Electronic SupplementaryMaterial
Journal:Journal of Pest Science
Article title: Invasion biology of Spotted Wing Drosophila (Drosophila suzukii): a global perspective
and future priorities
Article Authors:Mark K. Asplen*, Gianfranco Anfora, Antonio Biondi, Deuk-Soo Choi, Dong Chu, Kent M. Daane, Patricia Gibert, Andrew P. Gutierrez, Kim A. Hoelmer, William D. Hutchison, Rufus Isaacs, Zhi-Lin Jiang, Zsolt Kárpáti, Masahito T. Kimura, Marta Pascual, Christopher R. Philips, Christophe Plantamp, Luigi Ponti, Gábor Vétek, Heidrun Vogt, Vaughn M. Walton, Yi Yu, Lucia Zappalà, Nicolas Desneux*
Article Doi: 10.1007/s10340-015-0681-z
Corresponding authors:M.K. Asplen: Natural Sciences Department; Metropolitan State University; 700 East 7th Street; Saint Paul, MN 55106, USA; phone: (1) 651-793-1518; e-mail:
N. Desneux: French National Institute for Agricultural Research; 400 route de Chappes; Sophia-Antipolis, France; phone (33) 4 9238 64 27; e-mail:
Tabular Data for SWD Detections in Germany and Eastern Europe
Table S1. SWDoccurrence in Germany, as of December 2014.
Location / First occurrenceRheinland-Pfalz
Baden-Württemberg / 2011
Bayern
Hessen
Nordrhein-Westfalen / 2012
Niedersachsen
Brandenburg
Berlin
Mecklenburg-Vorpommern / 2013
Schleswig-Holstein
Sachsen
Hamburg
Saarland / 2014
Sachsen-Anhalt
Thüringen
Table S2. Recent confirmation of SWD in Hungary, Serbia and Slovakia (all collections in 2014).
Country / Location / End date of trapping / Male(s) / Female(s)*Hungary / Sóskút / 19. Oct. 2014 / 2 / 2
Balatonszárszó / 19. Oct. 2014 / 2 / 3
Győr / 20. Oct. 2014 / 3 / 2
Szekszárd / 20. Oct. 2014 / 2 / 2
Visegrád / 18. Oct. 2014 / 3 / 2
Szigetszentmiklós / 02. Nov. 2014 / 3 / 2
Diósd / 21. Oct. 2014 / 2 / 2
Tiszaalpár / 16. Nov. 2014 / 2 / 2
Szajol / 27. Oct. 2014 / 3 / 3
Gyula / 30. Oct. 2014 / 2 / 3
Röszke / 24. Oct. 2014 / 1 / 1
Ősagárd / 25. Oct. 2014 / 2 / 1
Kecskemét / 09. Oct. 2014 / 2 / 2
Serbia / Senta / 09. Nov. 2014 / 3 / 2
Horgoš / 31. Oct. 2014 / 2 / 1
Slovakia / Štúrovo / 18. Oct. 2014 / 4 / 3
Rimavská Sobota / 26. Oct. 2014 / 1 / 1
*Sample size for males and females is also based on confirmation of SWD identifications by A. Blanton, University of Minnesota.
Modeling spotted winged drosophila (SWD) A.P. Gutierrez and L. Ponti
The methods for physiologically-based demographic modeling (PBDM) have been widely reviewed (see Gutierrez and Baumgärtner 1984; Gutierrez 1996; Gutierrez and Ponti 2013) (see appendix below). The ultimate goal of the model is to predict prospectively the phenology and geographic range and abundance of the fly using observed and/or predicted weather patterns. The biological data used to formulate a preliminaryPBDM for the spotted winged Drosophila(Drosophilasuzukii, hereafter SWD)are incomplete or highly variable. Unless otherwise indicated, the data usedto develop the PBDM were estimated from published graphs and tables (see data below).
Biology ofSWD
Like other Drosophila species, SWDhas multiple generations per year,does not have a true diapause stage and overwintersas “reproductively quiescent” long-lived adults (see Izquierdo1991; Dalton et al. 2011, and below). During favorable periods, growth, development and reproduction of fly life stages depend on temperature, likely measures of the drying power of the air, and host age and availability.Adult flies are assumed to enter facultative reproductive quiescence when hosts are unavailable or temperatures are low (see below).In the model, we assume that host materials (fruit, dehisced fruit, etc.) are available for reproduction when temperatures are in the favorable range. This is an obvious deficiencythat can be easily corrected by including the phenology of hosts in an area.
Rate of development of SWD during favorable periods
The data and function for the SWD developmental rate for the egg to adult period (subscript e-a) on temperature (T) are illustrated in Fig. S1 (eqn. 1;cf., Brière et al.1999).
[1]
Fig. S1. Developmental time in days of the egg to adult period (days) and the developmental rate (1/days) plotted on temperature.Data from P. Gibert and C. Plantamp (unpublished) (◌, ●) and Tochen et al. (2014) (red).
A lower thermal threshold of 5.975 °C was estimated using data from Table S3, with the developmental rate declining to zero at about 31.5°C (an estimate reported byKinjo et al. 2014). Using the 5.975 °C threshold, average developmental times for the adult longevity in degree days [dd] were computed in the midrange of favorable temperatures (e.g.,) (Table S3). The daily change in dd for development at time t is computed asusing the egg to adult period (243 dd) as the base.
Table S3.Developmental times in degree days above5.975°C computed in the favorable linear mid-range of temperature (P. Gibert and C. Plantamp [unpublished] and Kinjo et al. [2014]).
Stage degree days
egg19.025
larval121.76
pupal93.22
adult1,050 (75days at 20C; P. Gibert and C. Plantamp, unpublished)
Temperature dependent mortality rate
Dalton et al. (2011) concluded that long-term survival of D. suzukii is unlikely at temperatures below 10°C. The mortality rate of SWD life stagesacross temperatures (Fig.8 in the main article) was estimated from survivorship data from theP. Gibertlaboratory for the egg to adult stage in the 10-30°Crange (unpublished), adult survival at temperatures in the -2-10°Crange (Dalton et al. 2011),and adult data at higher temperatures (25-33°C) (Kinjo et al. 2014).The Dalton et al. (2011) data at low temperatures are inconsistent at 1 and 3 °C (note square values in Fig. 8b in the main article),possibly due to difficulties in accurately determining mortality. Higher mortality rates in the Dalton et al. (2011) data may also be explained by use of an alternative food source to that used in the other two studies (fruits versus optimized artificial medium). A second analysis was thus performed with a subset of these low temperature values (compare Figs. 8b and 8c in the main article). The mortality rates of adult flies taken from different temperature regimes and exposed to -2°C for 7 days were roughly the same before and after the cold temperature treatment, and cold hardeningof individuals was not apparent.The values at 10°C were the same for the Daltonet al. and P. Gibertlaboratory data, and hence we combinedthe three data sets to estimatetwo mortality functions.
An inelegant polynomial function (eqn. 2i) captures the mortality data rather well despite the inconsistencies at 1 and 3°C,
[2i]
A simple convex function(eqn. 2ii; eqn. 2 in the main article; Fig. 8 in the main article) centered on 15ºCwas also fit but did not capture the data as well.
[2ii]
Mortality rate data from Tochen et al. (2014) in the range 10 – 30C were higher inthe upper range,and suggest that asimple convex function may be appropriatefor the Dalton et al. (2011) -Tochen et al. (2014) data:
).[2iii]
These early attempts at modeling highlight inconsistencies in empirical studies of SWD temperature-dependent mortality, such that the current data cannot be adequately explained by a single function.In general, the results suggest that additional experiments using the same developmental stages (e.g., egg-adultemergence) across the full range of temperature are required to clarify the discrepancies, and the illustrate the issues with using data from across different studies, locations, and life stages.
Reproduction
Fecundity for females of a French population was estimated as the number of eggs (±std, n=20) laid every two day at 21°C on a banana lab medium (see Chabert et al. 2013). Data on the average age-specific (x = days) oviposition rate (f(x) = eggs /♀at 21 ºC [data from Chabert et al. 2013, Fig.S2]) werefit using the model proposed by Bieri et al. (1983) (eqn. 3). The pre-oviposition period is less than a day,after which f (x) increases to 4.5 eggs d-1 at x=20 days, and then begins to declines. This figure captures the general pattern, but likely underestimates the fly’s potential.Data for D. melanogasterreported in David and Clavel (1965) are nearly 10 fold higher, and peak earlier.
,[3]
Fig. S2. Age-specific fecundity of SWD at 21C (from Chabert et al. 2013). The heavy solid line is eqn. 3.
The total number of eggs (F(t,T)) produced at time t (i.e., day) by the population of females (N(x,t)) is affected by age (x), sex ratio (sr = 0.5), temperature () and relative humidity((see below).
Fig. S3. Scalars for the effect of temperature (a; see Tochen et al. 2014) and assumed relative humidity (b) on SWD reproduction. The up arrows in (S3a) represent the proportion of adult flies entering or exiting reproductive quiescence. The open symbol O=2 (dashed box) in Fig. S3a is likely an error.
Theeffects of temperature and humidity on SWD fecundity are assumed to be concave (Fig. S3). With the exception of the value at 18C, data from Tochen et al. (2014) suggest that the normalized scale is concave in the temperature range12.75-29°C, with the peak at20.65°C, (eqn. 4, Fig.S3a).
[4]
Observations on the effects of relative humidity on fecundityarenot currently available in the literature,hence we assumeasymmetrical concave scalarin the range 25-100%RH (Fig. S3b;eqn. 5).
[5]
The effects of and on populationlevel reproductionare indicated ineqn. 6.
[6]
Reproductive quiescence
Entry and exit from reproductive quiescence is assumed to hinge on 12.75ºC.When hostsare rare and/or temperature drops below theoviposition thresholdof12.75C, females become reproductively quiescent. The proportion of females that enter the quiescentphasein response to 5-d average temperature (Tavg)(eqn. 7i)and the individuals are transferred to the same age class in a separate population array (e.g., Fig. S4). Specifically, the proportion enteringquiescence is:
if Tavg(t)12.75 then 0enter quiescence = 0.2*(12.75-Tavg(t)) 1.0[7i]
As temperatures warm above 12.75C, theproportion of females exiting thequiescent phaseis
if Tavg(t)12.75 then 0exit quiescence = 0.2*(Tavg(t)-12.75) 1.0.[7ii]
Entry or exit from reproductive quiescence occurs between equivalent age classes (i=1…K), butflies continue to age and die overtime at temperature dependent rates (i.e., individuals move down the aging array or exit as deaths, seeFig. S4). For example,adultsmay enterreproductive quiescenceat age i at time t, but may become reproductive again at some future time. The daily effects of temperature on the dynamics of SWD life stages are captured using a time-invariant, distributed-maturation time demographic model of Manetsch (1976) andVansickle (1977) (see also Di Cola et al., 1999, pp. 523-524).
Fig. S4. Aging and transfer between reproductive and quiescent adult female SWD (see text).
The mortality rate of quiescent females is unknown, but appears to be roughlythe sameas that of reproductive females at temperatures below 10 ºC. The longevity of reproductivelyquiescent females is assumed to be longer (1050 vs.1200dd; Dalton et al. 2011).
Model predictions
The model is incomplete and major assumptions were made. Despite this, the model was used to simulate the continuous dynamics of the fly across several yearsat hundreds of locations in the US-Mexico, Europe, and the Mediterranean Basin. The observed distribution of SWD in the US is illustrated in Fig. S5.
Fig.S5. The observed distribution of SWD in the US.[ map.php?code=IOAPAUA#., accessed on 15 October 2014].
The observed dynamics of SWD are illustrated in Fig. S6afor Stockton, San Joaquin County,CA, while the simulated prospective dynamics during 2000-2003 are illustrated in Fig. S6b. The irregularities in the simulated dynamicsreflect the influence of daily weather, especially during summer and winter on survival and reproduction.Adultreproduction is most affected by temperature and host availability, while temperature affects all stages. The prospective densities are relative and should be viewed as indices of abundance. The simulated,double-peaked yearly patterns for 2000-20003 are similar to that reported by Dalton et al. (2011) for San Joaquin Co. (compare Figs. S6a and S6b). The population declines duringwinter and summer, depending on the presence of adverse temperatures.
Fig.S6. Observed (redrawn from Dalton et al. 2011) and simulated dynamics of SWD San Joaquin Co, CA andat Stockton, San Joaquin Co, CA. Temperature data from the original Fig.S6a have been smoothed.
GIS maps of theprospective distribution of SWD in the US-Mexico(Fig. S7) and the Mediterranean Basin (Fig. S8)
Sound data on SWD reproductive quiescence and winter survival arerequired to fully capture the dynamics of the fly as affected by weather; hence anyresults must be viewed with caution. As currently modeled, cold winter temperatures limit the fly in northern areas and hot summer temperatures limit it in hotter areas (e.g., Arizona and the desert of California; see Fig. S7). Highest cumulative egg densities (i.e.,a metric of favorability) are predicted in Florida and subtropical-tropical Mexico. Favorability is inversely related to the coefficient of variability of population size (see insert). Seasonal migration of SWD from warmer climesto colder northern areas during late summer may also be a factor, but the occurrence or extent of this is unknown. Maps of cumulative mortality rates (e.g., eqn. 2ii) above and below 15C are shown in Fig.S7 c,d.
Fig. S7. Prospective distribution and relative abundance of SWD based on current parameter estimates (a) in California, and (b) in the USA, and the cumulative mortality rates () at temperatures above and below 15C (c and d respectively). The subfigure plots average cumulative pupae on the coefficient of variation (CV).
Using the same model, prospective populations of SWD are simulated across the Mediterranean Basin and Europe (Fig. S8).
Fig S8. Simulation runs for the Mediterranean Basin are summarized (i.e., average pupae/ year, standard deviation (std) and coefficient of variability (CV(%)) and show areas of relative favorability (i.e., clear to red on the scale).
Summary of Data Used to Develop the Model (P. Gibertand C. Plantamp, unpublished)
TableS4.Development of SWD at different temperatures for populations from France and Spain. Strains were collected in Fall 2012 by M. Pascual in Spain and J. David in France, and kept using isofemale lines in the lab for about 8 generations before being used for experiments. For each population, 5 isofemale lines were used and reared at 5 constant developmental temperatures (10,15, 20,25 and30°C) with LD 12:12, 60% humidity. For each line, 50 eggs were put in a vial with standard Drosophila medium. Development time of adults was checked daily. Viability was calculated as the ratio between the number of adults and the number of eggs in each vial. Since the number of emerging adults was checked only once a day, at high temperature, most of the flies were already dead.We added results for two more temperature 27 and 29 °C for French populations only.
Table S5. Combined Dalton (D)–Gibert laboratory (P. Gibert and C. Plantamp, unpublished) –Kinjo (K) data on temperature dependent mortality. F= after freeze treatment (see eqn. 2i).
Author_stage / temp / mort/dayD_Adult / 1 / 0.167
D_Adult / 3 / 0.167
D_Adult / 5 / 0.029
D_Adult / 7 / 0.036
D_Adult / 10 / 0.009
D_F- Adult / 1 / 0.160
D_F- Adult / 3 / 0.160
D_F- Adult / 5 / 0.036
D_F- Adult / 7 / 0.036
D_F- Adult / 10 / 0.017
G _ Immature / 15 / 0.014
G _ Immature / 21 / 0.026
G _ Immature / 25 / 0.041
G_Immature / 10 / 0.014
G _Immature / 15 / 0.026
G _Immature / 20 / 0.026
G _Immature / 25 / 0.052
G _Immature / 27 / 0.040
G _Immature / 30 / 0.093
K_ Adult / 25 / 0.019
K_ Adult / 28 / 0.046
K_ Adult / 31 / 0.114
K_ Adult / 33 / ?
References cited in the Electronic Supplementary Material
Bieri M, Baumgärtner J, Bianchi G, Delucchi V, Von Arx R (1983) Development and fecundity of pea aphid (Acyrthosiphon pisum Harris) as affected by constant temperatures and pea varieties. Mitt Schweiz Ent Ges 56:163-171
Briére JF, Pracros P, Le Roux AY, Pierre JS (1999) A novel rate model of temperature-dependent development for arthropods.Environ Entomol28:22-29
Chabert S, Allemand R, Poyet M, Eslin P, GibertP (2012) Ability of European parasitoids (Hymenoptera) to control a new invasive Asiatic pest , Drosophila suzukii.Biol Control, 63:40–47
Chabert S, Allemand R, Poyet M, Ris N, Gibert P (2013) Drosophila suzukii, vers une lutte biologique contre ce ravageur des fruits rouges. Phytoma 660:34-38
Dalton DT, Walton VM, Shearer PW, Walsh DB, Caprile J, Isaacs R (2011) Laboratory survival of Drosophila suzukii under simulated winter conditions of the Pacific Northwest and seasonal field trapping in five primary regions of small and stone fruit production in the United States. Pest Manag Sci 67:368–1374
David JR, Clavel MF (1965) Interaction entre le génotype et le milieu d’élevage. Conséquences sur les caractéristiques du développement de la Drosophile. Bull Biol Fr Belg99:369–378
Di Cola G, Gilioli G, Baumgärtner J (1999) Mathematical models for age-structured population dynamics, pp. 503-534. In C. B. Huffaker & A. P. Gutierrez [eds.], Ecological entomology, Second ed. John Wiley and Sons, New York
Gutierrez AP (1996) Applied population ecology: A supply-demand approach. John Wiley and Sons, Inc.: New York, New York, USA; Chichester, England, UK
Gutierrez AP, Baumgärtner JU (1984) Multitrophic level models of predator-prey energetics: II. A realistic model of plant-herbivore- parasitoid-predator interactions. Can Entomol 116:933-949
Gutierrez AP, Ponti L(2013) Eradication of invasive species: why the biology matters. Environ Entomol 42:395-411
Izquierdo JI (1991)How does Drosophila melanogaster overwinter? Entomol Exp Appl59:51-58
Kinjo H, Kunimi Y, Nakai M (2014) Effects of temperature on the reproduction and development of Drosophila suzukii (Diptera: Drosophilidae). Appl Entomol Zool 49:297-304
Manetsch, TJ(1976) Time-varying distributed delays and their use in aggregate models of large systems. IEEE Trans Syst Man Cyb 6:547-553
Tochen S,Dalton DT, Wiman N, Hamm C, Shearer PW, Walton VM (2014)Temperature-related development and population parameters for Drosophila suzukii (Diptera: Drosophilidae) on cherry and blueberry. Environ Entomol 43:501-510
Vansickle J (1977) Attrition in distributed delay models,IEEE Transactions on Systems, Man, and Cybernetics 7:635-638
APPENDIX – Physiologically based demographic models
(PBDM)
Review of approaches for estimating the distribution and impact of weather and climate change on biological systems
The recent literature speaks clearly to the urgent need for approaches to estimate the geographic distribution and relative abundance of species that experience novel climates due to range expansion or climate change. Specifically, Section 4.3 (“Assumptions about future trends”) by Working Group 2 in the fourth assessment report (AR4) of IPCC outlines the shortcomings of widely used standard methods based largely on the climate envelope approaches (i.e., ecological niche models, ENMs) used to assess the impact of climate change on ecosystem. Among the gaps identified in IPCC AR4 were: the “inability to account for species interactions, the lack of physiological mechanisms, and the inability to account for population processes”. Including multi-trophic interactions in assessments of climate effects on biological systems has been an ongoing major challenge. Species interactions may constrain the geographic range of species even on an evolutionary time scale, and play a key role in host-parasite relationships in general.
Physiologically-based demographic system models (PBDM) explicitly capture the mechanistic weather-driven biology and dynamics of species at all trophic levels to predict the weather driven phenology, dynamics, and distribution of species across wide geographic areas on a daily basis – a time step rarely used in macro-ecological modeling(see Gutierrez and Baumgärtner 1984; Gutierrez 1996; Gutierrez and Ponti 2013). The model captures, via sub-models, the processes of resource acquisition and allocation, as well asbirth-death rates. PBDMs are sufficiently detailed to be realistic, and yet complexity is kept to a minimum by applying the same dynamics model and process sub models to all trophic levels. The complexity enters the model at the conceptual level, and running the model requires minimal computational capacity. These models have contributed to basic theory, and have helped solve many applied field problems, because they bridge the gap between purely theoretical analytic models and overly complicated simulation models. Physiological analogy across trophic levels is a powerful conceptual tool and is used as a way to tackle the huge challenges facing global ecosystem modeling.
The distributed maturation time demographic dynamics model
The biology of resource acquisition and allocation is embedded in a distributed maturation time demographic model(Manetsch 1976; Vansickle 1977) used to simulate the dynamics of age-mass structured populations, where time (t) and age (a) in the model are in physiological time units (e.g., proportional development or say degree days). Other dynamics model could also be used.