Figures
Abstract
To escape or alleviate low temperatures in winter, insects have evolved many behavioral and physiological strategies. The purple stem borer, Sesamia inferens (Walker) is currently reported to be expanding their northern distributions and causing damage to summer maize in Xinxiang, China. However, their method of coping with the lower temperature in the new northern breeding area in winter is largely unknown. This paper investigates the overwinter site of S. inferens, and identifies the cold hardiness of larvae collected from a new breeding area in winter and explores a potential distribution based on low temperature threshold and on species distribution model MaxEnt. The results show that the overwintering location of the S. inferens population is more likely to be underground with increasing latitude and the population gradually moved down the corn stalk and drilled completely underground in later winter (February) in the north. The cold hardiness test shows the species is a moderate freeze-tolerant one, and Supercooling Points (SCP), Freezing Points (FP) and the incidence of mortality during the middle of winter (January, SCP: -7.653, FP: -6.596) were significantly lower than early winter (October) or late winter (March). Distribution in the new expansion area was predicted and the survival probability area was below N 35° for the Air Lower Lethal Temperature (ALLT50) and below N 40° for the Underground Lower Lethal Temperature (ULLT50). The suitable habitat areas for S. inferens with MaxEnt were also below N 40°. This study suggests the overwinter strategies of S. inferens have led to the colonization of up to a five degree more northerly overwintering latitude.
Citation: Huang J, Li G, Lei H, Fan C, Tian C, Chen Q, et al. (2020) Low-temperature derived temporal change in the vertical distribution of Sesamia inferens larvae in winter, with links to its latitudinal distribution. PLoS ONE 15(7): e0236174. https://doi.org/10.1371/journal.pone.0236174
Editor: Yonggen Lou, Zhejiang University, CHINA
Received: February 29, 2020; Accepted: June 30, 2020; Published: July 28, 2020
Copyright: © 2020 Huang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: J.H.’s visiting scholarship to the University of Exeter was funded by the Youth Exchange Funding of Henan Academy of Agricultural sciences. This research was supported by the National Natural Science Foundation of China (Grant No. 31401731) and Youth Funding of Henan Academy of Agricultural sciences (No.2020YQ09), Supported by the National Key Research and Development Program of China (Grant Nos. 2016YFD0300705, 2018YFD0200602, and 2018YFD0200605).
Competing interests: The authors have declared that no competing interests exist.
1 Introduction
Temperature is the main abiotic factor that determines the growth and breeding of ectotherms. The distribution of most insects is directly related to extreme temperature [1]. The minimum temperature in winter which determines the survival rate of insect wintering populations is an important factor limiting the potential geographical distribution of insects [2, 3], particularly in ectotherms, setting northern range limits [3–5]. The low temperature in winter constrains the behavioral strategies of colonial insects [6], such as migration, drilling holes in refuges or making a thick cocoon, to avoid winterkilling. Insects also avoid chill injury by regulating physiological metabolism and accumulation of cold-tolerant substances, usually resulting in a lower Supercooling Point (SCP) of the overwintering population in mid-winter [7]. The warming of the global climate has also led to poleward distribution in some species' ranges, and may result in insects evolving new overwintering strategies in different breeding areas [8, 9].
The purple stem borer, Seramia inferens Walker, belongs to the order Lepidoptera, Noctuidae, and Sesamia. They are polyphagous insects and their hosts are mainly gramineous crops or weeds such as rice, wheat, maize [10]. With the changes in the farming system, crop distribution, and use of pesticides, S. inferens is gradually becoming a significant rice pest in many parts of China [11]. S. inferens is distributed in rice-producing countries, mainly in Asia. In China, the species is distributed south of N 34 degrees latitude [12]. The ranges are consistent with research by Ezcurra et al. [13]. Based on the cold resistant characteristics and field experiments of S. inferens, the distribution area was also agreed to be south of N 34 degrees latitude by Gu [14]. However, adult S. inferens were first found by light trap and also caused maize plant damage in Xinxiang City, Henan Province, China (E: 113.696°, N: 35.021°) in 2014 [15]. Further investigation illustrated that S. inferens can produce three generations in total and overwinter in local areas with mainly 3–4 larval instars [16]. They must have developed tolerance to cold to establish a population in the new northern breeding area in China in winter. Past research mainly focuses on overwintering in rice stubble in southern areas. It states that S. inferens overwinter as late instar larvae, and it is an exclusively freeze-tolerant insect [17, 18]. Significant damage to maize crops has also been reported in southern China [19]. However, studies on overwintering strategies of the pest in the maize planting area and cold tolerance of S. inferens in the new expansion area remain inadequate.
In this study, the behavioral strategies were confirmed by investigating the overwinter site and survival of S. inferens in early-middle-late winter in the field. The diapause larvae were also collected and the cold hardiness was measured, and a regional distribution based on low temperature threshold and species distribution model MaxEnt were used to generate a potential distribution of S. inferens. This research aims to identify the methods of S. inferens larvae to tolerate cold temperatures and to study the northern distribution boundary in the expansion zone.
2 Materials and methods
2.1 Latitudinal overwintering position of S. inferens
Field surveys were carried out and all S. inferens (dead or alive) larvae were examined in the late overwinter period on the 5th of February 2016 in Nanjing (NJ) (32.65°N, 119.43°E) and the 5th of February 2016 in Changsha (CS) (28.47°N, 113.35°E) in remaining rice stubble. The S. inferens population was also investigated on the 18th of February 2016 in Xinxiang (XX) (35.02°N, 113.69°E) in remaining maize stalks. The number of S. inferens at vertical locations of two hosts was recorded. Two vertical positions of the remaining rice or maize stalk were devised: above ground (AG) and underground (10 cm deep). The data of Xianyou (XY) (25.36°N, 118.69°E) samples cited in Gu [20] were documented in a field survey on the 2nd and 11th of February in 2013.
2.2 Underground-forward behavior on corn stalks
The vertical distribution of S. inferens larvae in maize plants was destructively sampled to determine survival in field conditions during the winters of 2015–2016 on the 6th of October, 2nd of November, 10th of December and 18th of February in XX. We examined the large-scale remaining corn stalks. They usually were the same height, standing at 150 cm or a little over 150 cm, which meant the same positions of the stem stalk represented the same heights above the ground. The numbers of living and dead S. inferens larvae and the position of those insects in the maize stalk were recorded. Six vertical positions of the maize stalk, which remained completely in the field except the corn ear, from top to bottom are defined as positions 1–6. There are 4 positions (positions 1–4) above ground, and two positions underground (positions 5–6). Position 1 represents the high maize ear section (up to 120 cm). Position 2 represents two sections of the maize ear (70–120 cm). Position 3 represents two sections under the maize ear (30–70 cm). Position 4 represents three sections above the soil surface (0–30 cm). Position 5 represents the base of the maize stem and maize root under the soil surface about 10 cm deep. Position 6 represents the surrounding soil near the root, with a diameter of 10 cm.
2.3 Cold hardiness during winter
Wild S. inferens larvae were collected on remaining summer maize stalks in Xinxiang City (32°N, 119°E) in different winter periods on October 31st and November 10th in 2015, January 27th, February 18th and March 30th in 2016. Each larva replicate was placed in a 0.5 ml plastic tube with a 1–2 mm diameter ventilation hole in the lid and labeled. A weight test was carried out (Mettler Toledo ME204, Zurich, Switzerland). SCP and Freezing Points (FP) for each larval replicate were then measured by a small thermocouple thermometer [21] (Temp 20, Beijing Pengcheng Electronic Technology Center), which could automatically record the dropping temperature of an insect body per second. The SCP represents the first lowest temperature value of the body. After the SCP point, the body temperature jumps as the water within the insect cells starts to freeze, generating heat. The peak temperature represents the FP and the temperature begins to decrease after this. The thermometer was placed into the plastic tube, touching the larvae bodies, immobilized with cotton wool. Thermocouples were attached to a multichannel data logger (USB-TC, Measurement Computing, Norton, MA), and recorded and logged by using Tracer-DAQ software (Measurement Computing). The tubes were then placed in a -20°C freezer for 30 minutes. The body temperature of the larvae was gradually reduced, so that all larval replicates could have the SCP and FP recorded, as in Leather [22]. There were about 15 minutes from the test start to the reach point of the SCP and the cooling rate of the body was about 0.2–0.5°C/min. After testing, death was assessed by the lack of mandibular and body movement after 6 hours recovery at 24°C, and the incidence of mortality was identified by the chill-coma recovery numbers. All sample larvae were finally dried at 65°C for 48 h in a drying oven and then weighed again, to determine the water content of each larva.
2.4 Lethal low temperature experiment
Larvae were collected in Anyang (35.57°N, 114.85°E) in the same field on the 5th of November 2015 and each larva was placed in a separate 0.5 ml plastic tube as mentioned above. Twenty centimeters of soil was placed in a container (L605×W410×H240 mm), along with moist cotton wool at the bottom of the container in order to guarantee the same humidity as the overwintering period. The tubes with larvae were then placed vertically into soil in the container. Two centimeters of soil was placed on the tubes to simulate the overwintering environment of S. inferens underground. The container was placed in an outdoor environment. Two microclimate sensors (TH11R, HHW) were installed in air (10 cm from the soil surface) with a shelf, and at 2 cm soil depth to record the temperature of both sites until March 30th 2016. Low-temperature exposure experiments were carried out to measure the thresholds for the long-term survival of S. inferens at a constant temperature, separated into 4 times (Nov 9th, Dec11th, Jan 28th, Feb 19th) and seven gradient low temperatures (0°C, -5°C, -10°C, -15°C, -20°C, -25°C, and -30°C). All replicates of 140 in total were placed in a 0°C freezer for a 12 hour cold treatment, and then 20 replicates were taken out and placed at ambient temperature for 24 hours to determine whether they were alive or dead. The number of dead larvae was recorded. Those remaining replicates of 120 then continued to experience -5°C for 12 hours by reducing the freezer temperature, and then 20 replicates were taken out to check if alive, and so forth until all 7×20 larval replicates were tested. Ultimately, the incidence of mortality of different temperatures was identified. A low temperature threshold: Air Lower Lethal Temperature (ALLT50), which is defined as the temperature where 50% of individuals die in the low-temperature exposure experiments mentioned above, was calculated by a logistic linear regression [23]. In this study, the data of the sample dates were taken together to calculate the ALLT50, and the ALLT50 value was -9.7°C.
2.5 Distribution based on low temperature threshold
Daily minimum temperatures in 47 different locations (Henan province, Hebei province, Tianjing city and Beijing city, formed a south-north belt in the north plain of China, in order to illustrate the northern overwinter site of S. inferens) from 2007 to 2014 were collected from national weather stations (https://data.cma.cn/). The winter temperatures from November to March were recorded in the air (10 cm from the soil surface) and underground (at 2 cm soil depth) by the two microclimate sensors mentioned above and it was found that there is an obvious significant linear relationship between lowest daily air and lowest underground temperature (R2 = 0.9161, p<0.001, n = 100). Based on this linear model, the minimum daily air temperatures in different locations were translated into the minimum daily soil temperature at 2 cm underground because the underground data were not available at national weather stations, and the landscape and soil composites of the 47 locations were very similar. Another low temperature threshold, Underground Lower Lethal Temperature (ULLT50) was also translated by linear equation and the value was -4.5°C, i.e. when the air temperature was -9.7°C the soil temperature postulated reached -4.5°C. Forty-seven location survival rates of air or underground in each year were speculated by predicting all larvae would die when the annual daily minimum temperature was lower than the ALLT50 or ULLT50, respectively, i.e. if the annual minimum temperature is lower than the ALLT50 or ULLT50, 0% of the population will survive. One hundred percent will survive if the annual minimum temperature is higher than the ALLT50 or ULLT50. The calculation of survival rates in different locations from 2007 to 2014, resulted in identification of the total proportion of insect survival over the 8 years. Based on the proportion of each location, a potential distribution based on annual minimum temperature threshold could be produced with ArcMap (version 10.2, ESRI, Redlands, CA, USA) [24].
2.6 The distribution based on MaxEnt
A species distribution model MaxEnt (version 3.3.3k) was used to generate a potential district-level distribution map of S. inferens. Nineteen bioclimatic layer data from the WorldClimte dataset were obtained [25] (http://www.worldclim.org/). Occurrences at 147 districts (locations) were obtained from published articles or websites, including the new northern breeding region in China [26], (S1 and S2 Tables). Sixty percent of total district (city) occurrences were used for model calibration (training data: 98 districts) and the remaining for model validation (test data: 49 districts). GoogleEarth was used to generate approximate coordinates. The area under the ROC (receiver operating characteristic) curve metric was used to evaluate the model performance [27, 28] (S1 Fig). Current and potential distribution maps were generated using ArcMap and three arbitrary categories were defined as low (<0.15), medium (0.15–0.53) and high (>0.53) based on predicted habitat suitability.
2.7 Statistics
Data were tested for normal distribution using the Shapiro-Wilk test and normally distributed data was compared by an independent samples t-test and Pearson’s correction test. The differences in the incidences of mortality exposed in different sub-zero temperatures between different sampling periods in winter were compared with two factors across anaylsis of variance. The differences between factors were evaluated using Tukey-Kramer groupings comparison in the least squares means and Non-normally distributed data were compared by nonparametric tests of group differences. The linear or nonlinear models were identified with p<0.05 for the significance, and all data were analyzed with the software program R 3.5.3 (R Core Team 2017).
3 Results
3.1 Geographical variation in overwinter position
The living S. inferens population was closer to the ground with increasing latitude in late winter and 100% and 75% of overwinter individuals were underground in northern locations in XX and NJ, respectively. But in southern regions, CS and XY, they were mainly above ground (Fig 1).
UG: Underground; AG: above ground, XX: Xinxiang(n = 24), NJ: Nanjing(n = 74), CS: Changsha(n = 33); XY: Xianyou(n = 13).
3.2 Overwintering behavioral in maize stalk
In early winter (October 6th and November 2nd), 100% of the S. inferens population was alive and the highest numbers of insects were at positions 2 and 3. In later winter (December 10th and February 18th), the incidences of population mortality were 25.5% and 20.83% respectively, and the highest numbers of insects were both at position 5 (Fig 2). In the early winter, most of the S. inferens population was distributed above ground. The location of the whole population shifted to the ground when the winter became colder, and most (83.3%) of the S. inferens population (dead and alive) was distributed underground by the 18th of February (Fig 3). On the 10th of December, the ratio of dead individuals above ground and underground was 21:1 and on the 18th of February, it was 4:1. All of the living individuals were only found at underground position 5 and all the remaining population above ground in winter were dead by February 18th (Fig 2).
3.3 Cold hardiness of S. inferens
There were significant differences in SCP and FP of S. inferens larvae over time (SCP: F4,91 = 4.993, p = 0.003; FP: F4,91 = 10.26, p<0.001). The mean SCP (±SE) of S. inferens larvae ranged from the highest (-5.731±0.405°C) in October to the lowest (-7.653±0.952°C) in January (Fig 4) and the mean and minimum value of the SCP first decreased and then rose with the cold temperature event. The lowest value -20.02°C appeared on the 27th of Jan (Table 1). Similarly, the mean of the FP first decreased and then rose. The means of the FPs in middle winter (December: -5.49±0.447; January: -6.596±0.889) were significantly lower than early and end of winter (November, October and March). In contrast, there was no significant difference in the fresh weight of larvae during the cold temperature period (F4,91 = 1.997, p = 0.102), but there was a significant gradual decrease in dry weight (F4,91 = 3.934, p = 0.006) and water content of the larvae (F4,87 = 3.19, p = 0.017). Larvae dry weight showed a decreased tendency throughout the winter and the water content decreased at the beginning of winter and slightly increased at the end of winter (Table 1).
Same pattern bar with different letters on top shows a significant difference at p< 0.05.
After a 30 minute SCP test, 91.667% of the tested S. inferens larvae were dead when returned to 24°C immediately after freezing on October 31st. About sixteen percent died in the January 27th SCP test, and 76.471% died in the March 31st test. The number of dead larvae after the test was significantly negatively correlated with the daily air mean temperature on the sampling date (r = -0.915, p = 0.029) (Table 1). The species was therefore determined to be a moderately freeze-tolerant insect.
3.4 Current and potential distribution of S. inferens
There were no significant differences between different sampling periods of S. inferens in winter (F3,24 = 0.002, p = 0.998), but there were significant differences between the length of time the insects were exposed to subzero temperatures (F1,24 = 19.650, p = 0.002). Therefore, we combined the dataset and only considered the effects of exposure temperatures. The ALLT50 was -4.5 based on a nonlinear regression: y = (1.03054)x (y: Incidence of mortality, x: Temperature (°C), R2 = 0.878, p<0.001, n = 24) (Fig 5A). The ULLT50 was -9.713°C after being translated by a linear model between minimum daily air and underground temperature: y = 0.6628*x+2.6267 (y: Underground, x: Air, R2 = 0.9161, p<0.0001, n = 100) (Fig 5B). The regional survival ratio was calculated with both ALLT50 and ULLT50 based on low temperature threshold. ≥50% probability survival regions were below N 35° due to the ALLT50 and below N 40° due to the ULLT50 (Fig 6A).
(A) The incidence of mortality of S. inferens larvae after lethal low temperature experiments and (B) the relationship between daily minimum air and underground temperature during the overwinter period.
(A) Potential distribution based on the low temperature threshold, the size of the circle in A means the survival probability, and the blue and purple circles are sourced from ALLT50 and ULLT50, respectively. (B) Current overwinter sites and predicted suitable habitat area modeling by MaxEnt. The darker color in B represents the better predicted habitat suitability. The white points are the new northern overwinter sites. The map was produced using ArcGIS 10.2 under a CC BY license, with permission from ESRI (www.esri.com) and shapefile reprinted from GADM database under a CC BY license, with permission from Global Administrative Areas (www.gadm.org).
The MaxEnt model predicted potential distribution of S. inferens with a high accuracy test AUC value of 0.97 and training AUC value of 0.981 (S2 Fig), and three temperature-variables: BIO6 (minimum temperature of coldest month), BIO10 (mean temperature of warmest quarter), and BIO2 (mean of monthly difference of maximum and minimum temperature), which showed a top three permutation importance on the model (S3 Table). Model predictions closely matched the new overwinter site and also showed potentially suitable districts in north regions, and predicted suitable habitat areas of S. inferens below N 40°. The most suitable areas were below N 38° (Fig 6B). This means that the underground overwinter temperature boundary defined by the ULLT50 is reliable, and overwinter strategies preventing winterkilling led to five latitude wider overwintering regions.
4 Discussion
There are many studies which show the latitudinal pattern of insects facing cold temperatures in winter. Higher latitude populations possess a stronger cold tolerance, a shorter chill-coma recovery time and a higher survival rate [29–34]. This research shows the overwintering location of the S. inferens population was closer to the ground with increasing latitude. In the new northern breeding area in winter, with the temperature decreasing, S. inferens population gradually climbed down and drilled into the base of the maize stalk under the soil surface before the coldest temperature arrived. Larvae remaining above ground in winter were all winterkilled in later winter. Compared to the year-round site at N 19° in the south [35], it is an obvious behavioral strategy of S. inferens to survive, to find the warmer microclimate of the overwintering site in the north in winter. The different latitudinal position and underground-forward behavior observed may be entirely lower temperature driven. A further indoor experimental test is needed to confirm that populations from the south exhibit the same behavior if exposed to colder temperatures, so that a population behavior could be defined. Insects usually select warmer micro-habitats to survive. Coccinella septempunctata L., Ceratomegilla undecimnotata (Schneider), Hippodamia variegata (Goeze) Harmonia axyridis (Pallas) and Coccinella septempunctata select human houses, isolated grass tussocks or under covered stones as overwintering sites [36, 6, 37]. Similarly, Sesamia nonagrioide, the same family of the insects in the study but another species, also show underground-forward behavior in winter in Landes (South-west of France, 45°N, 0.7°W) [38].
Physiological adaptation of insects for low temperatures is by partial dehydration, increasing body fluid osmolality, and the accumulation of a complex mixture of winter specific metabolites, which strengthen their cold hardiness [39, 37]. In this paper, the mean SCP of larvae which were collected from the field was -7.653±0.952°C (lowest), and the incidence of mortality of the population was 15.789% (lowest) after a 30 minute cold hardiness test in January (middle winter). Meanwhile, the water contents gradually decreased and were lowest on Feb 18th which shows the diapaused insects transfer fluidity into antifreeze proteins and glycoproteins [40]. When insects begin to recover in spring the fluidity increases. In this research, insects survived when body temperature was under freezing point and showed an extreme low temperature tolerance capability meaning that the population of S. inferens larvae in the northern breeding area was a moderately freeze-tolerant one. These results correspond with Guo et al. [17] and Sun [18] where the pest hosts on rice. Cold hardiness is also related to the host insects have fed on [41, 42] and the differences in the cold tolerance mechanisms between maize and rice need to be further investigated in the new breeding region.
Minimum temperatures in the coldest month are linearly related to insect mortality [29]. ALLT50 or the lethal temperature (LTe50) of insect cold exposure are the best predictors of cold distribution limits and are usually used to assess the potential distribution according to the annual minimum temperature [43, 23]. In this research, the ALLT50 and ULLT50 were -9.7°Cand -4.5°C, respectively. The method particularly included the effect of different exposure durations in different winter periods. In natural conditions, there were two days when the daily mean temperatures in the soil were subzero in 2015–2016 in the middle-south district (Xinxiang), but in north, there were more subzero days. Therefore, the larvae in the field will experience different durations of subzero temperatures in the soil in different latitudinal places. The ALLT50 we identified was the subzero temperature effect after experiencing different exposure durations; the value was -9.7°C. This is more realistic than the lower ALLT50 (-6.12°C) of larvae which was sourced through consistent exposure duration (2 h) collected on rice in Oct 2012 in Yangzhou by Sun [18], particularly when assessing the distribution of insects on a large scale in winter. Other research also shows that short-term lower temperature interruption procedures causes higher death and results in a higher ALLT50 [44]. This study used the lower temperature criteria which discriminated by up or below the ALLT50/ULLT50 in annual daily minimum temperatures, and show a northern distribution considering the ALLT50 in locations under 35°N. However, depending on the ULLT50, the results showed a northerly distribution below 40°N when insects overwinter underground. This was also in agreement with the results of the MaxEnt modeling, which show below 40°N predicted suitable habitat areas. Distribution of insects is highly impacted by climatic factors (temperature, moisture, humidity and their variations), especially the effects of temperature [45]. MaxEnt integrates insect occurrence records with climatic and other environmental variables and the MaxEnt model of this study shows a similar result to de la Vega [46], that the minimum temperature of the coldest month was the important abiotic factor restricting the geographic distribution of Triatoma infestans and Rhodnius prolixus. The MaxEnt model also included many other important parameters such as the precipitation, which showed a high contribution in the model. It is usually used to predict potential distributions of insect pests [27]. Although artificial because the linear model based on the point data might not represent the actual regional patterns, the linear relationship between air and soil temperature has indeed been studied in different areas [47, 48]. Hong [24] has taken advantage of this linear relationship from 96 weather stations in Shanxi, China, and classified the overwintering sites of the southern root-knot nematode. These results also show that the underground low temperature threshold defined by the ULLT50 could be extrapolated to other underground overwinter species.
The population behavior of animals which escape cold weather usually by large- scale migration into a warmer habitat, will lead to relocated populations regionally, but not all individuals can successfully find suitable places. S. inferens have a weak flight tendency and capability in the field [49]. This paper shows another behavioral strategy of S. inferens larvae by locally searching for a warmer place by drilling into the ground in early winter. The microclimate of the overwintering site could result in the S. inferens colonizing more northerly latitudes and the insect’s physiological and biochemical changes may be a factor favoring this northern expansion leading to a higher winter survival chance. Past studies have shown that S. inferens are distributed below N 34° degrees north latitude. This study also documents a new location for S. inferens: a northern distribution record of N 40°. The overwintering boundary of many insects has also been reportedly spreading poleward due to global warming [50]. It cannot be overlooked that the warmer changes in northern China in these two centuries have driven the change of animal redistribution [51], but the strategy of behavioral or physiological changes in this study to overcome winter will certainly help to colonize a five degree more northerly latitude. For maize producers, ploughing the maize stubble in autumn after harvesting will lead to an increased mortality of S. inferens larvae due to cold exposure. This study allows us to propose a simple non-pollutant pest control method in northern China.
Supporting information
S1 Fig. The Receiver Operating Characteristic (ROC) curve of the MaxEnt model.
https://doi.org/10.1371/journal.pone.0236174.s001
(DOCX)
S2 Fig. AUC of different environmental variables based on results of jackknife tests in the MaxEnt model.
https://doi.org/10.1371/journal.pone.0236174.s002
(DOCX)
S1 Table. The live number of S. inferens larvae at northern overwinter sites.
https://doi.org/10.1371/journal.pone.0236174.s003
(DOCX)
S2 Table. The worldwide distribution of S. inferens.
https://doi.org/10.1371/journal.pone.0236174.s004
(DOCX)
S3 Table. The permutation importance and relative contributions of the environmental variables in the MaxEnt model.
https://doi.org/10.1371/journal.pone.0236174.s005
(DOCX)
Acknowledgments
We thank Chao Han and Mingyong Ma for carrying out the overwinter investigation in Nanjing and Changsha, respectively. We thank those who contributed to discussions of the ideas presented in this paper. Instructive comments from the anonymous reviews have greatly improved this manuscript.
References
- 1. Sunday JM, Bates AE, Dulvy NK. Thermal tolerance and the global redistribution of animals. Nat Clim Chang. 2012; 2(9): 686–690
- 2. Bale J S. Insect cold hardiness: a matter of life and death. Eur J Entomol 1996; 93(3): 369–82
- 3. Stahl K, Moore RD, McKendry IG. Climatology of winter cold spells in relation to mountain pine beetle mortality in British Columbia. Canada Clim Res. 2006; 32(1):13–23
- 4. Gray DR. The relationship between climate and outbreak characteristics of the spruce budworm in eastern Canada. Clim Change. 2008; 87(3–4): 361–383
- 5. Calosi P, Bilton DT, Spicer JI, Votier SC, Atfield A. What determines a species geographical range? Thermal biology and latitudinal range size relationships in European diving beetles (Coleoptera: Dytiscidae). J Anim Ecol. 2010; 79(1): 194–204 pmid:19761459
- 6. Labrie G, Coderre D, Lucas E. Overwintering strategy of multicolored Asian lady beetle (Coleoptera: Coccinellidae): cold-free space as a factor of invasive success. Ann Entomol Soc Am. 2008; 101(5): 860–866
- 7. Ward SF, Venette RC, Aukema BH. Cold tolerance of the invasive larch casebearer and implications for invasion success. Agricultural and Forest Entomology, 2019; 21(1): 88–98.
- 8. Reynolds DS, Reynolds WC, Brandenburg RL. Overwintering, oviposition, and larval survival of hunting billbugs (Coleoptera: Curculionidae) and implications for adult damage in north carolina turfgrass. J Econ Entomol. 2015; 109(1): 240–248.
- 9. Hu CX, Hou ML, Wei GS, Shi BK, Huang JL. Potential overwintering boundary and voltinism changes in the brown planthopper, Nilaparvata lugens, in China in response to global warming. Clim Change. 2015; 132(2): 337–352
- 10.
Guo YY. Crop pests and diseases in China (The 3nd press). Beijing: China Agricultural Press. 2017.
- 11. Chen XJ, Lu DH. Study Advances in Occurrence and Control of the Polyphagous Pink Stem Borer Sesamia inferens. China Agric Sci Bull. 2015; 31(25): 171–175. Chinese
- 12. Zhang SM. Discussion the boundary line between the ancient North and the Eastern regions in the east of the Qinling Mountains in China from the distribution of some agricultural insects. Acta Entomol Sin.1965; 14(4): 411–419. Chinese.
- 13. Ezcurra E, Rapoport EH, Marino CR. The geographical distribution of insect pests. J Biogeogr.1978; 5(2): 149–157
- 14. Gu NH. Preliminary study on the characteristics of hibernation of pink rice borer. Acta Ecol Sin. 1985; 5(1): 64–70. Chinese
- 15. Huang JR, Feng HQ. Preliminary report on damage of summer corn seedlings by the purple stem borer, Sesamia inferens (Walker) (Lepidoptera: Noctuidae) in northern Henan province, Plant Protect,2015, 41(2):231–233. Chinese.
- 16. Huang JR, Huang B, Li GP, Tian CH, Feng HQ. Preliminary study on the occurrence of pink stem borer, Sesamia inferens (Walker) (Lepidoptera: Noctuidae) in northern Henan Province. Plant Protect. 2017; 43(2):156–162. Chinese.
- 17. Guo H F, Li Q, Feng J C, Zhang H. Comparison of cold tolerance of three specie of overwintering rice stem borers in Nanjing area. J Jiangsu agric Sci. 2002; 18(2):85–88. Chinese
- 18. Sun M, Tang XT, Lu MX, Yan WF, Du YZ. Cold tolerance characteristics and overwintering strategy of Sesamia inferens (Lepidoptera: Noctuidae). Fla Entomol. 2014; 97(4): 1544–1553
- 19. Xie DP, Cai RX, Zhao FZ, Li TC, Xie TY. A study of the bionomics of the pink stem borer of maize, Sesamia inferens walker and its control by insecticides. J Sichuan Agric C. 1985; 3(1): 33–40
- 20.
Gu L. Characteristics of rice pest in Xianyou county [dissertation], Fujian: Fujian Agriculture and Forestry University. 2014
- 21. Liu YQ. Zheng XX, Ma HF, Xia R X, Li YP, Zhang QR. Supercooling capacity and cold tolerance of the wild silkworm, Antheraea pernyi (Lepidoptera: Saturniidae). J Econ Entomol. 2016; 109(4): 1619–1627 pmid:27371710
- 22.
Leather SR, Walters KFA, Bale JS. The ecology of insect overwintering. Cambridge: Cambridge University Press; 1993.
- 23. Andersen JL, Manenti T, Sørensen JG, MacMillan HA, Loeschcke V, Overgaard J. How to assess Drosophila cold tolerance: chill coma temperature and lower lethal temperature are the best predictors of cold distribution limits. Funct Ecol. 2015; 29(1): 55–65
- 24. Hong B, Zhang F, Li YM, Zhang SL, Chen ZJ, Gao F, et al. GIS-based regional classification for overwintering of southern root-knot nematode in Shanxi Province. Acta Ecol Sin. 2014; 34(16): 4603–4611
- 25. Fick SE, and Hijmans RJ. WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. Int J Climatol. 2017; 37(12): 4302–4315
- 26. Zhang YC, Feng HQ, Huang JR. A new northern distribution record of the purple stem borer, Sesamia inferens (Walker) (Lepidoptera: Noctuidae) in China. Entomol News. 2019; 128(5): 460–462.
- 27. Phillips SJ, Anderson RP, Schapire RE. Maximum entropy modeling of species geographic distributions. Ecol Model. 2006; 190 (3–4): 231–259
- 28. Kumar S, Graham J, West AM, Evangelista PH. Using district-level occurrences in MaxEnt for predicting the invasion potential of an exotic insect pest in India. Comput Electron Agric. 2014; 103(2): 55–62
- 29. Hoffmann AA, Anderson A, Hallas R. Opposing clines for high and low temperature resistance in Drosophila melanogaster. Ecol Lett. 2002; 5(5): 614–618
- 30. David JR, Gibert P, Moreteau B, Gilchrist GW, Huey RB. The fly that came in from the cold: geographic variation of recovery time from low-temperature exposure in Drosophila subobscura. Funct Ecol. 2003;17(4): 425–430
- 31. Chen B, Kang L. Variation in cold hardiness of Liriomyza huidobrensis (Diptera: Agromyzidae) along latitudinal gradients. Environ Entomol. 2004; 33(2): 155–164
- 32. Rochefort S, Berthiaume R, Hébert C, Charest M, Bauce é. Effect of temperature and host tree on cold hardiness of hemlock looper eggs along a latitudinal gradient. J Insect Physiol. 2011; 57(6): 751–759 pmid:21356214
- 33. Kleynhans E, Mitchell KA, Conlong DE, Terblanche JS. Evolved variation in cold tolerance among populations of Eldana saccharina (Lepidoptera: Pyralidae) in South Africa. J Evol Biol. 2014; 27(6): 1149–1159 pmid:24773121
- 34. Tian B, Xu L, Zhang M, Feng Y, Zong S. Supercooling capacity and cryoprotectants of overwintering larvae from different populations of Holcocerus hippophaecolus. CryoLetters, 2016; 37(3): 206–217. pmid:27393957
- 35. Zhou CB, Chen AF. Observation on the biology and wintering situation of Sesamia inferens in northern Hainan province. China J Appli Entomol. 1985; 5(1):199–201. Chinese
- 36. Honek A, Martinková Z, Pekár S. Aggregation characteristics of three species of Coccinellidae (Coleoptera) at hibernation sites. Eur J Entomol. 2007; 104(1): 51
- 37. Koštál V, Miklas B, Doležal P, Rozsypal J, Zahradníčková H. Physiology of cold tolerance in the bark beetle, Pityogenes chalcographus and its overwintering in spruce stands. J Insect Physiol. 2014; 63(1): 62–70
- 38. Gillyboeuf N, Anglade P, Lavenseau L, Peypelut L. Cold hardiness and overwintering strategy of the pink maize stalk borer, Sesamia nonagrioides Lef (Lepidoptera, Noctuidae). Oecologia. 1994; 99(3–4), 366–373. pmid:28313892
- 39. Rozsypal J, Koštál V, Zahradníčková H, Šimek P. Overwintering strategy and mechanisms of cold tolerance in the codling moth (Cydia pomonella). PLoS One. 2013; 8(4): e61745 pmid:23613923
- 40. Rozsypal J, Koštál V, Berková P, Zahradníčková H, Šimek P. Seasonal changes in the composition of storage and membrane lipids in overwintering larvae of the codling moth, Cydia pomonella. J Therm Biol. 2014; 45: 124–133 pmid:25436961
- 41. Morey AC, Venette RC, Nystrom Santacruz EC, Mosca LA. Hutchison WD. Host‐mediated shift in the cold tolerance of an invasive insect. Ecol Evol. 2016; 6(22): 8267–8275 pmid:27878094
- 42. Alford L, Androdias A, Franco T, Pierre JS, Burel F, van Baaren J. Potential host manipulation by the aphid parasitoid Aphidius avenae to enhance cold tolerance. PloS one. 2016; 11(12): e0168693 pmid:28006018
- 43. Ungerer MJ, Ayres MP, Lombardero MJ. Climate and the northern distribution limits of Dendroctonus frontalis Zimmermann (Coleoptera: Scolytidae). J Biogeogr. 1999; 26(6):1133–1145
- 44. Nedvĕd o, Lavy D, Verhoef H A. Modelling the time-temperature relationship in cold injury and effect of high temperature interruptions on survival in a chill-sensitive collembolan. Func Ecol. 1998; 12(5): 816–824
- 45. Aregbesola OZ, Legg JP, Sigsgaard L, Lund OS, Rapisarda C. Potential impact of climate change on whiteflies and implications for the spread of vectored viruses. J Pest Sci, 2019; 92(2): 381–392.
- 46. de la Vega GJ, Medone P, Ceccarelli S, Rabinovich J, Schilman PE. Geographical distribution, climatic variability and thermo-tolerance of Chagas disease vectors. Ecography. 2015; 38(8):851–860
- 47. Parton WJ, Logan JA. A model for diurnal variation in soil and air temperature. Agric Meteorol. 1981; 23(3): 205–216
- 48. Horton B. Models for estimation of hourly soil temperature at 5 cm depth and for degree-day accumulation from minimum and maximum soil temperature. Soil Res. 2012; 50(6): 447–454
- 49. Han LZ, Peng YF, Wu KM. Studies on larval dispersal ability in the field and adult flight capacity of the pink stem borer, Sesamia inferens. Plant Prot. 2012; 38(4): 9–13. Chinese
- 50. Wheatley CJ, Beale CM, Bradbury RB, Pearce-Higgins JW, Critchlow R, Thomas CD. Climate change vulnerability for species-Assessing the assessments. Glob Chang Biol. 2017;23(9): 3704–3715 pmid:28660715
- 51. Wu J. Detection and attribution of the effects of climate change on bat distributions over the last 50 years. Clim Chang. 2016; 134(4): 681–696.