Temperature plays an important role in the growth and development of arthropods, and thus the current trend of climate change will alter their biology and species distribution. We used Chaetodactylus krombeini (Acari: Chaetodactylidae), a cleptoparasitic mite associated with Osmia bees (Hymenoptera: Megachilidae), as a model organism to investigate how temperature affects the development and voltinism of C. krombeini in the eastern United States. The effects of temperature on the stage-specific development of C. krombeini were determined at seven constant temperatures (16.1, 20.2, 24.1, 27.5, 30.0, 32.4 and 37.8°C). Parameters for stage-specific development, such as threshold temperatures and thermal constant, were determined by using empirical models. Results of this study showed that C. krombeini eggs developed successfully to adult at all temperatures tested except 37.8°C. The nonlinear and linear empirical models were applied to describe quantitatively the relationship between temperature and development of each C. krombeini stage. The nonlinear Lactin model estimated optimal temperatures as 31.4, 32.9, 32.6 and 32.5°C for egg, larva, nymph, and egg to adult, respectively. In the linear model, the lower threshold temperatures were estimated to be 9.9, 14.7, 13.0 and 12.4°C for egg, larva, nymph, and egg to adult, respectively. The thermal constant for each stage completion were 61.5, 28.1, 64.8 and 171.1 degree days for egg, larva, nymph, and egg to adult, respectively. Under the future climate scenarios, the number of generations (i.e., voltinism) would increase more likely by 1.5 to 2.0 times by the year of 2100 according to simulation. The findings herein firstly provided comprehensive data on thermal development of C. krombeini and implications for the management of C. krombeini populations under global warming were discussed.
*Scientific Article No. 3278 of the West Virginia Agricultural and Forestry Experiment Station, Morgantown, West Virginia
Citation: Ahn JJ, Son Y, He Y, Lee E, Park Y-L (2016) Effects of Temperature on Development and Voltinism of Chaetodactylus krombeini (Acari: Chaetodactylidae): Implications for Climate Change Impacts. PLoS ONE 11(8): e0161319. https://doi.org/10.1371/journal.pone.0161319
Editor: Raul Narciso Carvalho Guedes, Universidade Federal de Vicosa, BRAZIL
Received: March 3, 2016; Accepted: August 3, 2016; Published: August 17, 2016
Copyright: © 2016 Ahn 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 paper and its Supporting Information files.
Funding: The study was supported by the USDA National Institute of Food & Agriculture (NIFA) Grant NIFA Award: 2014-67004-21631 (http://nifa.usda.gov/program/agriculture-and-food-research-initiative-afri). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Climate change affected population dynamics of arthropods in natural and agricultural systems such as temporal asynchrony of interacting populations [1,2]. The Coupled Model Intercomparison Project Phase 5 (CMIP5) predicted that significant future climate change and variability would cause severe impacts on various ecosystems. Among various events caused by climate change, temperature increase could be one of the most influential drivers of ecosystem function . The global average surface temperature has increased over the 20th century by ca. 0.6°C  and the rate of increasing surface temperature from 1895 through 2013 was about 0.72°C in the contiguous 48 states of the United States .
Temperature is a key abiotic environmental factor that influences growth, development, survival and reproduction of poikilothermic arthropods including insects and mites [5–10]. Although some social insects (e.g. honey bees) can decrease the temperature of their nests by fanning their wings, most of arthropods do not have the ability to manipulate temperature within their nests, making them more vulnerable to temperature increase. Arthropod development occurs within a limited range of temperatures and the thermal range varies among arthropod species. Various empirical models have been applied to describe the relationship between temperature and developmental parameters including the lower and upper developmental thresholds and optimal temperatures [11–14]. In addition, developmental parameters often have been applied for optimizing mass rearing systems, predicting phenology, modeling population growth, determining geographical distribution, and simulating climate change impact [15–18]. Various studies already showed that development of arthropods is sensitive to temperature changes [5,19,20] and a small change in temperature could result in spatial and temporal changes in their phenology.
Based on the future climate scenarios of Representative Concentration Pathways (RCPs) , annual mean surface temperatures in the eastern United States are predicted with the greatest increase from 2006 to 2100 under RCP 8.5 scenario (+5.7°C), followed by RCP 6.0 (+4.1°C), RCP 4.5 (+3.2°C), and RCP 2.6 (+1.2°C) scenarios (S1 Fig). These four RCP scenarios are new ones presented in the Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment Report (AR5) and provide time-dependent projections of atmospheric greenhouse gas concentrations . The simulated increases in surface air temperature over the eastern United States during the 21st century would be significant enough to impact growth and development of arthropods based on temperature increase in the United States during the last century (< +1°C).
The Krombein’s hairy-fingered mite, Chaetodactylus krombeini (Acari: Chaetodactylidae), is a cleptoparasitic mite of Osmia bees (Hymenoptera: Megachilidae) which are major or supplementary bees for orchard pollination. Osmia bees collect pollen in the field and provision in the nest to feed their larvae. The mite steals and consumes the pollen (i.e. cleptoparasitic), which causes the decline of Osmia bee populations . At high density of over 50 adults per cell in an Osmia nest , C. krombeini can kill the bee eggs  and larvae  by attacking and consuming directly. Even if the bee eggs and larvae survive under heavy infestation of the mites within the same cell, the adult bees suffer from reduced fecundity  and their pollination efficiency is reduced because the bees get irritated by the mites on their body. Therefore, among the pests of Osmia bees, C. krombeini is known as a key pest that affects the maintenance and propagation of Osmia bees for orchard pollination in the United States [22,26,27]. If the mite infestation is not controlled in an orchard, Osmia populations could decrease to half from previous year’s size .
Developmental stages of C. krombeini life cycle include egg, larval, protonymph, deutonymph, tritonymph, and adult stages. Chaetodactylus krombeini overwinters in Osmia bee nests as a deutonymph in two different morphs: encysted and phoretic morphs . Major cures to trigger overwintering of C. krombeini have not been revealed yet although daylight length, temperature, and food availability could be potential factors. The encysted deutonymphs remain in the same nest over winter until some environmental queues trigger their activity . In spring when temperature increases, the phoretic deutonymphs become active and disperse from old nests to new nests primarily by hitch-hiking on Osmia adults . Upon arrival into a new nest of the bee, the phoretic deutonymphs become adult females after molting and each female lays a single egg which develops into a male . Then, a typical life cycle (i.e. egg, larva, protonymph, tritonymph and adult) is repeated and the mites build up their population throughout summer. Later in the season when temperature decreases or pollen provisions are scarce in the nest, protonymphs become either phoretic or encysted deutonymphs for overwintering [24,29].
Despite the economic importance of C. krombeini as a key pest of Osmia bees, little is known about the effect of temperature on the developmental biology of C. krombeini. This mite does not have any temperature regulation system in its body or behavior adaptation, which makes it an ideal organism to measure the impact of temperature change. This study was conducted (1) to investigate stage-specific development of C. krombeini at constant temperatures, (2) to develop empirical models to describe thermal development of C. krombeini, and (3) to predict the spatial and temporal changes in voltinism (i.e., the number of generations) of C. krombeini under the future climate scenarios.
Materials and Methods
Chaetodactylus krombeini was collected from the nests of the Osmia cornifrons (Hymenoptera: Megachilidae) colony on the West Virginia University Organic Farm in Morgantown, West Virginia (USA) with permission by the WVU Experiment Station. To establish laboratory colonies of C. krombeini, adult mites were isolated from bee nests and transferred into wells of ELISA plates (E&K Scientific, Santa Clara, CA) mimicking Osmia bee nests. Pollen provisions collected from the bee nests in the field were supplied into the wells as a food source for mites. The plates with mite colonies were kept in a 1,000-cm3 cardboard box to provide darkness and kept under 20 ± 2°C and 60 ± 10% RH.
Newly-laid C. krombeini eggs (<1 day old) were collected randomly from the laboratory colonies to create seven sets of mite colonies for experiments. In each set, 50–100 eggs were individually transferred into an ELISA well provisioned with pollen collected from O. cornifrons nests. The seven sets of mite colonies were placed into environmental chambers (Percival Scientific, Perry, IA; Fisher Scientific, Dubuque, IA) with seven constant temperatures (i.e. 16.1±0.28°C, 20.2±0.73°C, 24.1±0.06°C, 27.5±0.33°C, 30.0±0.08°C, 32.4±0.51°C, and 37.8±0.48°C). Temperature inside each environmental chamber was recorded with a HOBO data logger (HOBO® onset® UX 100 Temp/RH, Cape Cod, MA). Developmental stage (i.e., egg, larva and nymph) of each C. krombeini was checked and recorded daily. A stereomicroscope (Olympus SZ-ST, Tokyo, Japan) was used to distinguish life stages of C. krombeini based on size, shape, and quiescence periods between two stages (i.e., larva to nymph and nymph to adult). The adult stage was identified based on the presence of seta on abdomen and sclerotized structure underneath the abdomen . Checking and recording the developmental stages of C. krombeini needed to take C. krombeini development out of the chambers. To minimize the potential effect of temperature changes on C. krombeini development, recording developmental stages was done within 10 min. under 20±0.5°C, 60±10.2% RH.
Developmental distribution model
The effect of temperature on the duration of development was examined by using analysis of variance (ANOVA) . Regression analyses were used to model temperature-dependent development of each C. krombeini stage and the model parameters were obtained by using TableCurve 2D Automated Curve Fitting program .
Cumulative frequency distribution (%) of stage-specific emergence was plotted against development time (d) at each temperature. The relationship between time and the frequency was described by the cumulative Weibull model [33,34]: where f(x) is the cumulative frequency distribution (%) at cohort age x and α and β are scale and shape parameters, respectively. The median development time (i.e., time to 50% cumulative frequency) was calculated as α × [–ln (0.5)]1/β. To generate a temperature-independent distribution model for each stage, the cumulative frequency (%) was plotted against the normalized time which was calculated by dividing the development time by the median development time at each temperature (i.e., days / median). The normalized data were pooled across temperatures and fit to the Weibull model .
Development rate model
Development rate at each temperature was calculated as the reciprocal of median development time (1 / median) in days, which was obtained by the Weibull distribution model at each temperature. Over the entire thermal range, the relationship between temperature and the development rate was described by the nonlinear Lactin model , which was modified from Logan type I model : where R(T) is the rate of development at temperature T; ρ can be interpreted as a composite Q10 value for enzyme-catalyzed biochemical reactions; TL is the lethal maximum temperature; Δ is the width of the decline phase in development rate above the optimum temperature . Optimum temperature (Topt), at which the mite develops at its maximal rate, was calculated using the estimates of model parameters as Topt = TL− Δ. The thermal range with more than 80% performance (B80) was also determined . The 3-parameter Lactin model was selected because it consistently provided significant fits to the temperature-dependent rate data for all C. krombeini stages, in contrast to alternative 4-parameter nonlinear models.
Data at mid-range of the Lactin model were used to develop a linear model: y = a + bx, where y is the rate of development at temperature x; a is the y-intercept; and b is the slope. Using the parameter estimates, the lower developmental threshold (LDT) was calculated as—(a / b) and the thermal constant in degree-days (DD) was calculated as 1 / b .
Simulation of C. krombeini adult emergence
Assuming that a cohort of C. krombeini eggs were continuously exposed to constant-temperature condition, adult emergence in terms of daily frequency (%) was simulated in relation to the temperature (°C) and the time (day) by incorporating two models of C. krombeini development: the Weibull model for development frequency distribution and the Lactin model for the temperature dependent rate [5,35]. The Lactin model determines the daily rate of development at a given temperature and the Weibull model determines the cumulative frequency (%) at a given temperature and time. Therefore, the mathematical expression of the simulation is: where F(x, T) is the cumulative frequency (%) of the C. krombeini adult emergence at time x and constant temperature T; R(T) is the nonlinear Lactin model for the temperature-dependent development of C. krombeini (egg to adult); α and β are the scale and shape parameters estimated of the Weibull model against normalized time (egg to adult). From the simulated results in the cumulative frequency (%), the emergence of C. krombeini adults in daily frequency between time x and x +1 was calculated by subtracting the cumulative percentage of the cohort at time x from that at time x + 1.
Modeling voltinism of C. krombeini
Determining the number of generations (i.e., voltinism) obtained based on the climate data and developmental parameters, is useful to estimate the establishment potential for invasive species or introduced biological control agents [16,38]. To estimate the voltinism of C. krombeini, the Kearneysville Fruit Tree Research and Education Center (KFTREC) of West Virginia University located in Kearneysville, West Virginia (USA) (39° 23’ 45” N, 77° 53’ 44”; 160 m in elevation) was selected as a site for modeling and simulation of voltinism. The center had been utilized Osmia bees for orchard pollination and experienced problems associated with C. krombeini infestation during the past ten years. In addition, historical weather data (e.g., hourly temperature) directly collected at the center were available. The developmental parameters estimated herein were applied to calculate degree-days using daily minimum and maximum temperature data of ambient air from weather stations at KFTREC. Specifically, we used the lower development threshold (LDT) and upper developmental threshold (UDT) of 12.4°C and 37.8°C, respectively, for C. krombeini development from eggs to adults found in this study. Degree-days were calculated by using single sine/vertical cutoff methods  and cumulative degree-days (CDD) were obtained by summing calculated degree-days over a period of time. The voltinism of C. krombeini was calculated by dividing CDD per specific period (month or year) by the thermal constant requirement of C. krombeini development from eggs to adults.
In addition, the voltinism of C. krombeini was estimated based on two different types of biofix for degree-day accumulation each year (i.e., determining the first date for degree-day accumulation): one with calendar date of January 1st and the other with the predicted date for the emergence of O. cornifrons adults in spring. Osmia cornifrons adults generally emerge in April in West Virginia  and phoretic deutonymphs of C. krombeini need to move to a new bee nest by hitchhiking on newly-emerging Osmia bees. Therefore, the biofix using the predicted date of spring emergence for O. cornifrons adults could reflect more realistic life cycle of C. krombeini in the field. Spring emergence date of O. cornifrons adults was predicted by calculating the degree-day accumulation up to 126.1 DD with LDT and UDT of 10.5°C and 46.2°C, respectively, which were reported in a previous study on O. cornifrons . Degree-day calculations were done with single sine/vertical cutoff methods . After obtaining the predicted date of O. cornifrons adult emergence as a biofix, the degree-days for C. krombeini development were accumulated.
Projected voltinism of C. krombeini under the future climate scenarios
The projected voltinism of C. krombeini throughout the eastern United States (32–48°N and 66–87°W) was calculated by using the developmental parameters (i.e., LDT and UDT) of C. krombeini found in this study and the future temperature data for 2006–2100 based on various RCP scenarios: RCP 8.5, RCP 6.0, RCP 4.5, and RCP 2.6. Daily minimum and maximum surface temperatures were obtained from the coupled simulations of the Community Earth System Model version 1 using the Community Atmosphere Model version 5 (CESM1-CAM5)  under the four RCP scenarios. Based on the simulated model outputs covering 192 (latitude) by 288 (longitude) grid cells (i.e., 0.9375° by 1.25°) for 2006–2100, daily mean temperature was calculated by averaging the daily minimum and maximum temperatures.
To calculate the voltinism of C. krombeini in each grid cell over the eastern United States from 2016 to 2100, the daily mean surface air temperatures from the coupled CESM1-CAM5 simulations were used and two steps of calculations (i.e., bee emergence and C. krombeini development) were implemented (Fig 1). First, because C. krombeini starts its development after Osmia bees emerge so that it can hitchhike the bees to move to a new bee nest for colonization and development, the Julian day of bee emergence was required to be determined first. The LDT and thermal constant for O. cornifrons were set as 8.6°C and 179.8 DD, respectively, which were reported in White et al. . Degree-days were calculated by subtracting LDT from daily mean air temperature and used to obtain CDD for the bee emergence in each grid cell of the eastern United States. No accumulation of degree-days was made in days when daily mean air temperature was lower than LDT. Dates for the bee emergence in spring were obtained by determining the date when CDD was greater than or equal to thermal constant. Second, to calculate the voltinism of C. krombeini, LDT and thermal constant found in this study were used. Because it takes ca. two days for hitchhikes bees, move to a new nest, and lay eggs, the CDD for C. krombeini was calculated by adding two days into the Julian day of bee emergence (i.e., n + 2 in Fig 1). The voltinism of C. krombeini was determined by dividing CDD by thermal constant of C. krombeini. The calculated number of generations per year in each grid cell was displayed over the eastern United States based on the four RCP scenarios. A series of maps for the voltinism of C. krombeini were generated for the years of 2015, 2050, and 2100. In addition, the differences in the voltinism between 2015 and 2100 under different future climate scenarios were calculated to estimate the spatio-temporal changes in C. krombeini voltinism.
Ti, temperature at i-th date; LDT, lower temperature threshold; CDD, cumulated degree days; UDT, upper temperature threshold; TC, thermal constant.
To investigate the temporal variability in C. krombeini voltinism, a linear regression analysis was conducted by using the time series of the projected number of generations from 2006 to 2100. The number of generations per year in each grid cell over the eastern United States for the period of 2006–2100 was calculated, and then averaged the over the grid cells corresponding West Virginia (37–41°N and 77–83°W) and its surrounding regions. The trend of time series was analyzed with a regression analysis and the significance of the trend was tested by a Student’s t test . The same procedure described above was applied for all the four RCP scenarios.
Results are presented herein as means ± SEM, unless otherwise noted. Chaetodactylus krombeini developed from egg to the adult at all tested temperatures except the highest temperature (37.8°C), where no eggs hatched. The developmental duration of each stage was significantly influenced by temperatures: egg (F5, 276 = 107.17, P < 0.01), larva (F5, 182 = 183.70, P < 0. 01), nymph (F5, 146 = 70.61, P < 0.01), and egg to adult (F5, 145 = 241.28, P < 0.01) (Table 1). Mean time required for development from egg to adult emergence ranged from 8.8 days at 30.0°C up to 38.2 days at 16.1°C. Cumulative frequency distribution of C. krombeini development times at each temperature was well described by the Weibull model (r2 > 0.910, P < 0.01) and the median development time was obtained at each temperature (Table 1).
Developmental distribution model
Against the normalized time (days / median), the stage-specific frequency distribution of C. krombeini was also described by the Weibull model: egg (F1,39 = 1563.9, P < 0.01, r2 = 0.976), larva (F1,40 = 1847.5, P < 0.01, r2 = 0.979), nymph (F1,50 = 992.0, P < 0.01, r2 = 0.953), and egg to adult (F1,54 = 935.2, P < 0.01, r2 = 0.946) (Table 2, Fig 2).
A: egg, B: larva, C: nymph, and D: egg to adult.
For all the stages, temperature-dependent pattern of C. krombeini development rate over the entire range showed a typical skewed-bell shape, with sharp decline of the rates at high temperatures above the optimal temperature (Fig 3). The nonlinear Lactin model provided significant fit to the temperature-dependent rate of C. krombeini development: egg (F3,6 = 125.4, P < 0.01, r2 = 0.984), larva (F3,6 = 47.2, P < 0.01, r2 = 0.959), nymph (F3,6 = 7.8, P < 0.05, r2 = 0.795), and egg to adult (F3,6 = 27.4, P < 0.01, r2 = 0.932) (Table 3, Fig 3). The stage-specific optimal temperatures (Topt) with the maximal rate were estimated to be 31.4, 32.9, 32.6 and 32.5°C for egg, larva, nymph, and egg to adult, respectively. At the optimal temperatures, the earliest development completion would occur 2.9, 1.6, 3.3 and 8.5 days for egg, larva, nymph, and egg to adult, respectively. The stage-specific thermal ranges (B80) with ≥ 80% of the maximal rates were calculated to be 26.1–34.7°C, 28.8–35.4°C, 28.3–35.3°C and 28.1–35.2°C for egg, larva, nymph, and egg to adult, respectively.
A: egg, B: larva, C: nymph, and D: egg to adult.
Linear regression models were fit to the development rate data in the mid-range: egg (F1,4 = 176.4, P < 0.01, r2 = 0.983), larva (F1,5 = 45.9, P < 0.01, r2 = 0.920), nymph (F1,5 = 7.94, P < 0.05, r2 = 0.665), and egg to adult (F1,4 = 28.7, P < 0.01, r2 = 0.878) (Table 3, Fig 2). Based on the pattern of the nonlinear Lactin model, the data ranges chosen for linear regression analysis were 20.2–32.4°C for egg and 16.1–32.4°C for all the other stages. The LDTs were estimated to be 9.9, 14.7, 13.0 and 12.4°C for egg, larva, nymph, and egg to adult, respectively. The thermal constant (DD) over the stage-specific LDT were 61.5, 28.1, 64.8 and 171.8 DD for egg, larva, nymph, and egg to adult, respectively.
Given a cohort of C. krombeini eggs under constant temperatures, the predicted daily frequency (%) of adult emergence was presented in relation to temperature (°C) and time (day) (Fig 4). The predicted values showed that adult emergence of C. krombeini would occur earlier at the optimum temperature in much shorter time but the adult emergence at both ends of temperatures would occur later over extended periods. From a cohort of C. krombeini eggs, for instance, the adult emergence would occur in 8–10 days at 34°C, whereas it would take 70–84 days at 12°C and 21–25 days at 37°C.
Simulation of adult emergence
The mean number of generations per month, estimated from degree-days available per month at KTFREC, was shown in Table 4. We found differences in the degree-days accumulated depending on the calendar months (F11, 107 = 276.30, P < 0.01) and the number of generations possible for C. krombeini per month (F11, 107 = 248.18, P < 0.01). The number of generations highly varied depending on the season, with the lowest (0.01) in February and the highest (2.16) in July. At least one generation per month would be possible during May–September. Based on the biofix of January 1st, the voltinism of C. krombeini ranged from 8.6 to 10.7 generations at KTFREC over the past nine years (Table 5). Similarly, predicted dates of O. cornifrons emergence in spring varied from March 28th to May 1st, depending on the year.
By using the host emergence date as the biofix each year, the numbers of generations per year were estimated to be less (i.e., 8.1 to 10.1 generations per year), compared to the estimated values using the biofix of January 1st. However, the difference was not significant (P > 0.05); the small difference appears to have resulted from the lack of degree-days under the low temperatures prevailing at KTFREC during January through April (Table 4).
Projected voltinism of C. krombeini under the future climate scenarios
Spatial distributions of projected number of generations of C. krombeini over the eastern United States showed significant differences (Fig 5) among the years of 2015, 2050, and 2100 under the four RCP scenarios. In 2015, the voltinism from the four RCP scenarios was estimated to be 6–10 generations in the eastern part of West Virginia where KTFREC was located. This estimation of voltinism based on the future climate scenarios was consistent with that from our simulation using two different biofixes at KTFREC. The voltinism of C. krombeini was generally higher in the lower latitude regions in the eastern United States. In 2015, the voltinism was consistent among the four RCP scenarios with gradual increasing from four to eighteen generations from the northern to southern parts of the eastern United States. The number of C. krombeini generations increased by 2050 (Fig 5B–5E, 5H and 5K) and more apparent increase was predicted by 2100 (Fig 5C, 5F, 5I and 5L). The rates of changes in C. krombeini generations in the eastern United States from 2015 to 2100 were different among the four RCP scenarios (Fig 6). The voltinism increased by 0–2 and 4–10 generations per year in RCP 2.6, and RCP 8.5 scenarios, respectively, by the year of 2100 (Fig 6A–6D). The voltinism based on the RCP 4.5 (Fig 6C) and RCP 6.0 (Fig 6B) scenarios were increased by 2–4 and 2–5 generations, respectively, by 2100. In all four climate scenarios, the rate of increase in voltinism of C. krombeini was higher in the lower latitudes of the eastern United States.
To examine the temporal dynamics of the C. krombeini voltinism under the future climate scenarios, we performed the liner regression analysis by using the time series of the generation number. The number of C. krombeini generations was estimated to be significantly (P < 0.01) increased by 2100 under the four RCP scenarios (Fig 7). The greatest rate of increase in voltinism was 0.83 generations per decade based on the RCP 8.5 over the 95 years from 2006 to 2100. The predicted rates of increase in voltinism based on the RCP 6.0 and 4.5 were 0.55 and 0.40 generation per decade, respectively. In RCP 2.6, the lowest rate of increase in voltinism was 0.16 generation per decade and the time-series trend appeared relatively stationary beyond 2050. The results under the projected climate change indicated that the number of C. krombeini generations in the central Appalachian regions would increase more likely up to 1.5–2.0 times above their current levels (i.e., 8–10 generations per year) by the end of the 21st century.
This study firstly demonstrated the stage-specific thermal development of C. krombeini by applying empirical models, which provided developmental parameters crucial to predict phenology and damage of the mite. Development rate models in this study showed that C. krombeini development from egg to adult emergence would occur within the thermal range of 12.4°C and 37.8°C (i.e., between LDT and UDT) and the fastest growth was observed at 32.5°C (i.e., optimum temperature for development). At the optimum temperature, it would take 8.54 days for the mite egg to develop into adult, which also indicates the maximum number of generations would be possible in a season when the ambient temperature prevails near the optimum temperature or at locations where temperature in regional climate stays longer near the optimum temperature. When an actual field temperature was used to calculate the voltinism more realistically (Table 5), the mite would be able to complete up to 8–10 generations per year in Kearneysville, West Virginia (USA). Seasonal pattern of voltinism at the location showed much variation depending on the months with higher values in summer (May-September), which clearly suggests that the early application of control measures on the infested nest must be considered to prevent the rapid population increase in summer.
In commercial orchards, bundles of nests (e.g. cardboard tubes or reeds) are typically used to facilitate easy maintenance of Osmia bees [23,28] including O. cornifrons and O. lignaria. Such method, however, provides an ideal habitat for the mites [22,26]. Currently, C. krombeini is controlled by cultural and chemical methods such as removing bee cocoons from bee nests to remove C. krombeini , cleaning bee cocoons with bleach solution to clean C. krombeini [22,41], and using essential oils  to fumigate bee nests and kill C. krombeini directly. Alternatively, treating bee nests with very low or high temperatures would negatively affect the survival and population of C. krombeini. Because Osmia bees’ development and survival also depend on temperature [43,44], determining the optimal temperature that maximize both the bee’s survival and maximize the mite’s mortality would be important to use temperature as a physical control method. Results of our study suggest the potential of using high temperature as a control measure against C. krombeini for Osmia bee propagation and maintenance, based on the difference in optimal temperatures and UDTs between the Osmia bee and the C. krombeini. In spring, optimum temperature and UDT of O. cornifrons adult emergence are 35.7 and 46.2°C, respectively , whereas those for C. krombeini development herein are 32.5 and 37.8°C, respectively. This indicates that C. krombeini would suffer more than O. cornifrons if exposed to high temperatures (e.g. 35.7–37.8°C). During summer diapause of O. cornifrons, this method would be more effective because the diapausing O. cornifrons becomes tolerant to heat while the mites still remain susceptible to high temperatures . The effects of exposure duration and high-temperature treatment on both C. krombeini and O. cornifrons survival have yet to be determined in a future study in order to find out the optimal combination of the exposure duration and the treatment temperature, which can maximize the mortality of C. krombeini but have little or no impact on O. cornifrons survival. Moreover, heat treatment may be also considered as a post-treatment option after chemical control, if a certain pesticide in apiculture may have positive temperature-dependent toxicity [46,47].
The voltinism of C. krombeini estimated with the daily mean surface air temperature from the coupled CESM1-CAM5 simulations were consistent with those from the experimental results in Kearneysville, West Virginia. The spatial patterns from the simulated climate data well captured the latitudinal gradients of C. krombeini voltinism depicted by the increased number of generations from north to south over the eastern United States. Additionally, the lower numbers of generations in the central and southern Appalachian regions, compared to the surrounding regions in the same latitudes, were generally explained by showing the isolines bending southward in the mountainous regions (Fig 5). Although the climatic data used in this study represented well the overall spatial patterns of the voltinism in the eastern United States with the daily temporal interval and approximately 100 km2 spatial resolution, a higher spatial resolution should be beneficial to capture the local and regional changes in the mite generations under the future climate change.
In addition to temperature, water vapor deficit (VPD) can be an important factor that affects C. krombeini under future climate change. VPD is the difference between saturation vapor pressure and actual vapor pressure . Actual vapor pressure and relative humidity are linearly associated when air temperature is constant and saturation vapor pressure increases exponentially as temperature increases . Therefore, under the condition with increased VPDs at higher temperature, insects and mites can lose higher amount of moisture to the air by evaporation. Although the effect of VPD on C. krombeini has not been reported, previous studies documented increased mortality of some predatory mites (e.g. Amblyseius cucumis and Neoseiulus califonicus) at higher VPDs [48,50–55]. We also observed a higher mortality rate of C. krombeini at higher temperatures in our study, which also could be caused by higher VPDs because relative humidity was kept constant at 60±10% in all growth chambers (i.e. same or very similar actual vapor pressures across chambers). However, no studies have found that VPD could affect the rate of insect or mite development which is a key function in modeling temperature-dependent development of C. krombeini in our study although effects of VPD on mite development was studied . Cook et al.  successfully simulated the increased VPD with the RCP 8.5 scenarios at the global scale, and thus additional studies to investigate the effect of various VPDs on mite development could lead to incorporating VPD in the development model and generating more realistic prediction of future biological events of the mite (e.g. development period and voltinism).
Operative thermal ranges of C. krombeini and the Osmia bee clearly show that the host and the parasite have a discreet thermal sensitivity profile for development , indicating that Osmia bees and C. krombeini would respond differently exposed to the same temperature in the field. While the short-term exposure of high temperature is of great interest for the management of C. krombeini (i.e., sensitivity to high temperature), the relatively small and realistic change of climate in the long-term may alter drastically the host-parasite interaction, geographical distribution, and abundance as shown in other species [15, 58]. Despite sensitivity to high temperature extremes, the mite may acquire tolerance to the extreme by acclimation following non-lethal condition and the slowly-elevated temperature may allow an increase in the voltinism per year due to faster development and longer season available for multiple reproductive phases.
Results herein provide fundamental information regarding the thermal development of all the life stages of C. krombeini. C. krombeini development from egg to adult emergence would occur within the thermal range of 12.4°C and 37.8°C (i.e., between LDT and UDT) and the fastest growth was observed at 32.5°C (i.e., optimum temperature for development). This study also showed how the mite’s voltinism could be affected by not only future temperature increases but also host bees’ temperature-dependent spring emergence which is also affected by the future temperature increase. Temperature can play a major role in altering voltinism of C. krombeini according to the future climate change scenarios; the number of generations of the mite would increase by 1.5–2.0 times by the year of 2100.
In addition, findings in this study firstly provided comprehensive data on thermal development of C. krombeini and implications for an effective control and mitigation strategy against C. krombeini for the propagation and management of Osmia bees for pollination. Operative thermal ranges of C. krombeini and the Osmia bee are different and they have a discreet thermal sensitivity profile for development, indicating the potential of using temperature as a method to control C. krombeini; high-temperature treatment may have potential for C. krombeini control because the mite is more sensitive to high temperatures than Osmia bees. Future studies may be needed to determine the optimal combination of the exposure duration and the treatment temperature.
We are very grateful to Dr. Pavel Klimov (University of Michigan) for confirming that Chaetodactylus krombeini used in this study was correctly identified. The study was supported by the USDA National Institute of Food & Agriculture (NIFA) Grant (NIFA Award: 2014-67004-21631). We acknowledge the World Climate Research Programme's Working Group on Coupled Modelling, which is responsible for CMIP, and we thank the climate modeling groups, especially the CESM groups, for producing and making available their model output. For CMIP the United States Department of Energy's Program for Climate Model Diagnosis and Intercomparison provides coordinating support and led development of software infrastructure in partnership with the Global Organization for Earth System Science Portals.
- Conceptualization: YLP EL.
- Data curation: YLP EL.
- Formal analysis: YS JJA YH.
- Funding acquisition: YLP EL.
- Investigation: JJA YS YH EL YLP.
- Methodology: YLP EL.
- Project administration: YLP.
- Resources: YLP.
- Software: JJA YS YH.
- Supervision: YLP.
- Validation: YLP YS.
- Visualization: EL YLP.
- Writing - original draft: YS JJA YH EL YLP.
- Writing - review & editing: YS EL YLP.
- 1. Field CB, Lobell DB, Peters HA, Chiariello NR (2007) Feedbacks of terrestrial ecosystems to climate change. Annu Rev Environ Resour 32: 1–29.
- 2. Pautasso M, Dehnen-Schmutz K, Holdenrieder O, Pietravalle S, Salama N, Jeger MJ (2010) Plant health and global change—some implications for landscape management. Biol Rev Camb Philos Soc 85: 729–755. pmid:20105153
- 3. Schweiger O, Settele J, Kudrna O, Klotz S, Kühnet I (2008) Climate change can cause spatial mismatch of trophically interacting species. Ecology 89: 3472–3479. pmid:19137952.
- 4. IPCC (2013) Summary for policymakers. In: climate change 2013: the physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
- 5. Allen JC, Yang Y, Knapp YL (1995) Temperature effects on development and fecundity of the citrus rust mite (Acari: Eriophyidae). Environ Entomol 24: 996–1004.
- 6. Angilletta MJ Jr. (2009) Thermal adaptation: A theoretical and empirical synthesis. Oxford University Press. New York. 289 p.
- 7. Danilevskii AS (1965) Photoperiodism and seasonal development of insects. Oliver and Boyd LTD, Edinburgh and London. 283 p.
- 8. Gadino AN, Walton VM (2012) Temperature-related development and population parameters for Typhlodromus pyri (Acari: Phytoseiidae) found in Oregon vineyards. Exp Appl Acarol 58: 1–10. pmid:22527839
- 9. Gotoh T, Sait M, Suzuki A, Nachman G (2014) Effects of constant and variable temperatures on development and reproduction of the two-spotted spider mite Tetranychus urticae (Acari: Tetranychidae). Exp Appl Acarol 64: 465–478. pmid:25047962
- 10. Park H-H, Ship L, Buitenhuis R, Ahn JJ (2011) Life history parameters of a commercially available Amblyseius swirskii (Acari: Phytoseiidae) fed on cattail (Typha latifolia) pollen and tomato russet mite (Aculops lycopersici). J Asia-Pac Entomol 14: 497–501.
- 11. Briére JP, Pracros P, LeRoux AY, Pierre JS (1999) A novel rate model of temperature-dependent development for arthropods. Environ Entomol 28: 22–29.
- 12. Lactin DJ, Holliday NJ, Johnson DL, Craigen R (1995) Improved rate model of temperature-dependent development by arthropods. Environ Entomol 24: 68–75.
- 13. Logan JA, Wollkind DJ, Hoyt SC, Tanigoshi LK (1976) An analytic model for description of temperature dependent rate phenomena in arthropods. Environ Entomol 5: 1133–1140.
- 14. Shi P, Ge F, Sun Y, Chen C (2011) A simple model for describing the effect of temperature on insect development rate. J Asia-Pac Entomol 14: 15–20.
- 15. Hance T, van Baaren J, Vernon P, Boivin G (2007) Impact of extreme temperature on parasitoids in a climate change perspective. Annu Rev Entomol 52: 107–126. pmid:16846383
- 16. Hart AJ, Tullett AG, Bale JS, Walters KFA (2002) Effects of temperature on the establishment potential in the U.K. of the non-native glasshouse biocontrol agent Macrolophus caliginosus. Physiol Entomol 27: 112–123.
- 17. Hatherly IS, Hart AJ, Tullett AG, Bale JS (2005) Use of thermal data as a screen for the establishment potential of non-native biological control agents in the UK. BioControl 50: 687–698.
- 18. Ims RA (1990) The ecology and evolution of reproductive synchrony. Trends Ecol Evol 5: 135–140. pmid:21232341
- 19. Son Y, Nadel H, Baek S, Johnson MW, Morgan DJW (2012) Estimation of developmental parameters for adult emergence of Gonatocerus morgani, a novel egg parasitoids of the glassy-winged sharpshooter, and development of a degree-days model. Biol Contr 60: 233–240.
- 20. White J, Son Y, Park Y-L (2009) Temperature-dependent emergence of Osmia cornifrons (Hymenoptera: Megachilidae) adults. J Econ Entomol 102: 2026–2032. pmid:20069827
- 21. Moss RH, Edmonds JA, Hibbard KA, Manning MR, Rose SK, van Vuuren DP et al. (2010) The next generation of scenarios for climate change research and assessment. Nature 463: 747–756. pmid:20148028
- 22. Bosch J, Kemp WP (2001) How to manage the blue orchard bee as an orchard pollinator, Sustainable Agricultural Network, Handbook No. 5, Beltsville. 98 p.
- 23. Qu D-C, Maeta Y, Goubara M, Nakatsuka KJ, Kitamura K (2002) Reproductive strategy in the two species of Cleptoparasitic Astigmatid mites, Chaetodactylus nipponicus and Tortonia spp. (Acari: Chaetodactylidae and Suidasiidae), infesting Osmia cornifrons (Hymenoptera: Megachilidae) I. Invasion/infestation patterns and partitive use of the host food. Jap J Entomol 5: 121–141.
- 24. Krombein KV (1962) Natural history of Plummers island Maryland: XVI. Biological notes on Chaetodactylus krombeini Baker, a parasitic mite of the Megachilid bees, Osmia (Osmia) lignaria Say (Acarina: Chaetodactylidae). Proc Biol Soc Wash 75: 237–250.
- 25. Sugiura N, Maeta Y (1989) Parental investment and offspring sex ratio in a solitary mason bee, Osmia cornifrons (Radoszkowski) (Hymenoptera, Megachilidae). Jap J Entomol 57: 861–875.
- 26. Park Y-L, Kondo V, White JB, West T, McConnell B, McCutcheon T (2009) Nest to nest dispersal of Chaetodactylus krombeini (Acari, Chaetodactylidae) associated with Osmia cornifrons (Hym. Megachilidae). J Appl Entomol 133: 174–180.
- 27. McKinney MI, Park Y-L (2013) Distribution of Chaetodactylus krombeine (Acari: Chaetodactylidae) within Osmia cornifrons (Hymenoptera: Megachilidae) nests: implications for population management. Exp Appl Acarol 60: 153–161. pmid:23100109
- 28. Batra SWT (1998) Hornfaced bees for apple pollination. Am Bee J 138: 364–365.
- 29. Klimov PB, O’Connor BM, Knowles LL (2007) Museum specimens and phylogenies elucidate ecology’s role in coevolutionary associations between mites and their bee hosts. Evolution 61: 1368–1379. pmid:17542846
- 30. Klimov PB, O’Connor BM (2008) Morphology, evolution, and host association of bee-associated mites of the family Chaetodactylidae (Acari: Astigmata) Miscellaneous Publications 199. Museum of Zoology, University of Michigan, Ann Arbor MI. 243 p.
- 31. SAS Institute (1995) SAS user’s guide statistics version 9.1ed. SAS Institute, Cary, NC.
- 32. Scientific Jandel (1996) TableCurve User’s Manual San Rafael, CA.
- 33. Weibull W (1951) A statistical distribution functions with wide applicability. J Appl Mech 18: 293–297.
- 34. Régnière JA (1984) A method of describing and using variability in development rates for the simulation of insect phenology. Can Entomol 116: 1367–1376.
- 35. Wagner TL, Wu HI, Sharpe PJH, Schoolfield RM, Coulson BN (1984) Modeling insect development rates: A literature review and application of a biophysical model. Ann Entomol Soc Am 77: 208–225.
- 36. Campbell A, Frazer BD, Gilbert N, Gutierrez AP, Mackauer M (1974) Temperature requirements of some aphids and their parasites. J Appl Ecol 11: 431–438.
- 37. Lutterschmidt WJ, Hutchison VH (1997) Critical thermal maximum: history and critique. Can J Zool 75: 1561–1574.
- 38. Mills N (2005) Selecting effective parasitoids for biological control introductions: codling moth as a case study. Biol Control 34: 274–282.
- 39. UC IPM (2014) UC IPM Statewide Integrated Pest Management Program—How to manage pests: run models and calculate degree days. Available: http://www.ipm.ucdavis.edu/WEATHER/ddretrievetext.html
- 40. Meehl GA, Washington WM, Arblaster JM, Hu A, Teng H, Kay JEet al. (2013) Climate Change Projections in CESM1(CAM5) Compared to CCSM4. J Climate 26: 6287–6308.
- 41. Krunić M, Stanisaljevic L, Pinzauti M, Felicioli A (2005) The accompanying fauna of Osmia cornuta and Osmia rufa and effective measures of protection. Bull Insect 58: 141–152.
- 42. White J, Park Y-L, West TP, Tobin PC (2009) Assessment of potential fumigants to control Chaetodactylus krombeini (Acari: Chaetodactylidae) associated with Osmia cornifrons (Hymenoptera: Megachilidae). J Econ Entomol 102: 20290–2095. pmid:20069836
- 43. Bosch J, Sgolastra F, Kemp WP (2010) Timing of eclosion affects diapause development, fat body consumption and longevity in Osmia lignaria, a univoltine, adult-wintering solitary bee. J Insect Physiol 56: 1949–1957. pmid:20816851
- 44. Radmacher S, Strohm E (2011) Effects of constant and fluctuating temperatures on the development of the solitary bee Osmia bicornis (Hymenoptera: Megachilidae). Apidologie 42: 711–720.
- 45. Yamada M (1990) Control of the Chaetodactylus mite, Chaetodactylus nipponicus Kurosa, an important mortality agent of hornfaced Osmia bee, Osmia cornifrons Radoszkowski. Bull Aomori Apple Exp Stat 26: 39–77.
- 46. Boina DR, Onagbola EO, Salyani M, Stelinski LL (2009) Influence of posttreatment temperature on the toxicity of insecticides against Diaphorina citri (Hemiptera: Psyllidae). J Econ Entomol 102: 685–691. pmid:19449650
- 47. Johnson RM, Ellis MD, Mullin CA, Frazier M (2010) Pesticides and honey bee toxicity—USA. Apidologie 41: 312–331.
- 48. Ghazy NA, Suzuki T, Shah M, Amano H, Ohyama K (2012) Effect of long-term cold storage of the predatory mite Neoseiulus californicus at high relative humidity on post-storage biological traits. BioControl 57: 635–641.
- 49. Eaton M, Kells SA (2009) Use of vapor pressure deficit to predict humidity and temperature effects on the mortality of mold mites, Tyrophagus putrescentiae. Exp Appl Acarol 47: 201–213. pmid:18979171
- 50. Kramer DA, Hain FP (1989). The effect of constant-humidity and variable-humidity and temperature regimes on the survival and developmental periods of Oligonychus ununguis (Acarina: Tetranychidae) and Neoseiulus fallacis (Acarina: Phytoseiidae). Environ Entomol 18: 741–746.
- 51. Shipp JL, Ward KI, Gillespie TJ (1996) Influence of temperature and vapor pressure deficit on the rate of predation by the predatory mite, Amblyseius cucumeris, on Frankliniella occidentalis. Entomol Exp Appl 78: 31–38.
- 52. Shipp JL, van Houten Y (1997) Influence of temperature and vapor pressure deficit on survival of the predatory mite Amblyseius cucumeris (Acari: Phytoseiidae). Environ Entomol 26: 106–113.
- 53. Ghazy NA, Suzuki T, Amano H, Ohyama K (2014) Air temperature optimization for humidity-controlled cold storage of the predatory mites Neoseiulus californicus and Phytoseiulus persimilis (Acari: Phytoseiidae). Pest Mang Sci 70: 483–487.
- 54. Ghazy NA, Suzuki T, Amano H, Ohyama K (2012) Effects of air temperature and water vapor pressure deficit on storage of the predatory mite Neoseiulus californicus (Acari: Phytoseiidae). Exp Appl Acarol 58: 111–120. pmid:22527835
- 55. Ghazy NA, Suzuki T, Shah M, Amano H, Ohyama K (2012) Using high relative humidity and low air temperature as a long-term storage strategy for the predatory mite Neoseiulus californicus (Gamasida: Phytoseiidae). Biol Contr 60: 241–246.
- 56. Mangini AC jr, Hain FP (1991) Vapor pressure deficit differentially affects laboratory populations of Metaseiulus occidentalis and Neoseiulus fallacis (Acarina: Phytoseiidae) reared together. Environ Entomol 20: 823–831.
- 57. Cook BI, Smerdon JE, Seager R, Coats S (2014) Global warming and 21st century drying. Clim Dyn 43: 2607–2627.
- 58. Gutierrez AP, Ponti L, d’Oultremount T, Ellis CK (2008) Climate change effects on poikilotherm tritrophic interactions. Clim Change 87 (Suppl 1) S167–S192.