking the entire stem (Holuša, Resnerová, and Kula 2021). It occurs at altitudes ranging from 400–2400 m a.s.l. (Postner 1974; Grodzki 2020). In the past, local outbreaks have been recorded in Europe (Krehan, 2004; Krehan and Cech 2004; Grodzki 2008; Grodzki and Kosibowicz 2009; Arač and Pernek 2014; Grucmanová et al. 2014), for example, in various regions in the Czech Republic the most severe outbreak of I. cembrae was recorded between 2015 and 2019 (Špoula and Kula 2023b), resulting in 47 000 m³ of larch wood infested (Anonymous 2023). In contrast to Ips typographus, outbreaks occur more rarely, especially during drought periods (Resnerová et al. 2022). The pest starts to attack the stem of standing trees at the stem base and after the stem base is populated the infestation continues gradually up into the crown (Holuša, Resnerová, and Kula 2021; Špoula and Kula 2023b). At low population density, I. cembrae prefers to breed on windfalls and stem breakages, where it can multiply its population and attack healthy trees (Holuša et al. 2014). European larch represents the seventh largest volume among the forest tree species in Europe (Danek and Danek 2022) and is found in Europe at altitudes ranging from 180–2500 m a. s. l. (Da Ronch et al. 2016). Therefore, most European larch stands may be at risk as I. cembrae occurs at altitudes from 400–2400 m a. s. l. (Grodzki 2020). Although larch is planted less than Norway spruce (Picea abies (L.) Karst), it is considered an important tree species in European forestry (Ellison et al. 2005). After the spruce bark beetle outbreak in the Czech Republic larch has been studied as a tree species suitable for the reforestation of clearcut areas (Kulla and Sitková 2012), where it could replace Norway spruce under conditions of global climate change (Zeidler et al. 2022). Ips cembrae is an aggressive bark beetle of larch that requires the application of standard control and defense measures to protect larch stands (Resnerová et al. 2020). Standard trap trees with no pheromone lure felled 8 weeks before the swarming of I. cembrae (Holuša, Resnerová, and Kula 2021) or the use of poisoned tripods baited with Cembräwit (Resnerová et al. 2020) are effective to control its population. The identification of I. cembrae aggregation pheromone (Stoakley et al. 1978) allowed the application of control measures that include a pheromone lure in forest protection (Hlásny et al. 2019). Male I. cembrae emit a mixture of ipsdienol, ipsenol and isoprenol (Stoakley et al. 1978; Rebenstorff and Francke 1982) to aggregate other males and females of the species. Renwick and Dickens (1979) indicate a high chiral selectivity of the (–)-ipsdienol enantiomer for the synthesis of ipsenol by I. cembrae. At present, the commercial pheromone lure Cembräwit (©Witasek, Pflanzenschutz GmbH, Feldkirchen, Austria) contains a mixture of ipsenol, ipsdienol, methyl butenol and amitinol (Zühlke and Mueller 2007) and is applied for monitoring and mass trapping I. cembrae.

Slot traps, standard trap trees, poisoned lying trap logs and standing trap trees baited with pheromone lures are commonly used to monitor and reduce bark beetle populations in Europe (Grégoire and Evans 2004; Wermelinger 2004; Zahradník and Knížek 2007; Holuša, Resnerová, and Kula 2021). Slot traps are barrier traps baited with pheromone lures used to monitor the population density of bark beetles (Kasumovic, Hrasovec, and Jazbec 2016). Poisoned trap logs are logs with an optimal length of 4 m, poisoned with a contact insecticide applied immediately before the swarming of bark beetles and baited with a pheromone lure (Blaženec, Jakuš, and Mezei 2015). Poisoned trap trees are healthy trees with stems poisoned with an insecticide from the stem base up to 4–6 m and baited with a pheromone lure (Juha and Turčáni 2008; Šotola and Kula 2022). The effectiveness of poisoned trap logs and trap trees for trapping the spruce bark beetle (Ips typographus (L.)) has been studied by many authors (e.g. Richter 1991; Niemeyer 1992; Raty et al. 1995; Schlyter et al. 2001, Lubojacký and Holuša 2014b; Kula et al. 2022), but has not been studied for I. cembrae. An alternative method to poisoned trap trees can be a poisoned tripod, created by securing three or more poisoned 1–2 m long logs assembled in the shape of a tripod (Lubojacký and Holuša 2014a).

Insecticide applied on the stems of living trees is partly protected from vertical precipitation by the crown, which makes the effects of the insecticide last longer than on the poisoned trap logs (Juha et al. 2012). The insecticide application on trap trees reduces migration of bark beetles during ovipositon and creation of sister broods (Juha et al. 2012). Poisoned traps may have unlimited capturing capacity when the insecticide is effective (Blaženec, Jakuš, and Mezei 2015). The recommended height of insecticide application on the stems of trap trees is influenced by the ecology of bark beetle species and by the position of the pheromone lure on the stem (Šotola and Kula 2022). For I. typographus, Juha et al. (2012) recommend the insecticide application up to 5 m of stem above the ground, Blaženec, Jakuš, and Mezei (2015) up to 6 m and to place a pheromone lure at the stem base.

Poisoned trap trees have a negative effect on non-target insects, especially on predators and parasitoids of bark beetles (e.g. Lubojacký and Holuša 2013; Lubojacký and Holuša 2014a; Blaženec, Jakuš, and Mezei 2015; Kula et al. 2022). Skrzecz et al. (2015) emphasized the threat to adults of the genus Thanasimus of using the Storanet® insecticide net containing alpha-cypermethrin. Lubojacký and Holuša (2014a) described the mortality of T. formicarius on poisoned tripods and logs treated with the insecticide Vaztak 10 SC. Slot traps usually capture fewer non-target organisms (Babuder, Pohleven, and Brelih 1996; Grodzki 2007) than measures that involve the application of an insecticide (Koleva et al. 2012; Lubojacký and Holuša 2014a).

Thanasimus formicarius (L.) is a generally widespread predator of 16 genera of bark beetle e. g. Cryphalus, Dendroctonus, Hylurgops, Ips, Orthotomicus, Pityogenes, Pityographus, Pityokteines, Scolytus, Tomicus and Trypodendron) (Gauss 1954; Gidaszewski 1974; Bakke and Kvamme 1981; Weslien and Regnander 1992; Weslien 1994; Schroeder 2001, Sarikaya and Avci 2009; Pernek, Kovač, and Lacković 2020; Bracalini et al. 2021; Meshkova et al. 2022) that attack coniferous and deciduous trees (Pfeffer & Knížek, 1996). Thanasimus formicarius can detect (R)-, (S) ipsdienol, and (R)-, (S)-ipsenol (Bakke and Kvamme 1978; Bakke and Kvamme 1981; Etxebeste et al. 2012) as a kairomone to find its bark beetle prey (Kenis, Wermelinger, and Grégoire 2004; Warzée and Grégoire 2006). The aggregation pheromone of most bark beetle species of the genus Ips consists of ipsdienol and ipsenol (Marco-Contelles 2021). The flight activity of T. formicarius begins in April simultaneously with the swarming of bark beetles ­belonging to the genera Hylurgops, Tomicus and Trypodendron (Thomaes et al. 2018). Adults live from 4 to 10 months (Gauss 1954). During oviposition, females lay 106–162 eggs into bark beetle galleries (Weslien and Regnander 1992; Dippel et al. 1997). Thanasimus femoralis (Zett.) is a widespread predator of bark beetles that naturally occurs together with T. formicarius (Kerchev et al. 2022). Little is known about the ecology of T. femoralis, but authors characterize this species less abundant than T. formicarius (Thomaes et al. 2018; Meshkova, Vorobei, and Omelich 2022). Thanasimus femoralis is active during the entire growing season and its ecology is similar to T. formicarius (Kerchev et al. 2022)

The fact that I. cembrae starts to attack larch trees from the stem base was the impetus for our experimental design to test standing poisoned trap trees with a pheromone lure. The objectives of this study were: (i) to test the effectiveness of poisoned trap trees, trap logs and slot traps baited with Cembräwit for the control of I. cembrae; (ii) to test the effect of the pheromone lure position on the stem in the top/bottom part of the poisoned trap tree and on the edge/centre of the trap log, and (iii) to determine the response of species of the genus Thanasimus to poisoned trap trees, trap logs and slot traps baited with the Cembräwit.

Effectiveness of trap trees and poisoned trap logs for the control of and defense against I. cembrae will be the same. We hypothesize that the effectiveness of trapping I. cembrae adults will be unequal across the surface of trap trees and will be affected by the position of the pheromone lure on the trap trees and logs. Poisoned trap trees baited with the Cembräwit will be assessed for their impact on clerid predators (T. formicarius and T. femoralis) who will be attracted to the treated areas.

Materials and methods

Study sites

The research was conducted in 2021–2022 at 5 study sites in the Děčín Forest District (Sněžník, Děčín Region, Czech Republic) (Figure 1). The study sites (AI, AII, BI, BII and CI) were located in larch monocultures aged 36–42 years at altitudes ranging between 460–600 m a.s.l., with different volumes of salvage-logged wood infested by I. cembrae in the period from 2018 to 2020. The localities were selected according to the occurrence of I. cembrae in 2020 (Table 1).

Figure 1. Study sites in the Czech Republic (Copernicus Land Monitoring Service 2019).

Table 1. Study site characteristics.

Localities	GPS	Altitude	Stand age	FSTC^a	IW (m3)^b
		m a.s.l.			2018–2019	2020	2021	2022
AI	50.7869478N	600	42	6K^c	17	6	0	0
	14.0621175E
AII	50.7861303N	600	42	6K^c	0	0	0	0
	14.0630844E
BI	50.8122508N	510	36	5K^d	45	6	1	1
	14.1228736E
BII	50.8105917N	510	36	5K^d	0	0	0	0
	14.1240081E
C	50.8200232N	460	36	6P^e	137	2	0	0
	14.1249445E

^aFSTC – Forest site type complex (Viewegh 2003)

^bIW – infested wood by I. cembrae and logged in the period from the beginning of August to the end of March of the next year

^c5K – 450 to 650 m a.s.l.; annual mean temperature (AMT) 5.5 to 6 °C; annual mean precipitation (AMP) 800 to 900 mm; growing season (GS) 130 to 140 days, acidic soil

^d6K – (500 m on sandstones) 650 to 950 m a.s.l.; AMT 4.5 to 5.5 °C; AMP 900 to 1,050 mm; GS 115 to 130 days; acidic soil

^e6P – 650 to 950 m a.s.l.; AMT 4.5 to 5.5 °C; AMP 900 to 1050 mm; GS 115 to 130 days, stagnic soil

Climatic data were obtained from the Sněžník weather station (GPS: 14°5′7.2954ʺE 50°47′48.5874ʺN; 569 m a. s. l.). Average temperature was higher in 2022 (8.3 °C) than in 2021 (7.5 °C), but mean annual precipitation in 2021 reached 582 mm, and in 2022 only 364 mm. Total annual temperature in 2022 was higher than the climatological normal of the Sněžník weather station for 1989–2022 (7.5 °C) and total annual precipitation was lower in both years than the climatological normal of 819 mm (ČHMÚ 2022) (Figure 2).

Figure 2. Daily total precipitation and average daily temperatures recorded at the Sněžník weather station (ČHMÚ 2022).

Trap trees, trap logs and slot traps

At each of the study sites, eleven healthy larch (Larix decidua Mill.) trees were selected randomly and marked with a number from 1 to 11. The trees were then treated as follows: (1) a baited poisoned trap tree (BPT), (2) a non-baited poisoned trap tree (NPT), (3–5) three poisoned control trees, and (6–11) six non-poisoned control trees (Figure 3). Trees 1–5 were poisoned with the insecticide up to a height of 4 m (1% solution of Vaztak Active; Agrospol Czech, s. r. o.; active substance: alpha-cypermethrin 50 g^·l-1; application of 5 l·m⁻³ of wood, applied every month from April to September) according to Zahradníková and Zahradník (2016). Two catching frames (0.8 × 0.8 m) at a height of 0.2 and 2.0 m were placed on the stems of BPT and NPT. The catching frames were protected against birds by netting. The Cembräwit (©Witasek, standard dispenser) was placed at a height of 1.8 m above the ground on the stem of the BPT. Diameter in the middle of the stem (mean ± SD) was 22.8 ± 0.9 cm (BPT) and 22.3 ± 1.4 cm (NPT). Average distance between BPTs and NPTs was 3.6 ± 1 m.

Figure 3. Position of poisoned trap trees, logs, slot traps and control trees at study sites (FD Děčín, 2021–2022).

Other trees in the group (3 poisoned, 6 non-poisoned) with an average diameter of 23.4 ± 5.2 cm and an average distance of 5.2 ± 2.4 from BPTs and 4.9 ± 2.3 m from NPTs were observed for the ­presence of bore dust regularly during the growing season for possible infestation by I. cembrae. The tree groups at all sites remained consistent during the study (2021–2022).

At each study site one trap log (PL) (length = 4 m), with an average diameter at the midpoint of the stem of 17.4 ± 1.8 cm (2021) and 16.7 ± 0.6 cm (2022), was placed 0.3–0.5 m above the ground and 10–20 m from the forest edge at each of the study sites. The insecticide was applied along the whole length of the logs. The Cembräwit was placed in the centre of the logs. On the edge and in the central part of PLs catching frames (1 × 1 m) were placed, which were protected against birds by netting. In addition, one slot trap (ST) was placed 10–20 m from the forest edge at each study sites (Figure 4).

Figure 4. Baited poisoned trap tree (BPT) (left) and baited poisoned trap log (PL) (right).

At each replicate study site (AI, AII, BI, BII and C), 7 treatments were made: bottom (1) and top (2) of BPTs, bottom (3) and top (4) of NPTs, edge (5) and centre (6) of PLs, and STs (7) (Figure 5). Eight samples were collected every year from each treatment. The main variables tested were the number of captured adults of I. cembrae and T. formicarius by treatments in every replication site. At each study site, no data was collected from the three poisoned trees (numbers 3–5) and six non-poisoned control trees (numbers 6–11), as they were just used for the early detection of a potential infestation.

Figure 5. Different treatment at study site.

For BPTs and NPTs the number of captured adults of I. cembrae and genus Thanasimus spp. is the sum of adults captured in the bottom and the top part. The number of captured adults of I. cembrae and genus Thanasimus on the whole stem (4 m) of trap logs was calculated using the formula: N = {[(2 × adults captured on edge) + adults in the central part]/3} × 4. We hypothesize that both edges of PLs will capture the same number of adults. The catching frame in the centre captured adults from 1 m of trap log, so total captures on 3 m of trap logs were divided by 3 and the result was multiplied by 4 to interpolate the total captures on whole length of trap log (4 m).

Cembräwit dispensers (Witasek PflanzenSchutz GmbH) were attached to trap trees and slot traps on 16^th April 2021 and 16^th April 2022 to attract the overwintering generation and replaced on 27^th June 2021 and 11^th June 2022 for the summer generation of I. cembrae. Samples were collected from STs, NPTs, BPTs and PLs every 14 days in the period from 30^th April to 5^th August 2021 and 2022 (in total 8 samples from every replicate site per year). The collected adults were stored in a 75% ethanol solution. Numbers of I. cembrae, T. formicarius and T. femoralis adults in the samples were determined in the laboratory. The species T. formicarius and T. femoralis were identified according to the colour of theventral part of mesothorax (Thomaes et al. 2018).

Data analysis

Data analysis was performed in the TIBCO Statistica (version 14.0.0.15) software. The normality of data was tested using a Shapiro–Wilk test (SW). A non-parametric Kruskal–Wallis test (KW) was used to identify differences in capture numbers of I. cembrae, T. femoralis and T. formicarius of trap trees, logs and slot traps. The degrees of freedom were calculated by the formula: $\mathit{\text{df}}=(n-1)$, where n is the number of groups.

A Mann–Whitney U test (U) was used to identify differences in capture numbers between years, bottom and top parts of BPTs and NPTs and edge and centre of PLs. All tests were performed at the confidence interval of α = 0.05.

Results

Effectiveness of poisoned trap trees, trap logs and slot traps

During our study, trees around trap trees baited with Cembräwit were not infested by I. cembrae. Sums and numbers of adults captured in every sample showed that baited poisoned trap trees (BPT) and baited trap logs (PL) were most effective in capturing I. cembrae adults. (SW: W = 0.45, p = 0.0000; KW: H (df = 4, N = 320) = 144.41, p = 0.000) (Figure 6). According to the total number of I. cembrae adults captured by trap trees, logs and slot traps, the population density was similar in both years (2021: 20 426; 2022: 19 558 adults) (Table 2). The differences in captured adults between poisoned trap trees and slot traps were the same in both years (2021: SW: W = 0.38, p = 0.0000; KW: H (df = 4, N = 160) = 70.70, p = 0.0000; 2022: SW: W = 0.63; p = 0.0000; KW: H (df = 4, N = 160) = 73.65, p = 0.0000). Although the non-baited trap trees (NPT) were in close proximity to the BPTs, only a total of 675 I. cembrae adults were captured by them. Slot traps (ST) captured on average 28.25 ± 51.6 adults. There were no significant differences in the number of adults in the top and bottom part of BPTs (SW: W = 0.738, p = 0.0000; U test: z = 0.232, p = 0.816), so the effectiveness of BPTs in capturing adults was similar along the entire profile of the stem. No significant differences were found between the number of I. cembrae adults captured in the top and bottom parts of BPTs between the study sites (SW: W = 0.74, p = 0.0000; KW: H (df = 4, N = 160) = 7.22, p = 0.125) or between years 2021 and 2022 (SW: W = 0.73, p = 0.0000; U test: z = −0.68, p = 0.494). No significant differences were found in the numbers of captured I. cembrae adults in the bottom (4.6 ± 14 adults) and top (3.9 ± 6 adults) parts of NPTs (SW: W = 0.44, p = 0.0000; U test: z = 0.72, p = 0.48). The mean value of captured I. cembrae adults was 105 ± 127.3 (mean ± SD) in the bottom part and 95 ± 117.1 in the top part of BPTs. The mean number of adults captured by BPTs and NPTs was 202.4 ± 239 and 8.4 ± 13, respectively.

Figure 6. Number of I. cembrae adults captured by poisoned trap trees, logs and slot traps. Squares indicate medians, rectangles indicate interquartile range and whiskers indicate minimum and maximum values. (ST – slot trap; NPT – non-baited poisoned trap tree; BPT – baited poisoned standing trap tree; PL – baited poisoned trap log).

Table 2. Mean numbers of captured I. cembrae ± standard deviation (adults).

	Non-baited poisoned trap trees (NPT)		Baited poisoned trap trees (BPT)		Trap logs (PL)		Slot traps (ST)
Site	2021	2022	2021	2022	2021	2022	2021	2022
AI	4 ± 3.9	7.5 ± 10.5	136.75 ± 87.8	274.5 ± 312.9	43.8 ± 34.2	176.3 ± 195	1.8 ± 2.5	18 ± 20.1
AII	9 ± 12.4	19.1 ± 27.9	120 ± 63.4	295.6 ± 342.5	88.5 ± 82.9	192 ± 144.4	5.1 ± 4.9	59.6 ± 76
BI	18.4 ± 12.7	9.6 ± 6.7	297.8 ± 350.6	231.6 ± 234.2	300.2 ± 438.9	383 ± 385.1	21.6 ± 21.9	26.4 ± 25.5
BII	3.1 ± 2.8	7 ± 9.2	100.4 ± 75.6	123.9 ± 97	381.7 ± 580.4	195.8 ± 162.2	75.5 ± 106.1	19.5 ± 32.8
CI	4.4 ± 3.6	2.3 ± 2.2	190.6 ± 198.1	228 ± 226.4	695 ± 1074.5	151 ± 170.6	31 ± 35.8	23.9 ± 30.4

The number of captured adults was significantly higher (104.5 ± 166.9 adults) in the central part, where the pheromone lures were placed, than at the edges of the PLs (SW: W = 0.559, p = 0.0000; U test: z = −3.549, p = 0.0004) (Figure 7). The mean value of I. cembrae adults captured on the edge was 45.5 ± 93 (mean ± SD).

Figure 7. Number of I. cembrae adults captured on the trap logs. Squares indicate medians, rectangles indicate the interquartile range, circles indicate outliers and whiskers indicate minimum and maximum values.

Impact of trapping variants on Thanasimus spp.

Statistically significantly higher numbers of T. formicarius adults (SW: W = 0.503, p = 0.0000; KW: H (3, N = 160) = 72.24, p = 0.0000; 2022: SW: W = 0.61; p = 0.0000; KW: H (df = 4, N = 160) = 137.41, p = 0.000) (Figure 8) were captured by BPTs in both years compared with NPTs, PLs and STs (2021: SW: W = 0.49, p = 0.0000; KW: H (df = 4, N = 160) = 72.24, p = 0.0000; 2022: SW: W = 0.61; p = 0.0000; KW: H (df = 4, N = 160) = 69.31, p = 0.0000). A total of 4100 adults (2021: 2428; 2022: 1672 adults) of T. formicarius were captured by poisoned trap trees, logs and slot traps.

Figure 8. Number of T. formicarius captured adults by poisoned trap trees, logs and slot traps. Squares indicate medians, rectangles indicate the interquartile range, circles indicate outliers and whiskers indicate minimum and maximum values (Legend: see Figure 6).

The BPTs (32.38 ± 30.17) and NPTs (5.43 ± 6.77) differed in the mean capture numbers of T. formicarius adults (SW: W = 0.78, p = 0.0000; U test: z = −4.16, p = 0.00003) and statistically more adults were caught in their top parts compared with the lower parts (SW: W = 0.62, p = 0.0000; U test: z = −3.61, p = 0.0003) (Figure 9). More adults were captured in the centre of PLs than at the edges (SW: W = 0.425, p = 0.0000; U test: z = −2.223, p = 0.026). The total number of T. femoralis adults captured by BPTs, NPTs, PLs and STs was 83, with no significant differences between the treatments (SW: W = 0.44, p = 0.00001; KW: H (df = 4, N = 320) = 4.55, p = 0.21).

Figure 9. Number of T. formicarius captured adults in the bottom and top part of BPT and NPT. Squares indicate medians, rectangles indicate the interquartile range, circles indicate outliers and whiskers indicate minimum and maximum values (Legend: see Figure 5).

Discussion

Juha & Turčani (2008) illustrated the possibility of using poisoned trees to mass trap I. typographus, but they warned that the trees can be infested in the part of the stem that was not treated by insecticide. In our study, no trees around the Cembräwit baited trap trees were infested by I. cembrae, even the part of stem that was not treated by insecticide. This fact can be explained by the relatively low population density of I. cembrae on the study sites in 2021 and 2022, compared to 2020 (Špoula and Kula 2023a) At low population densities, bark beetles search for unhealthy trees, which are typically widely dispersed in forests under natural conditions (Grégoire and Evans 2004; Clarke 2012). Some species of bark beetle attack healthy trees during population outbreaks, when the numbers of beetles are high enough for a mass attack that can overcome the natural defense of healthy trees (Martinson et al. 2013). During an outbreak, I. cembrae may also attack healthy trees in the vicinity of the poisoned trap trees baited with the pheromone lure.

Raty et al. (1995) compared the effectiveness of poisoned trap trees and slot traps in the mass trapping of I. typographus. As in our study, the trap trees were more effective than slot traps. A higher effectiveness of PLs for catching the overwintering generation of I. cembrae was expected, because trap logs are artificially stressed after felling, which results in a higher emission of volatile substances from the trees (e. g. terpenoids) (Kalinová et al. 2014) attracting bark beetles (Netherer et al. 2021). However, no differences in the effectiveness of BPTs and PLs in capturing I. cembrae were recorded throughout the growing season. We assume that at a low population density of I. cembrae, the pheromone lure is the most important factor for aggregation of I. cembrae adults around the pheromone lure. That may explain why a total of 4.5 times more adults of I. cembrae were captured in the central part of the PLs, where the pheromone lure was placed, than on the edge. At a distance of 1–2 m from the pheromone lure, there was a significant reduction in the intensity of I. cembrae aggregation on the stems on the PLs. The capture efficiency of BPTs was balanced in the bottom and top parts of the stem, because the pheromone lure was placed closer to the bottom of the frame for the top part of BPTs.

We assume that the use of shortened 2 m poisoned logs will maintain the same effectiveness as with logs of 4 m in length. Poisoned trap logs baited with Cembräwit should be placed at a safe distance of 10–25 m from the forest edge to prevent infestation of healthy trees. According to our results, slot traps seem to be ineffective in capturing large numbers of I. cembrae compared to the levels achieved by BPTs and PLs. The same findings have been found regarding capturing I. typographus in slot traps, which are therefore only used for monitoring purposes (Kuhn, Hautier, and San Martin 2022).

All baited control and poisoned measures have an impact on non-target arthropods (Bakke 1989; Valkama, Räty, and Niemelä 1997; Galko et al. 2016), but no studies of the effects on entire populations of non-target organisms have yet been published. Although no studies on the capture of non-target arthropods with control measures for I. cembrae have been published yet, our results are similar to papers associated with poisoned measures installed against I. typographus, defining a major impact on T. formicarius and T. femoralis (e.g. Lubojacký and Holuša 2013; Skrzecz et al. 2015). For example, Lubojacký and Holuša (2014a) found higher numbers of adults of Thanasimus spp. in poisoned tripods than in slot traps. Kula et al. (2022) captured high numbers of Thanasimus spp. adults in poisoned spruce trap logs and tripods covered with Storanet insecticide nets. The effects of slot traps on T. formicarius during the trapping of I. typographus and I. sexdentatus were evaluated by some authors (e. g. Özcan, Eroglu, and Akinci 2011; Akkuzu et al. 2021; Heber et al. 2021; Šramel et al. 2021), who agree that the numbers of captured T. formicarius are affected by the composition of the pheromone lure.

In comparison to our study, the number of captured T. femoralis adults was higher in the studies by Lubojacký and Holuša (2014a) and Kula et al. (2022). The reason for the low number of captured T. femoralis adults in our study may be the lack of cis-verbenol in the Cembräwit (Zühlke and Mueller 2007), to which this species reacts with a higher intensity than to ipsenol and ipsdienol, unlike T. formicarius (Bakke and Kvamme 1981). At a low population density of I. cembrae, control measures are unwarranted, because of the impact on non-target predators (Werner, Hastings, and Averill 1983; Zahradník 2005; Zahradník and Knížek 2007; Lubojacký and Holuša 2011; Lubojacký and Holuša 2013). Felled trap trees can be used instead of poisoned trap logs and trap trees for the control of I. cembrae, because they have no impact on non-target arthropods, although they have limited capacity (Holuša, Resnerová, and Kula 2021; Špoula and Kula 2023b).

Conclusions

During the 2-year experiment, poisoned trap trees and trap logs with Cembräwit were more effective in capturing large numbers of I. cembrae adults than standard slot traps. There were no differences in the effectiveness of trap trees and trap logs throughout the growing season. The numbers of I. cembrae adults captured on the poisoned trap logs were not the same over the entire surface and the position of the pheromone lure influenced aggregation of I. cembrae. In forestry practice, the poisoned trap logs (4 m) can be shortened to 2 m while maintaining their efficiency. We assume that using trap logs would be safer in forestry practice, because trap trees (i.e. healthy trees with pheromone lure) might get infested by I. cembrae on the portion of stem not treated with an insecticide. The poisoned trap trees and logs had an impact on the predators of I. cembrae, especially on T. formicarius, while there was only a low occurrence of T. femoralis due to the composition of Cembräwit. Baited poisoned trees and logs installed for mass trapping of I. cembrae should be used only in extreme cases of outbreaks, similar to the situation with I. typographus.

Geolocation information

The research was conducted at 5 study sites around Sněžník (Děčín Region, Czech Republic) GPS: 0°47′28.5ʺN 14°05′13.7ʺE.

Acknowledgements

The authors would like to thank Dr. Martin Mullet (United Kingdom) for checking the English.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are openly available in “zenodo” (https://zenodo.org/) at https://doi.org/10.5281/zenodo.10420656.

Additional information

Funding

This work was supported by the Internal Grant Agency (IGA) of the Faculty of Forestry and Wood Technology, Mendel University in Brno under Grant IGA-LDF-22-IP-015. The authors would like to thank Internal Grant Agency (IGA) of the Faculty of Forestry and Wood Technology, Mendel University in Brno for funding the research.