Introduction

Weather parameters are the dominant factors determining the distribution and abundance of most insect species. Oscinella frit, (Diptera: Chloropidae) is a stem-borer which can cause considerable damage to cereals that sown in spring. Weather parameters are dominant factors determining the distribution and abundance of many insect species including O. frit. Spring wheat in is susceptible to frit fly infestation during a limited period at early growth stages in central Germany. Fluctuations in the population of a pest during the year produce concomitant variations in damage. Life cycle of frit fly is ca. 20–60 days, eggs are laid one or in groups (1–15 eggs) behind sheath of first leaf, fecundity reaches 25–30 eggs, and embryonic development lasts (4–10 days) Lindblad and Sigvald (1996). Larvae hatch (three instars) and feed by tunnelling in stems to crown near soil surface. The larvae of older instars feed on tissues of growing- points. The larvae period lasts 13–20 days (Volkmar and Wetzel 1984; Lindblad and Sigvald 1996). Tillers attacked by frit fly show the classic ‘dead heart’ symptoms of a yellow central leaf which can be easily pulled away from the surrounding green leaves. Frit fly maggots attack the young plants (1–4 leaves) this leads to plant loss and in severe attacks whole crops have had to be re-drilled where control measures have not been used. The maggots are able to survive in the decaying plant debris and can migrate to feed on the newly sown crop (Lindblad 1999).

The insect opportunities to exploit plants during these periods are affected by weather conditions in several ways (Kingsolver 1989). The synchronization between insect and plant phenologies varies due to weather variation. Furthermore, weather conditions during the susceptible period of the host plant directly influence infestation rates, by e.g. affecting the insect oviposition rate or mortality (Bentley and Clements 1989; El-Wakeil et al. 2009).

Coloured water traps are often used to sample flying insects. Insects are attracted visually by the colour of the traps and are then captured in the water in them. Studies have demonstrated the preferences of a variety of insect groups for particular trap colours (Finch 1992; Barker et al. 1997). Previous studies using coloured traps to collect adult frit flies have looked at colour preferences and support the use of blue traps Lindblad and Sigvald (1996).

The objectives of this work were to assess the role of some weather conditions (temperature, rainfall and solar radiation) on frit fly population in spring wheat fields, its infestation levels and the response of frit fly catches to different coloured water traps.

Materials and Methods

Spring Wheat Plots

The experiments were carried out sandy loam soil in research station (Lehr-und Versuchsstation, (LVS)) in Martin Luther University- Halle, Germany. An Egyptian wheat variety (Sakha 93) known to be drought resistant (Ibrahim et al. 2011), and high quality German variety (Triso) were chosen for these experiments, which were cultivated in the spring. The experimental area was designated as Randomized Complete Block, divided into four blocks (plot = 1.5 ´ 3 m); 10 replicates were conducted. Weather condition parameters were (19°C, 0.9 mm and 1913 J/cm2), (14.0°C, 0.9 mm and 1727 J/cm2) and (12.0°C, 3.1 mm and 1212 J/cm2) for mean daily temperature, rainfall and solar global radiation, respectively (personal communication Dr. Döring) as shown in Table 1.

Table 1 Mean of daily temperature, rainfall and solar global radiation during 2008, 2009 and 2010 in spring wheat- in LVS, Halle
Full size table

Water Traps

Three colour water traps (blue, white and yellow) were used to sample adult flies in spring wheat plots. The traps consisted of metallic dishes; 25 cm diameter and 6.5 cm deep fixed in the ground and were partly filled with water with a small quantity of detergent as described by Lindblad and Sigvald (1996). The traps were observed two times weekly and trapped flies were counted under binocular in the laboratory.

Frit Fly Infestation

Infestation assessment was conducted at different wheat growth stages (GS) GS 10, 12, 13, 17, 19, 21, 22, 27 and 30 based on Meier (2001), ca. 50 plants/variety. All sampled plants were examined visually for symptoms of frit fly infestation (yellow or missing central leaf) during the studied years 2008, 2009 and 2010.

Statistical Analyses

Infestation percents data was arcsine square-root transformed. The difference among treatments was analyzed by Generalized Linear Model ((GLM), ANOVA-repeated measures) using Statistix 9 program (Thomas and Maurice 2008) and following with Tukey test to compare means of varieties. Significant differences were observed at P < 0.05 for all trials.

Results

Water Traps

In general, there were significant differences (P = 0.001) among years in trap catches. Weather condition parameters significantly affected trap catches during 3 years.

A. 2008

There were two peaks on 6th and 13th May 2008 (GS 17 and 21) in frit flies number in the water traps. On 6th May, the caught flies were 32, 24 and 16 flies in blue, white and yellow traps, respectively, while on 13th May, the caught flies were 29, 58 and 36 flies in blue, white and yellow traps, respectively. The blue traps caught the highest flies (32) on 6th May, while the white traps caught the highest flies (58) on 13th May. The numbers then declined until end of May. There were significant differences (P = 0.0312) among total caught of frit flies, which were 128, 173 and 140 flies in different traps (blue, white and yellow), respectively (Fig. 1a). These numbers coincided with the wheat seedlings being at a suitable growth stage for oviposition and correlated with weather condition parameters (temperature, rainfall and solar global radiation/cm2) (Fig. 2a–c).

Fig. 1
figure 1

Number of frit fly catches by different colour water traps and their relation to global radiation (J/cm2) in spring wheat (a) 2008, (b) 2009 and (c) 2010

Full size image
Fig. 2
figure 2

Relations between weather parameters (a) Temperature, (b) rainfall & (c) solar radiation and frit fly trap catches during 2008–2010

Full size image

B. 2009

Trap catch were recorded in two peaks on 4th and 11th May 2009 (GS 17 and 21). On 4th May, the caught flies reached to 56, 49 and 34 flies/trap in blue, white and yellow traps, respectively; while on 11th May, the fly numbers reached to 18, 39 and 16 flies/ trap in blue, white and yellow traps, respectively. The highest flies were recorded in blue traps (56 flies/trap) on 4th May, while in the white traps reached to 39 flies/trap on 11th May. Then, those numbers declined daily until end of May (Fig. 1b). Total frit fly catches were 221, 229 and 153 flies in different traps (blue, white and yellow), respectively; which were differed significantly (P = 0.0381) among trap colours. The peaks correlated with weather condition parameters (temperature, rainfall and solar global radiation/cm2) (Fig. 2a–c), and also coincided with growth stages of wheat seedlings for oviposition.

C. 2010

The highest trap catches in the different water traps were recorded on 10th May 2010 (GS 17). The frit fly numbers reached to 30, 103 and 95 flies in blue, white and yellow traps, respectively. These catch numbers declined gradually until end of May. Total trap catches were significantly differed (P = 0.0242) among trap colours, the highest numbers were recorded in white traps (398 flies) followed by yellow traps (356 flies) and lastly came blue traps (117) (Fig. 1c). The fly numbers were correlated with suitability wheat seedlings for oviposition and also with mean of solar global radiation, temperature and rainfall (Fig. 2a–c).

Infestation Percents of Frit Fly

Generally, analyses of variance of infestation percents data showed significantly differences between wheat varieties (P = 0.0001) as well among the studied years (P = 0.001). Season effects may be due to the different weather conditions among years.

A. 2008

Mean of infestation percentages in Sakha 93 and Triso varieties were 2.7 and 4.1% at GS 10 respectively. These percents reached to 13.1, 28.5, 58.5 and 63.0% at GS 12, 13, 17 and GS 19 in Sakha 93 variety. While in Triso variety, these percents reached to 25.3, 54.0, 74.8 and 79.0% at GS 12, 13, 17 and GS 19, respectively (Fig. 3a). The analyses of data showed that there were significant differences between wheat varieties (P = 0.0027); the infestation percents in Triso variety were higher than Sakha 93 variety. These percents were reduced to 56.7, 41.2, 36.7 and 20.0% at GS 21, 22, 27 and GS 30 in Sakha 93 variety. While in Triso variety, these percents were 70.0, 56.3, 46.7 and 26.7% at GS 21, 22, 27 and GS 30, respectively (Fig. 3a).

Fig. 3
figure 3

Mean ± SE of infestation percents of frit fly at different growth stages of two spring wheat varieties during three seasons (a) 2008, (b) 2009 and (c) 2010. Different letters indicate significant differences

Full size image

B. 2009

The analyses of data showed that there were significant differences between wheat varieties (P = 0.0032). Mean infestation percents in Triso and Sakha 93 varieties were 1.5 and 2.4% at GS 10, respectively. These percents reached to 10.1, 34.0, 41.0 and 48.2% at GS 12, 13, 17 and GS 19 in Sakha 93 variety. While in Triso variety, these percents reached to 9.8, 15.5, 24.1 and 40.5% at GS 12, 13, 17 and GS 19, respectively (Fig. 3b). These percents were reduced to 63.8, 52.4, 42.4 and 32.1% at GS 21, 22, 27 and GS 30 in Sakha 93 variety. While in Triso variety, these percents were 52.3, 44.2, 36.0 and 27.3% at GS 21, 22, 27 and GS 30, respectively (Fig. 3b). This means that infestation percents in Sakha 93 variety were higher than Triso variety.

C. 2010

The infestation percents were significantly differed (P = 0.0038) between varieties; Sakha 93 had more infested tillers than Triso (Fig. 3c). Mean of infestation percentages in Triso and Sakha 93 varieties were 1.3 and 2.2% at GS 10, respectively. These percents reached to 13.7, 19.9, 33.2 and 47.9% at GS 12, 13, 17 and GS 19 in Sakha 93 variety. While in Triso variety, these percents reached to 10.3, 15.1, 26.6 and 32.7% at GS 12, 13, 17 and GS 19, respectively (Fig. 3c). These percents were reduced to 43.1, 41.8, 35.2 and 30.4% at GS 21, 22, 27 and GS 30 in Sakha 93 variety. While in Triso variety, these percents were 29.8, 25.9, 22.7 and 20.5% at GS 21, 22, 27 and GS 30, respectively (Fig. 3c).

Discussion

There were significant differences among trap catches in the three season-studies. The flies caught in water traps were recorded in two peaks in 2008 and 2009; while one peak was recorded in 2010. The results are consistent with previous findings by El-Wakeil et al. (2009). The high numbers were related to weather conditions as reported by Lindblad and Sigvald (1999), who stated that trap catches and infestation rates of frit fly in oats plants were related to daily weather conditions as well as plant growth stages. Similarly, Menéndez (2007) mentioned that climate changes affect biological and ecological systems of insects; as well Rygg (1966) and Barker et al. (1997) indicated that majority of flies caught in water traps were caught in early of May which had developed eggs. Trap catches were significantly differed among traps (blue, white and yellow) and also frit fly catches were correlated with solar global radiation. There was a positive correlation between blue trap catches and solar radiation and negative correlation with white and yellow traps in different studied years.

Our results show in agreement with Cuncliffe et al. (1925), that wheat seedlings were most susceptible to frit fly infestation during growth stages GS 11–13. Thereafter, the susceptibility rapidly decreased. Seedlings in GS 17 were considerably less susceptible than in earlier stages. However, we found that up to 30% of the main shoots may be infested in this growth stage if the infestation pressure is high. Hence, the risk of frit fly damage cannot be neglected until the spring wheat has reached GS 19–21. Spring wheat seedlings tended to be less infested in GS 10 than in the following stages (Fig. 3). This finding supports the hypothesis that very young plants are less attractive to the egg-laying female as confirmed by Jonasson (1977), but may also be due to a shorter exposure time.

Temperature and solar global radiation affected trap catches and infestation rates, whereas rainfall was less important. Few plants were infested during days with maximum temperatures lower than 12–13°C (Tab. 1 and Fig. 3). Our results are comparable with previous findings from studies of damage on autumn-sown cereals; where Riggert (1935) stated that egg-laying ceases at maximum temperatures below 15°C and that wind and rain normally do not affect infestation rate. Since spring wheat seedlings mainly are susceptible to infestation during GS 11–13, daily temperatures during these growth stages will highly influence the risk of frit fly infestation. The relationship between the development rate of spring wheat and temperature is approximately linear as mentioned by Saarikko and Carter (1996); Lindblad and Solbreck (1998) and the duration of these growth stages depends on weather conditions.

Triso had more infested tillers than Sakha 93 in 2008 and vice versa in 2009 and 2010 (infestation was in Sakha 93 higher than Triso variety in 2009 and 2010). This may be due to one variety grows faster than the other variety with respect to different weather conditions. Generally, infestation percents were higher in season 2008 and 2009 than 2010; this may be due to weather condition parameters which affected on frit fly activities (hibernation- frit fly emergence, egg oviposition and larval activity) as declared by Lindblad (1999). Frit fly adults has a good chance to lay their egg till GS 13, commonly occurs in early of season as confirmed by Southwood et al. (1961) and Tolley and Niemzcyk (1988); while plants are still young. Larval activity stopped at GS 19–21 then declined till GS 31–33.

Early spring wheat will escape in most cases from frit fly infestation even if the weather is warm during the most susceptible growth stages, because the flies have not yet emerged. The time of frit fly emergence can be predicted with relatively high accuracy, although forecasting errors may arise due to e.g. choice of weather station within a region as reported by Jonasson (1977); Clements et al. (1990); Lindblad and Sigvald (1996, 1999). However, our study shows that although significant damage of seedlings in GS 17–19 is uncommon, it cannot be totally excluded. In years with high abundance of frit flies and warm weather, early-sown crops may suffer unexpected damage due to infestation in this growth stage. Hence, to explain the incidence of frit fly damage, it is necessary to consider not only the timing between insect and crop phenologies, but also the influence of weather conditions at trap catches, oviposition, larval infestation and variation in population size between years. In conclusion, there were two stages; firstly, frit fly-overwinter-emergence which affected by rainfall and solar radiation and secondly, fly infestation that affected by temperature. Climate is the dominant factor determining the distribution and abundance of most insect species.