Overwintering Migration of the Double-Spined Spruce Bark Beetle Ips duplicatus (Sahlberg, 1836) (Coleoptera; Curculionidae)

: The double-spined bark beetle, Ips duplicatus , is an example of a species whose economic importance has increased in many European countries over the past decade. Many aspects of its life cycle remain unknown, complicating its management. I. duplicatus adults overwinter in forest litter, which makes the removal of infested trees during winter ineffective. Therefore, this study focuses on the mechanisms and timing of I. duplicatus overwintering migration into the forest litter. Only 2.2% of adult I. duplicatus overwintered under the bark of spruce logs stored outdoors from August 2016 to March 2017. In all, 94.4% of the beetles emerged and burrowed in forest litter. Migration began in late August, peaked in mid-September, and lasted until mid-October. At that time, forest litter collected at the base of an infested tree contained an average of 154.7 ± 23.9 I. duplicatus . Catches of migrating adults on glue stripes proved that the beetles walked down the trunk on days when the average daily temperature reached 10–15 ◦ C. Based on the results, we suggest that trees infested with I. duplicatus that have completed their development into adults should be removed no later than September. Sanitation cutting in the following months is not effective against this pest.


Introduction
In recent decades, Central European spruce stands have been negatively affected by climate change. Rising average temperatures [1,2] and long periods of precipitation deficits [3][4][5] have increased the susceptibility of trees to bark beetle attack. The low fitness of host trees and the more frequent occurrence of windthrows [6,7] resulted in a surplus of breeding substrate and large-scale bark beetle outbreaks, which doubled spruce mortality in Central Europe [8,9]. In 2019, the volume of harvested wood infested by bark beetles was nearly 20 million m 3 in the Czech Republic [10,11], 3.1 million m 3 in Slovakia [12], 4.3 million m 3 in Austria [13], 1.4 million m 3 in Poland [14,15], and 30 million m 3 in Germany [16]. While the effects of climate change on spruce stands in Central Europe are clearly negative, insects such as poikilotherms can benefit significantly from rising temperatures. Thus, shifts in phenology associated with climate change have been observed in numerous insects, including bark beetles, which have contributed significantly to the overall extent of damage. In recent decades, the spring emergence of the spruce bark beetle, Ips typographus, begins up to one month earlier than in the 1990s [17]. These extended growing seasons provided more time for development, and voltinism increased along the entire elevational gradient [18]. In addition, some bark beetle species, such as the doublespined bark beetle, Ips duplicatus, which were previously considered insignificant and management was not established, have increased their damaging potential [19,20], and local outbreaks and shifts in distribution have occurred in recent years [21,22]. This trend has been observed in many European countries (Germany [23]; Austria [24]; Slovakia [25]; and Romania [26]). Climatic extremes combined with the emergence of previously unknown pests complicate the management of commercial forests in Central Europe, and protective measures against bark beetles on spruce need to be adapted to current environmental conditions to maintain their effectiveness.
Currently, the most effective protective measure against bark beetles on spruce is the timely removal of attractive material (broken trees and windthrows) and infested trees from stands [27][28][29][30][31][32][33]. However, due to the number of infested trees and the lack of capacity to process them, infested stands are often logged in the fall and winter, which reduces effectiveness, as bark infested with bark beetles falls off standing trees and a large percentage of beetles overwinter in forest litter [33,34]. The removal of such debarked trees does not reduce the bark beetle population density on site and the damage proliferates the following spring. In addition, it is not uncommon for trees to be infested by several bark beetle species simultaneously, especially I. typographus and I. duplicatus, but also Pityogenes chalcographus and Ips amitinus, which should be taken into account when planning protective measures.
The double-spined bark beetle, I. duplicatus, is one of the species of increasing economic importance [19,22,35,36] that spread from its original range in the Euro-Siberian taiga to the rest of Europe in the first half of the 20th century [37,38]. Its rapid spread was favored by climate change and negatively affected non-natural spruce stands at lower and middle elevations [22,26,39]. The first outbreaks of I. duplicatus in the Czech Republic were recorded in the 1990s [38,40,41], and the economic importance of the species continues to increase [19,21]. The life cycle of I. duplicatus is very similar to that of I. typographus. Adults of both species emerge from their overwintering habitats at the same time in April and May and infest spruce trees, where they mate, and the development of daughter generation occurs [21,42]. While I. typographus prefers sunlit and water stressed trees at the edge of the stand, I. duplicatus predominantly attacks the upper parts of shaded trees inside the stand and develops much faster, making their early detection and timely removal difficult [39][40][41]43]. In addition, I. duplicatus adults overwinter in forest litter and previous authors hypothesized that the beetles migrate in the fall [44,45]. Therefore, the number of adults of I. duplicatus in the soil cannot be estimated from the percentage of bark that has fallen from the tree, as is the case for I. typographus.
The objective of the present study was to determine the timing at which adult I. duplicatus migrate into the forest litter under conditions of mid-elevation Central European spruce forests, to confirm the process of their active migration into the forest litter microhabitat, and to determine the percentage of the migrating population. These findings could answer questions about the effectiveness of the fall and winter logging of spruce trees that are predominantly infested with I. duplicatus: does such logging reduce population density or is it ineffective, assuming most beetles have already migrated to forest litter? What is the latest time to harvest infested trees?

Experiment 1 Overwintering Habitat of Ips duplicatus
In August 2016, three spruce trees infested with I. duplicatus were felled in the locality of Kružberk (GPS: 49.8339764 N, 17.6220542 E) in Northern Moravia. The study site was located at an altitude of 550 m a.s.l. in a 40-60 years old spruce monoculture. Twenty logs of one meter length were cut from the trees and transported toČeské Budějovice. Only logs with approximately the same diameter (17 cm), bark thickness (0.5 cm), and infestation density (2 boreholes per dm 2 ) were selected for the experiment. Ten randomly selected logs (half of the samples) were completely debarked immediately after transport and the total number of adults/pupae/larvae of I. duplicatus in each log was calculated. The remaining logs were placed in wooden cages covered with polypropylene netting. The bottom of the cages was made of hardboard. To mimic natural conditions, a 10 cm layer of mineral soil and a 10 cm layer of forest litter were placed on the hardboard floor. The cages were kept outdoors over the winter. At the end of March 2017, the logs were removed from the cages, debarked, and both forest litter and mineral soil layers were searched for I. duplicatus adults. The number of bark beetles found was recorded for both forest litter and mineral soil. Air temperature was measured at 30-min intervals during the experiment using Comet data loggers (Comet Systems Inc., Rožnov pod Radhoštěm, Czech Republic).
Normality of data was checked using the Shapiro-Wilk normality test and data were analyzed by one-way ANOVA and post hoc Tukey test. Data were analyzed in GraphPad Software (Version 6.0.0, San Diego, CA, USA).

Experiment 2 Number of Ips duplicatus Adults Overwintering in the Forest Litter
The experiment was conducted in 2017 at the Slunečná site (GPS: 49.8349039 N, 17.4638422 E) in Northern Moravia, approx. 25 km from the study site of Experiment 1. The study site was located at an altitude of 600 m a.s.l. in a 20-40 years old spruce monoculture. At the time of the experiment, most of the spruce trees were infested with I. duplicatus and had been left on site due to lack of logging capacity. The bark of trees infested in May 2017 fell off naturally because it was damaged by the development, maturation feeding and emergence of the F1 generation, and trees infested in late July/early August contained both parental beetles and the developing daughter generation. Forest litter at the base of infested trees was undisturbed by logging, so it could be collected weekly from late July through mid-October. Air temperature was measured at 30 min intervals during the experiment using Comet data loggers (Comet Systems Inc., Rožnov pod Radhoštěm, Czech Republic). The litter was removed with a shovel to a depth of 10-15 cm, usually to the hardened mineral soil. The collected litter was placed in ventilated plastic boxes and brought to the laboratory. On each sample day, litter was removed from around 10 trees. In the laboratory, litter was placed into barrel photoeclectors in a climate-controlled culture room with a controlled temperature of 22 • C, humidity of 60%, and photoperiod of 18:6 (light:dark). Photoeclectors were checked daily and the number of emerging beetles was recorded. If no beetles emerged for 14 consecutive days, the sample was removed from the photoeclector and discarded.
Normality of the data was tested using the Shapiro-Wilk normality test and data were analyzed by one-way ANOVA and post hoc Tukey test. Data were analyzed in GraphPad Software (Version 6.0.0, San Diego, CA, USA).

Experiment 3 Timing of Migration to the Forest Litter
In early August 2020, 10 trees infested with I. duplicatus were selected in the locality of Lísek in the Vysočina region (GPS: 49.5294886 N, 16.1620386 E), approx. 150 km from the study site of Experiment 1. The distribution of experimental sites illustrates the interannual shift of the outbreak of I. duplicatus. The study site was located at an altitude of 650 m a.s.l. in a 60-80 year old spruce monoculture. On each of the 10 trees, two (one at ground level and one at 2 m height) 10 cm wide strips of tree protection glue (Forestina, Mnichov, Czech Republic) were applied around the trunk. Bark beetles migrating from the trees into the forest litter were glued to the traps and collected weekly from mid-August to late October. Adhesiveness of the traps was checked on each collection day, and if it was reduced by dust and weather, the adhesive layer was renewed. Air temperature was measured at 30-min intervals during the experiment using Comet data loggers (Comet Systems Inc., Rožnov pod Radhoštěm, Czech Republic).
Normality of the data was tested using the Shapiro-Wilk normality test and the data were analyzed with the Kruskal-Wallis test and the Dunn's multiple comparisons test. Data were analyzed in GraphPad Software (Version 6.0.0, San Diego, CA, USA).

Experiment 1 Overwintering Habitat of Ips duplicatus
The average number of adults in the 10 one-meter logs debarked in August 2016 after being transferred from the field was 668.5 ± 99.98. At this time, the logs contained no immature stages of I. duplicatus. The bark of spruce logs overwintered outdoors did not fall off when debarked in March 2017. However, the number of adults recorded in the logs was significantly lower (ANOVA, p < 0.0001). A log contained an average of 8.5 ± 6.6 I. duplicatus adults and most beetles (365.7 ± 91.1) were found in the forest litter at the bottom of the cages. Only a few bark beetles managed to penetrate the mineral soil at the very bottom of the cage. On average, the mineral soil contained 13.2 ± 6.8 adults (Figure 1). Overall, 58% of the beetles were found compared to the August analysis numbers. Of these, 94.4% of I. duplicatus adults were found in forest litter, 3.4% in mineral soil, and 2.2% in logs. Temperatures during the experiment were relatively high. The only period when temperatures were below 0 • C was characterized by snow cover, which isolated the adults that migrated to the soil litter ( Figure 2).

Experiment 1 Overwintering Habitat of Ips duplicatus
The average number of adults in the 10 one-meter logs debarked in August 2016 after being transferred from the field was 668.5 ± 99.98. At this time, the logs contained no immature stages of I. duplicatus. The bark of spruce logs overwintered outdoors did not fall off when debarked in March 2017. However, the number of adults recorded in the logs was significantly lower (ANOVA, p < 0.0001). A log contained an average of 8.5 ± 6.6 I. duplicatus adults and most beetles (365.7 ± 91.1) were found in the forest litter at the bottom of the cages. Only a few bark beetles managed to penetrate the mineral soil at the very bottom of the cage. On average, the mineral soil contained 13.2 ± 6.8 adults (Figure 1). Overall, 58% of the beetles were found compared to the August analysis numbers. Of these, 94.4% of I. duplicatus adults were found in forest litter, 3.4% in mineral soil, and 2.2% in logs. Temperatures during the experiment were relatively high. The only period when temperatures were below 0 °C was characterized by snow cover, which isolated the adults that migrated to the soil litter ( Figure 2).

Experiment 2 Number of Ips duplicatus Adults Overwintering in the Forest Litter
The number of beetles that emerged from the forest litter differed between sampling dates (ANOVA, p < 0.0001). No beetles appeared in the photoeclectors in late July and early August. The first I. duplicatus adults emerged from litter collected on 21 August 2017. There was an average of 7 ± 4.8 beetles. From that date on, average catches increased with

Experiment 2 Number of Ips duplicatus Adults Overwintering in the Forest Litter
The number of beetles that emerged from the forest litter differed between sampling dates (ANOVA, p < 0.0001). No beetles appeared in the photoeclectors in late July and early August. The first I. duplicatus adults emerged from litter collected on 21 August 2017. There was an average of 7 ± 4.8 beetles. From that date on, average catches increased with each sampling. In early September, the average number of beetles emerged reached 50.2 ± 22.6 and doubled to 105.8 ± 21.8 during the next collection date. Another increase by half was recorded during the penultimate and final sampling dates, when forest litter samples contained 152.4 ± 152.4 and 154.7 ± 23.9 I. duplicatus adults, respectively (Figures 3 and 4). Generally, the first adults emerged from the soil litter after 10 to 14 days. The timing of emergence did not differ among samples.

Experiment 2 Number of Ips duplicatus Adults Overwintering in the Forest Litter
The number of beetles that emerged from the forest litter differed between sampling dates (ANOVA, p < 0.0001). No beetles appeared in the photoeclectors in late July and early August. The first I. duplicatus adults emerged from litter collected on 21 August 2017. There was an average of 7 ± 4.8 beetles. From that date on, average catches increased with each sampling. In early September, the average number of beetles emerged reached 50.2 ± 22.6 and doubled to 105.8 ± 21.8 during the next collection date. Another increase by half was recorded during the penultimate and final sampling dates, when forest litter samples contained 152.4 ± 152.4 and 154.7 ± 23.9 I. duplicatus adults, respectively (Figures 3 and 4). Generally, the first adults emerged from the soil litter after 10 to 14 days. The timing of emergence did not differ among samples.

Experiment 3 Timing of Migration to the Forest Litter
The number of I. duplicatus captured in glue traps differed significantly between collection dates (Kruskal-Wallis test, p < 0.0001). No beetles were captured during the first month of the experiment. The first specimens of I. duplicatus were caught on 1 September, but only on four of 10 trees. Two beetles were caught on each tree, all on the upper glue

Experiment 3 Timing of Migration to the Forest Litter
The number of I. duplicatus captured in glue traps differed significantly between collection dates (Kruskal-Wallis test, p < 0.0001). No beetles were captured during the first month of the experiment. The first specimens of I. duplicatus were caught on 1 September, but only on four of 10 trees. Two beetles were caught on each tree, all on the upper glue stripes. During the next two samplings, the average number of beetles captured increased to 5.3 ± 5.2 and 9.9 ± 10.5 adults per tree, respectively. Thereafter, the number of emerging beetles decreased to 7.2 ± 4.5 per tree in late September and 4.9 ± 4.0 per tree on October 1, when the last beetles were captured. No beetles were found in the traps after this date (Figures 5 and 6). Most of the captured I. duplicatus were on the upper glue stripes.

Discussion
Ips duplicatus is a species whose economic importance in Europe has increased in recent years [36]. Although its range was originally restricted to the Fennoscandia region

Discussion
Ips duplicatus is a species whose economic importance in Europe has increased in recent years [36]. Although its range was originally restricted to the Fennoscandia region and elevations up to 600 m above sea level [38], this species spread further westward during the 20th century. In the 21st century, I. duplicatus continues to spread westward Figure 6. The average number of beetles (+SD) that emerged from infested trees and were caught in glue traps. Different letters above columns indicate statistically significant differences (p < 0.05) (Dunn's multiple comparisons test).

Discussion
Ips duplicatus is a species whose economic importance in Europe has increased in recent years [36]. Although its range was originally restricted to the Fennoscandia region and elevations up to 600 m above sea level [38], this species spread further westward during the 20th century. In the 21st century, I. duplicatus continues to spread westward [36,46,47] and also to higher elevations, having been recorded at elevations above 800 m a.s.l. [48]. I. duplicatus usually co-occurs with I. typographus, and in some areas 80% of all infestations were caused by this species [19]. Currently, I. duplicatus is considered an invasive bark beetle species in European spruce forests (CABI 2019), occurring together with I. typographus in many locations [43,49]. When both species co-occur, the most effective method of forest protection is the timely removal of attractive breeding material (cuts, upheavals) and the regular search for infested trees and their removal [50]. Both of these methods are efficient if carried out before the new generation emerges, which is usually 4 to 6 weeks after infestation [39,51]. However, it can be difficult to find infested trees in time, especially if they are infested only by I. duplicatus, since this species mainly attacks younger, shaded trees in the stand [39,42]. The higher developmental rate and lower temperature requirements of I. duplicatus compared to I. typographus often result in the new generation of beetles emerging from infested trees before needle color changes become visible [39,51]. Therefore, when searching for infested trees, the most important sign is not the color change of the needles, but the appearance of boring dust at the base of the trunks. I. duplicatus is bivoltine in most parts of its range, but in recent years an increasing number of cases have been observed in which four generations developed during one growing season [19,36,39]. The last established generation may overwinter under the bark of infested trees when low fall temperatures slow development [52]. However, most adults overwinter in forest litter around infested trees, where their presence has been recorded in late October and spring of the following year [44]. This is consistent with the results of Experiment 1, where 94.4% of the overwintering population was detected in the forest litter in the spring and few beetles were found in the logs. Most of the adults resided in the 10 cm deep layer of the forest litter, which is also consistent with the results of previous authors who reported findings of overwintering adults from a depth of about 10 cm [44,45]. This depth, in combination with the snow cover, is sufficient insulation against low winter temperatures, so that I. duplicatus does not have to invest too much energy in developing and maintaining cold resistance mechanisms compared to some other members of the genus Ips (e.g., Ips acuminatus) [53][54][55]. However, overwintering in the forest litter has a disadvantage in that the beetles cannot replenish their energy reserves by feeding, as do bark beetles that overwinter under the bark of attacked trees [56]. In Experiment 1, the beetles that resided in the logs fed during the winter because fresh boring dust was detected near the logs. Fresh dust was also found in March 2017 during analysis.
The timing of migration of I. duplicatus from trees to forest litter has not been described in the literature. From the results of Experiment 2, it appears that the first adults migrated around the turn of August and September. In the following weeks, the number of migrating beetles increased, but the main factor affecting the migration of I. duplicatus is temperature. During Experiment 3, more I. duplicatus were caught when the true daily average temperature reached 10 to 15 • C. The temperature dependence of migrations into the forest litter is also confirmed by the observation of geographic directional preferences in I. duplicatus. The majority of beetles overwintering in forest litter were found in samples collected east and south of infested trees, i.e., in locations where solar radiation increased soil temperature [44,45]. Zhang et al. [44] hypothesized that migration into forest litter is active by climbing down the trunk, which was demonstrated in Experiment 3. The capture of I. duplicatus in glue traps shows that the beetles do indeed climb down the trunk and immediately bury themselves in the forest litter near the infested tree. This confirms the exponential decrease in the number of I. duplicatus adults with distance from the base of the trunk [44].

Conclusions
Based on the migration of I. duplicatus into forest litter, beetles can be expected to leave infested trees by the end of August, when they have completed their development. Migration does not depend on bark peeling and cannot be estimated from the external condition of infested trees, as is the case with I. typographus [34,45,57]. The timely removal of infested trees is strongly recommended during this period. Note that the timing of removal is directly proportional to the number of beetles in the forest litter. In places where trees were removed in September and October, an intensive search for newly infested trees is required the following spring. It is also possible to use some other forms of forest protection methods, such as pheromone traps or standing trap trees made more attractive with pheromone dispensers. Felled, debranched trap trees commonly used against I. typographus are not attractive to I. duplicatus [58][59][60]. The use of felled trap trees with branches may be attractive to I. duplicatus in some cases [60], as I. duplicatus reacts to the volatiles emitted from the needles and thin branches when searching for trees suitable for attack. However, a recent study evaluates them as completely ineffective and does not recommend them for use in forestry practice [61]. However, it is necessary to determine the proportion of I. duplicatus and other bark beetle species, especially I. typographus, at a given site. I. typographus mainly overwinters under the bark of infested trees [56,62], so the removal of infested trees at a later date is effective against this species. Felled trap trees, branched or unbranched, are also very efficient in reducing its population density [36]. Therefore, when planning protective measures, it is necessary to evaluate the spectrum of bark beetles at the site, although distinguishing individual species in the field can be difficult because of their similar biology and infestation characteristics [36,43,49].

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy.