Supplemental Material: Analysis of the invasiveness of spotted wing Drosophila (Drosophila suzukii) in North America, Europe, and the Mediterranean Basin

Andrew Paul Gutierrez1,2,3, Luigi Ponti1,4, Daniel T. Dalton5

1 Center for the Analysis of Sustainable Agricultural Systems (CASAS Global), 37 Arlington Ave., Kensington, CA, USA 94707

2 Division of Ecosystem Science, College of Natural Resources, University of California, Berkeley, CA, USA 94720-3114.

3 Corresponding author, e-mail:

4 Agenzia nazionale per le nuove tecnologie, l’energia e lo sviluppo economico sostenibile (ENEA), Centro Ricerche Casaccia, Via Anguillarese 301, 00123 Roma, Italy.

5 Department of Horticulture, Oregon State University, 4017 Ag and Life Sciences Bldg., Corvallis, OR 97331, USA

The supplemental materials contain the full multiple regression analysis of yearly summary simulation data for the USA and Mexico during the period 1980-2010 (Table S1), and summarizes the available data on the effects of temperature on mortality of summer and winter morph D. suzukii adults and the simulations across the USA and Mexico, and Europe and the Mediterranean Basin. Also included are the daily dynamics of reproduction by the summer and winter adult morphs at Benton WA.

The mortality rate for reproductive adults is the black line in Fig. S1, and is the same as in text Fig. 2a (eqn. 2i), while the mortality rate of winter morph adults is the red line (and red symbols). Note that as winter morph adults become reproductive, the mortality rate eqn. 2i is applied. The simulated effects of enhanced winter survival of winter morph adults on the geographic distribution and abundance of the fly in the USA and Mexico are illustrated in Fig. S2, while the results for Europe and the Mediterranean Basin are shown in Fig. S3. Average cumulative pupa density during 1990-2010 is the metric of favorability in these maps. To illustrate the population level effects of reduced winter morph mortality, the simulated daily dynamics of the egg stage and of the summer and winter adults at Benton, WA are compared in (Fig. S4a vs S4b). As expected winter survival of adult flies increases but the relatively high thermal threshold of ~12.75°C for reproduction delays reproduction in spring and induces development of winter morph adults early in fall. Thus despite enhanced winter survival, densities do not increase dramatically at Benton and likely in more northern areas because the season length remains relatively unchanged and D. suzukii has a low reproductive capacity. Hence, while the abundance increases modestly, the basic geographic distribution does not change dramatically (Fig. S2, S3).

Fig. S1. Mortality rate on temperature: the dark and patterned symbols and the black line are from Fig. 2a, while the red symbols are winter morph mortality rates estimated from Shearer et al. (2016) and Toxopeus et al. (2016) and the red line is a heuristic function that shifts eqn. 2i (see text) leftward 5.5°C for temperatures below the oviposition threshold of 12.75°C.

Fig. S2. Effects of winter morph mortality rates (see Fig. S1) on the geographic distribution of D. suzukii (pupae) in USA and Mexico using 1990-2010 weather data below 2,500 m elevation: (a) average (AVG), (b) coefficient of variation as a percent (CV%), (c) standard deviation (STD), (d) AVG clipped at 3000 pupae, (e) STD for the USA, and (f-h) AVG, CV% and STD for California.

Fig. S3. Effects of winter morph mortality rate on the geographic distribution of D. suzukii (pupae) in Europe and the Mediterranean Basin below 2000m elevation: (a) average density color clipped at 1300 (AVG) for ease of comparison to text Fig. 6, (b) frequency distribution of average values, (c) coefficient of variation as a percentage (CV%) and (d) standard deviation (STD).

Fig. S4. The simulated effects of temperature on egg, reproductive adults and winter morph adult dynamics during 2000 to 2010 at Benton, WA: (a) mortality using eqn. 2i in the text, and (b) shifting eqn. 2i leftward 5.5°C to simulated the enhance survival of winter morph adults. The horizontal dashed line is to aid in the comparison.

Table S1. Multiple regression analysis of yearly summary simulation data for the USA during the period 1980-2010

independent variable / mean / standard deviation / correlation x vs y / regression coefficient / std, error of reg. coef. / computed t - value
annual rain (mm) / 738.9 / 469.8 / 0.612 / 0.821 / 0.1915E-02 / 428.55
m>16°C / 6.484 / 563.4 / 0.382 / -35.44 / 0.23403 / -151.44
m<16°C / 31.84 / 33.34 / -0.436 / 2.083 / 0.4411E-01 / 47.236
dd>5.975 / 3452. / 1792 / 0.712 / 0.409 / 0.8166E-03 / 5000.53
m16°C ´ m<16°C / 89.97 / 76.16 / -0.304 / 0.4212E-01 / 0.1593E-01 / 2.645
dependent variable
pupae/year / 871.7 / 904.03

y-intercept -986.03

multiple correlation 0.8473

r square 0.7179

std. error of estimate 480.59

ANALYSIS OF VARIANCE FOR THE REGRESSION

source of variation degrees of sum of mean f value

Freedom squares squares

attributable to regression 5 0.186256E+12 0.372512E+11 161,284

deviation from regression 316,854 0.731825E+11 230966

total 316,859 0.259439E+12

1