BioControl DOI 10.1007/s10526-016-9783-7
Alyssum flowers promote biological control of collard pests Andre´ L. Ribeiro . Lessando M. Gontijo
Received: 30 August 2016 / Accepted: 28 December 2016 Ó International Organization for Biological Control (IOBC) 2017
Abstract Collard greens Brassica oleracea (L.) are often attacked by various pests including whiteflies, aphids and diamondback moth. Hitherto, the main method used to manage these pests in Brazil has been the application of a limited number of registered insecticides. The search for more sustainable pest management strategies is therefore warranted. In this context, the conservation biological control stands out as an appealing alternative. Conservation biological control is achieved, at least in part, by strip-cultivating and/or conserving flowering plants within the agroecosystem. The present study investigates how alyssum flowers Lobularia maritima (L.) could contribute to the attraction of natural enemies and to the management of collard pests. Two field experiments were conducted in different years. Each experiment consisted of two treatments and three replicates, which
were set up in a completely randomized design. The treatments were (1) collards alone, and (2) collards ? alyssum. We evaluated weekly the population density of natural enemies and pests on both treatments. The results show that the alyssum flowers attractiveness contributed to increase the abundance of generalist predators during both experiments, which in turn translated into a significant reduction of collards pests, especially aphids. Some of the main predators attracted/harbored by alyssum flowers were spiders, coccinellids, syrphids and Orius sp. Finally, strip intercropping alyssum with collards can be an important strategy to manage brassica pests and cope with the limited availability of insecticides registered for this vegetable crop. Keywords Conservation biological control Lobularia maritima Brassicas Intercropping
Handling Editor: Marta Montserrat.
Electronic supplementary material The online version of this article (doi:10.1007/s10526-016-9783-7) contains supplementary material, which is available to authorized users. A. L. Ribeiro L. M. Gontijo Instituto Federal Goiano, Campus Morrinhos, Morrinhos, GO 75650-000, Brazil L. M. Gontijo (&) Instituto de Cieˆncias Agra´rias, Universidade Federal de Vic¸osa – Campus Florestal, Rodovia LMG 818, km 06, Florestal, MG 35690-000, Brazil e-mail:
[email protected]
Introduction The provision of alternative food (i.e., pollen and nectar) to natural enemies has been widely accepted as means of enhancing biological control of herbivorous mites and insects (Hickman and Wratten 1996; Grafton-Cardwell et al. 1999; Gurr et al. 2004; Berndt et al. 2006; Ponti et al. 2007). This provision may be attained by ‘spraying’ complementary food on or near the target crop (Wade et al. 2008), by selective
123
A. L. Ribeiro, L. M. Gontijo
conservation of non-crop flowering vegetation (Amaral et al. 2013), and/or by cultivating strips of insectary plants along some rows of a crop (Patt et al. 1997; Gontijo et al. 2013). In addition to providing alternative food, some insectary plants can also provide natural enemies with shelter, which in turn helps to protect them from adverse environmental conditions, pesticides and intraguild predators (Tscharntke et al. 2005; Perdikis et al. 2011; Tixier et al. 2013). This conservation biological control approach is particularly important for highly disrupted agroecosystems, which seldom have the necessary resources to guarantee the attraction and establishment of natural enemies (Landis et al. 2000; Gardiner et al. 2009). In addition, the availability of alternative food could also be important in times of prey/host scarcity, which often leads natural enemies to leave the crop area (Rand et al. 2006). For example, Limburg and Rosenheim (2001) found out that cotton plants extrafloral nectar helps biological control by maintaining lacewing larvae Chrysoperla plorabunda (Fitch) in the field until aphid populations start increasing again. Although the provision of floral resources in general has shown to be important for the local conservation of natural enemies (Landis et al. 2000; Fiedler et al. 2008), there are certain plant attributes that need to be taken into consideration when choosing the most appropriate species of insectary plant. The best candidate should provide novel resource not already available within or nearby the agroecosystem, easily accessible to natural enemies and not advantageous to pests. In addition, from an agronomic standpoint, insectary plants must be able to grow and develop under normal crop management practices, not compete with crops for nutrients, light or water (Bone et al. 2009); and to reduce management costs, insectary plants should preferably be perennial, or if annual, be able to reseed itself. Alyssum Lobularia maritima (L.) is a flowering plant species that has been used quite often as insectary plant in the United States (Hogg et al. 2011; Gontijo et al. 2013; Brennan 2013, 2016). By contrast, little is known about the use of this plant species for conservation biological control in the tropics (i.e., Brazil). Additionally, all the work done with alyssum up to date has investigated its impact only on single target pests, thus leaving a question whether it could also help with the suppression of
123
multiple pests. Alyssum forms a low-growing mat of plants with a high flower density, attracts several generalist and specialist natural enemies and very few pests; and although it can re-seed itself it is unlikely to become a weed (Chaney 1998; Gontijo 2011). Thus, the cultivation of strips of alyssum could be an obvious way to enhance biological control of multiple pests, especially in organic farms where the insecticide options are few. Collard greens Brassica oleracea (L.) cultivar ‘manteiga’ is one of the most common cruciferous plants cultivated in Brazil (Novo et al. 2010), especially in small farms. However, this vegetable plant is attacked throughout the year by various insect pests, including aphids (e.g., Brevicoryne brassicae L.), diamondback moth Plutella xylostella (L.), brassica caterpillar Ascia monuste orseis (Godart) and whiteflies Bemisia tabaci (Gennadius), which cause direct damage to the marketable leaves. The frequent use of a limited number of registered insecticides to control these pests coupled with a recent increase of the organic vegetable market in Brazil (Oelofse et al. 2010) have warranted more research on devising sustainable pest management strategies. The biological control of the aforementioned collard pests can be attained by the attack of various natural enemies such as syrphids, minute pirate bugs, coccinellids, carabids, earwigs and parasitoids (Tukahirwa and Coaker 1982; White et al. 1995; Naranjo 2001; Oliveira et al. 2003; Furlong et al. 2004; Nieto et al. 2006; Firake et al. 2012). Thus, the conservation of such group of natural enemies is of paramount importance to promote pest biological control in the collard system. In the present study we examined whether or not the planting of alyssum strips could indirectly enhance the biological control of pests attacking collard greens. More specifically we assessed during two growing seasons the population density of aphids (B. brassicae and Myzus persicae (Sulzer)), P. xylostella larvae, B. tabaci, and natural enemies on collard greens cultivated alone or near strips of alyssum.
Materials and methods Two experiments were conducted in the field (17°480 50.400 S, 49°120 16.500 W) at the Instituto Federal Goiano, Campus Morrinhos in the state of Goia´s, Brazil at two different summer time periods: (1) from
Alyssum flowers promote biological control of collard pests
October 2014 to February 2015, and (2) from October 2015 to January 2016. These time periods are characterized by high daily temperatures (average 25–32 °C) and significant precipitation (Supplementary Table 1). Both experiments consisted of two treatments: (1) collards alone (monoculture), and (2) collards ? alyssum flowers (polyculture). There were three replicates per treatment during each experiment, which were arranged in a completely randomized design. Each replicate was represented by a raised soil bed (4.0 9 5.0 9 0.30 m; W 9 L 9 H) containing four rows of seven collard plants (cultivar ‘manteiga’) spaced apart by 100 cm between rows and 50 cm between plants (within rows). This yielded a total of 28 plants per replicate. The soil beds/replicates were spaced 20 9 20 m away from each other, which was considered the buffer zone. The collard seedlings were purchased from a commercial supplier (Viveiro Beira Mato, Morrinhos City, Goia´s, Brazil) and transplanted into the beds when they were 20 days old (15 cm of height approximately). Each soil bed received before the transplanting 2 kg/m-2 of cattle manure, and 350 g/m-2 of chicken manure at 15, 30 and 45 days after transplanting. The seedling transplant took place on October 11th, 2014 and on October 8th, 2015 for the first and second experiment, respectively. Some of the seedlings did not survive the first ten days, and were replaced immediately by others of the same age and size. Thereafter, approximately 100–150 g of alyssum seeds was sowed manually between the two middle rows of collards (strip 90 9 500 cm) within the beds of its respective treatment (collards ? alyssum). Water was provided daily (2 mm) in each replicate by a dripping irrigation system running along the collard rows. The control of weeds was conducted manually each week. The collard plants were naturally infested by insect pests occurring in the area. Nonetheless, all replicates had equal chances of being infested by ambient insect pests prior to alyssum blossom. Additionally, during the running of both experiments, data regarding temperature, humidity and rain precipitation were collected by a field weather station located nearby the experimental area (Supplementary Table 1). To assess the population density of insect pests and natural enemies in both experiments, weekly samples were taken in the morning (8:00–11:00 a.m.) on each replicate during the course of seven weeks using the following methods: (1) direct count on collard plants,
(2) beating tray on collard plants and alyssum flowers, (3) cardboard bands for earwigs, and (4) pitfall traps. The order of the data collection in each date was alternated between treatments to avoid any sampling bias. The samplings started 60 days after seedling transplanting in both experiments. The collard plants had an approximate height of 40 cm by the time of the first evaluation. Direct count (1): four random collard plants in each replicate were approached with care and one leaf per plant in the middle of the canopy was inspected visually for insect pests (adults of whitefly, larvae and adults of diamondback moth, and aphids) and natural enemies, using a hand lens when necessary. Beating tray (2): following the direct count, the canopy of four other random different collard plants were separately and gently tapped into a white plastic tray (35 9 25 cm) and insect pests and natural enemies that fell were counted. The bottom of the tray was dampened prior to each tapping to avoid the fast scape of flying specimens. This same process was also carried out on the alyssum flowers where two locations within each replicate were tapped for natural enemies and insect pests. Cardboard bands (3): earwigs were sampled on focal plants for 24 h per week using a 10 cm roll of cardboard tied to the main stem of the collard plant 10 cm above the soil level, as these opportunistic predators are not effectively sample using other means (Horton et al. 2002). Pitfall traps: these traps consisted of a plastic cup (10 9 15 cm; D 9 H) filled with 200–300 ml of water containing some drops of detergent to break the superficial tension. One trap was buried in the soil to the surface level in each replicate during 48 h, and thereafter taken to the laboratory for identifying and counting the collected predators. This sampling method was conducted at four dates evenly spaced in time during the conduction of each experiment. The number of adult parasitoids and parasitized insect pests was assessed only in the second experiment using the methodology described above (direct count for adults and mummies, and beating tray for adults also). Only aphid parasitoids were detected. The identification of the parasitoid species was done by taking a sample of 40 mummified aphids from both treatments into the laboratory and waiting for adult emergence. Thereafter, the adult parasitoids were killed in alcohol 70% and later mounted for identification using dichotomous keys.
123
A. L. Ribeiro, L. M. Gontijo
Statistical analysis
Results
The numbers of insects sampled within each replicate were summed by type (aphids, whiteflies, diamondback moth larvae, generalist predators, adult parasitoids and mummified aphids) and date. This sum comprises all the data from all types of sampling methods, except the beating tray on alyssum flowers to avoid inflating the number of natural enemies counted on that treatment (polyculture). Thus, since the type and number of samplings were standardized across treatments and dates each sum referred above represents one measure of pests or natural enemy density per experimental unit (three per treatment). Differences in population densities of insect pests (aphids, whiteflies, diamondback moth larvae) and natural enemies (not including those sampled on alyssum flowers) between treatments were assessed using repeated measures analyses of variance (PROC MIXED) for each experiment’s data. Covariance structures for the mixed model repeated measures were constructed, and Baysian Information Criterion (BIC) was used to assess which was the best model. For the first experiment the best models that fit the data were TOEPH (Heterogeneous Toeplitz) for aphid density, and ARH(1) (First-order Autoregressive Heterogeneous) for whitefly, diamondback moth larvae and predators density. By contrast, the model ARH(1) fit best all the data from second experiment regarding population density of whitefly, diamondback moth larvae, predators and parasitoids. Because we used the Satterthwaite df calculation method, which is an estimation based on the estimated variance–covariance G and R matrices (G for random effects, R for repeated measures), decimal values for the denominator df were obtained. Because the data from the second experiment regarding aphid density did not converge we could not carry out a repeated measures analysis of variance. Thus a regular ANOVA (PROC GLM) was conducted to assess the effects of treatments on the aphid population density. In this analysis dates were considered as blocks. All the analyses aforementioned tested the statistical effects of treatment, time and the interaction between them. All statistical analyses were conducted using the SAS program.
There was a significant effect of treatment, time, and treatment 9 time interaction on the population density of aphids on both experiments (Table 1). Despite the population fluctuation, it was noticed a significant higher aphid density on the collards-alone treatment during three of the evaluation dates on both experiments (Fig. 1a, b). Both Brevicoryne brassicae and Myzus persicae were found feeding on collard plants, and, although we did not try to measure their population size separately, a larger population of B. brassicae was perceived. There was also a significant effect of treatment and time, but no interaction, on the population density of adult whiteflies attacking collard plants (Table 1). The adult whiteflies tended to reach higher densities during the months of January and February of experiments 1 and 2, respectively, and regardless of treatment (Fig. 2a, b). Nonetheless, by comparison, the numbers of adult whiteflies tended to be higher on the collards-alone treatment during most of the dates (Fig. 2a, b). By contrast, there was no significant effect of neither treatment nor time on the population density of diamondback moth larvae during experiment 1, but only an effect of treatment 9 time interaction (Table 1). On the other hand, the density of diamondback moth larvae was significantly affected by treatment and time, but not by their interaction, during experiment 2 (Table 1). The population density of diamondback moth larvae was significantly higher on the collards-alone treatment during most evaluation dates in experiment 2 (Fig. 3a, b). Ascia monuste orseis (Godart) is also an important pest of collards that we a priori planned to sample. However, their numbers were negligible in both experiments. While sample was done for natural enemies on the alyssum plants, we did not notice the presence of any target pest (aphids, whiteflies or diamond back moth). There was a significant effect of treatment, time and their interaction on the population density of predators in both experiments (Table 1). It was noticed a higher predator population density on the ‘collard ? alyssum’ treatment during both experiments (Fig. 4a, b), therefore suggesting a significant potential of these flowers to attract predators of collard pests. The predator taxa encountered more often on the ‘collard ? alyssum’ treatment during the first experiment were spiders, carabids, Orius sp., syrphids and earwigs
123
Alyssum flowers promote biological control of collard pests Table 1 Results for the repeated measures ANOVA (PROC MIXED, SAS) regarding the effects of treatment, time and treatment 9 time interaction on the population density of pests and natural enemies
Response
Effect
F
d.f.
P
Experiment 1 (2014–2015) Aphid population
Whitefly population
Diomandback moth larvae pop.
Predator population
Treatment
18.23
1, 4.5
0.0114
Time
29.94
6, 5.17
0.0008
Treatment 9 time
16.28
6, 5.17
0.0034
Treatment
55.05
1, 4.66
0.0009
Time
59.60
6, 4.75
0.0002
Treatment 9 time
14.96
6, 4.75
0.0057
Treatment Time
2.13 3.04
1, 7.76 6, 4.88
0.1838 0.1247
Treatment 9 time
5.14
6, 4.88
0.0485 \0.0001
140.40
1, 6.45
Time
Treatment
33.68
6, 5.41
0.0004
Treatment 9 time
14.64
6, 5.41
0.0036
Treatment
Experiment 2 (2015–2016) Aphid population
Whitefly population
19.33
1, 12
0.0001
Time
5.50
6, 12
0.0007
Treatment 9 time
2.60
6, 12
0.0400
Treatment
4.70
1, 4.69
0.0339
24.45
6, 6.90
0.0003
3.57
6, 6.90
0.0611
Treatment
36.24
1, 4.15
0.0035
Time
10.65
5, 5.55
0.0072
Treatment 9 time Treatment
4.20 16.71
5, 5.55 1, 5.87
0.0600 0.0068
Time Treatment 9 time Diomandback moth larvae pop.
Predator population
Adult parasitoid population
Fig. 1 Mean number of aphids (Brevicoryne brassicae and Myzus persicae) per replicate at different evaluation dates. a Experiment 1 (Dec 2014–Feb 2015); b experiment 2 (Dec
Time
23.61
6, 7.15
0.0002
Treatment 9 time
10.34
6, 7.15
0.0035
Treatment
6.50
1, 4.05
0.0631
Time
1.25
6, 9.01
0.3693
Treatment 9 time
0.73
6, 9.01
0.6363
2015–Jan 2016). Each mean is the result of averaging the number of aphids that was summed across the sampling methods of direct count and beating tray on collards in each replicate
123
A. L. Ribeiro, L. M. Gontijo
Fig. 2 Mean number of adult whiteflies (Bemisia tabaci) per replicate at different evaluation dates. a Experiment 1 (Dec 2014–Feb 2015); b experiment 2 (Dec 2015–Jan 2016). Each
mean is the result of averaging the number of adult whiteflies that was summed across the sampling methods of direct count and beating tray on collards in each replicate
Fig. 3 Mean number of diamondback moth larvae (Plutella xylostella) per replicate at different evaluation dates. a Experiment 1 (Dec 2014–Feb 2015); b experiment 2 (Dec 2015–Jan
2016). Each mean is the result of averaging the number of larvae that was summed across the sampling methods of direct count and beating tray on collards in each replicate
(Table 2). Likewise, the most common predators encountered on the ‘collard ? flowers’ treatment during the second experiment were spiders, coccinelids, syrphids and earwigs (Table 2). The parasitoid population density was assessed only in the second experiment. The only parasitoids encountered were aphid parasitoids. There was no significant effect of treatment, time nor treatment 9 time interaction on the population of adult parasitoids (Table 1). Nevertheless, it was noticed a tendency for greater increase of adult parasitoid numbers on the ‘collard ? alyssum’ treatment, especially more towards the last evaluation dates (Fig. 5a). The number of parasitized aphids (‘mummies’) was
similar in both treatments (Fig. 5b). All the parasitoids emerging from field aphid mummies in the laboratory were identified as Diaeretiella rapae (Hymenoptera: Braconidae). Among the insect sampling methods used, the direct count on the plant and beating tray were the methods that allowed recording the greatest numbers of diurnal insects (herbivores and natural enemies) (Table 3). By contrast, the pitfall and cardboard bands were the methods that allowed a better sampling of nocturnal natural enemies such as carabids and earwigs, respectively (Table 3). The beating tray sampling carried out on alyssum flowers resulted in high number of natural enemies visiting the flowers (Table 3).
123
Alyssum flowers promote biological control of collard pests
Fig. 4 Mean number of generalist predators per replicate at different evaluation dates. a Experiment 1 (Dec 2014–Feb 2015); b experiment 2 (Dec 2015–Jan 2016). Each mean is the result of averaging the number of predators that was summed
across the sampling methods of direct count and beating tray on collards, cardboard bands and pitfall traps, in each replicate. It does not include the predators sampled directly on alyssum flowers (beating tray)
Discussion
Rosenheim 2001). Although in our study alyssum flowers contributed to an increase in the population of generalist predators, the pest biological control was not negatively affected, suggesting absence or nonsignificant intraguild predation. In fact, Tixier et al. (2013) found that increasing plant diversity could enhance biological control and reduce the chances for intraguild predation by providing alternative food and shelter to natural enemies. The presence of multiple prey species (i.e., aphids, whiteflies, diamondback larvae) attacking collards may also have contributed to the improved biological control by securing temporal availability of alternative prey food to generalist natural enemies. It is known that generalists can switch attacks among different pest species as each becomes more or less abundant during the course of the season (Symondson et al. 2002). In addition, the concomitant availability of both non-prey food (i.e., pollen and nectar) and alternative prey may work synergistically to enhance biological control of insect pests (Schmidt et al. 2013; Ramsden et al. 2015). The relative low number of adult parasitoids and parasitized aphids was similar between treatments. This may have happened because of the relative short distance between replicates (20 m), allowing parasitoids to visit both treatments equally often. Nonetheless, while this distance between replicates (buffer) may not be long enough to isolate flying natural enemies in the different treatments, it was still adequate enough to identify important and insightful
The results show that strip intercropping alyssum flowers with collards contributed to an increase in natural enemy abundance, which likely mediated a significant suppression of multiple insect pests. These results were even more striking with the aphid population, which was maintained at low numbers on the ‘collards ? alyssum’ treatment during both experiments (Fig. 1a, b). Previous studies have also indicated that alyssum flowers can mediate aphid suppression by attracting and providing alternative food to natural enemies (Hogg et al. 2011; Gontijo et al. 2013). The natural enemy population in our studies consisted primarily of generalist predators, which may have been important to prevent collard insect pests from reaching high densities since the beginning of the experiments. Because of their polyphagous behavior, generalists natural enemies can often survive on non-prey food before any pest colonizes a crop, thus forming a ‘‘first line of defense’’ as insect pests arrive in the area (Harwood et al. 2007). This is particularly important at moments of low pest density, when specialist natural enemies usually show a weak numerical response (Hassel and May 1986). Because generalist predators usually have a broader diet breadth, they are more prone to involve in intraguild predation (Rosenheim et al. 1993, Traugott et al. 2012), especially upon specialist natural enemies such as parasitoids (Heimpel et al. 1997; Colfer and
123
A. L. Ribeiro, L. M. Gontijo Table 2 Natural enemies sampled at different dates Predators
Collards ? alyssum 13 Dec.
20 Dec.
27 Dec.
Collards alone 10 Jan.
31 Jan.
07 Feb.
14 Feb.
13 Dec.
20 Dec.
27 Dec.
10 Jan.
31 Jan.
07 Feb.
14 Feb.
Experiment 1 (2014–2015) Spiders
23
23
27
36
41
16
51
9
9
3
9
15
10
30
Carabidae beetles
0
0
53
28
23
10
16
0
0
16
20
16
5
11
Chrysopidae adults
5
3
2
3
5
0
3
0
1
0
0
0
0
0
Chrysopidae larvae
0
0
2
1
3
0
0
0
0
0
0
0
0
0
Coccinellidae adults
11
8
23
14
29
8
16
2
3
3
4
2
0
2
Coccinellidae larve
4
4
4
4
6
0
7
0
1
1
1
1
0
3
Orius sp.
11
11
14
16
14
9
14
0
2
0
0
0
0
0
Syrphidae adults
16
15
18
18
29
14
24
5
7
3
3
4
2
7
Syrphidae larvae
2
2
1
1
2
0
1
0
1
0
0
1
0
0
Earwigs
18
19
70
34
27
16
25
0
5
20
11
12
6
11
Total
90
85
214
155
179
73
157
16
29
46
48
51
23
64
Natural enemies
Collards ? alyssum
26 Jan.
09 Jan.
23 Jan.
30 Jan.
9
20
16
2
1
1
05 Dec.
12 Dec.
19 Dec.
22
24
Collards alone 26 Jan.
09 Jan.
23 Jan.
30 Jan.
9
23
29
6
4
2
05 Dec.
12 Dec.
19 Dec.
17
15
17
Experiment 2 (2015–2016) Spiders
37
Carabidae beetles
7
Chrysopidae adults
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Chrysopidae larvae
0
0
0
0
2
0
0
0
0
0
0
0
0
0
Coccinellidae adults
11
9
9
3
3
8
0
6
3
1
0
2
0
0
Coccinellidae larve
3
1
1
1
1
2
2
0
0
0
0
0
0
0 0
Orius sp.
.
3
.
.
3
.
1
.
.
3
2
1
1
1
0
2
0
0
0
0
0
0
Syrphidae adults
13
10
12
15
3
19
23
11
11
10
9
6
5
9
Syrphidae larvae
0
2
0
3
0
0
0
0
0
1
0
0
0
0
Earwigs
35
15
19
10
15
12
5
20
7
14
8
11
5
2
Adult parasitoids
27
38
23
25
19
29
40
9
9
9
13
6
11
12
136
99
92
67
71
101
78
66
45
53
39
46
38
25
Total
Each value represents the sum of natural enemies collected using the sampling methods of direct count, beating tray on collards (not on alyssum), cardboard bands and pitfall traps
differences between treatments (e.g., differences in pest and predator density). Additionally, the data from beating tray on alyssum flowers show that various
123
predators and adult parasitoids are in fact attracted to those flowers (Table 3). Although we did not access predation directly (e.g., gut content analysis) we have
Alyssum flowers promote biological control of collard pests
Fig. 5 Mean number of adult parasitoids (Diaeretiella rapae) (a) and mummified aphids (b) per replicate at different evaluation dates. Each mean is the result of averaging the
number of adult parasitoids or mummies that was summed across the sampling methods of direct count and beating tray on collards, in each replicate
a reasonable degree of confidence that those predators attracted by alyssum flowers could prey upon aphids, whiteflies and diamondback moth larvae, as it has been documented elsewhere (Tukahirwa and Coaker 1982; Naranjo 2001; Furlong et al. 2004; Nieto et al. 2006; Firake et al. 2012). In general, both the population of collard pests and predators tended to reach higher densities during the first experiment, regardless of treatment. This was probably due to differences in environmental variables. For example, there was higher volume of pluviometric precipitation during the second experiment (Supplementary Table 1). We also found a significant treatment 9 time interaction for some of the statistical analyses (Table 1), thus making it somewhat difficult to interpret the data when considering all dates in one single analysis (repeated measures ANOVA). Nevertheless, by-date detailed data show significant differences between treatments in various dates regarding the density of the pests tested (Supplementary Table 2), thus indicating the potential of alyssum flowers to indirectly mediate collard pests suppression. The access of natural enemies to alyssum pollen and nectar may also be facilitated by the morphological structure of these flowers. Alyssum flowers have narrow and shallow corolla which allows natural enemies with different mouth parts to collect its content more easily (Fiedler and Landis 2007). Color has also been found to be an important cue for recognition and detection of flowers from a distance (Kevan et al. 1996). In a previous study, predatory syrphid flies were significantly drawn more often to
plots with white flowers (i.e., alyssum and buckwheat) (Gontijo et al. 2013). In addition, alyssum plants have a compact growth that facilitates the management of the garden floor (i.e., weeding and irrigation) and mitigates the potential for competition with the target crop. Because very few insecticides are registered for the control of collard pests in Brazil, the adoption of striping alyssum flowers with collards could be a viable manner to enhance the management of insect pests attacking this vegetable crop. Previous studies have investigated the best physical arrangements for intercropping alyssum with broccoli in respect to improving syrphid attraction and mitigating plant competition (Brennan 2016). While broccolis and collards are related, they are still morphologically very different, where collards have large leaves and relatively high canopy. Thus, collardspecific experiments that investigate best intercropping arrangement to avoid competition and guarantee natural enemy attraction are needed to further improve the outcome of our present study. In summary, our two-year study shows that strip intercropping alyssum with collards mediated a significant increase in the abundance of generalist predators, which in turn improved the biological control of multiple collard pests (aphids, whiteflies and diamondback moth larvae). The early arrival of generalist predators in the growing season can maximize the prevention of pest outbreaks. Thus, effective strip intercropping will entail methods that ensure that alyssum will be flowering at the time of transplanting the collard seedlings. Although some previous studies have similarly indicated the potential of alyssum
123
A. L. Ribeiro, L. M. Gontijo Table 3 Number of natural enemies summed across dates by different sampling methods Predators
Collards ? alyssum On collards
b.t. collards
Collards alone Pitfall
Bands
b.t. flowers
On collards
b.t. collards
Pitfall
346
26
50
9
Bands
Experiment 1 (2014–2015) Spiders
96
96
27
0
0
Carabidae beetles
0
0
59
0
0
0
0
27
0
Chrysopidae adults
5
16
0
0
23
1
0
0
0
Chrysopidae larvae
0
0
6
0
2
0
0
0
0
Coccinellidae adults
0
83
22
0
77
0
9
7
0
Coccinellidae larve
11
10
8
0
13
7
0
0
0
Orius sp.
66
0
0
0
98
2
0
0
0
Syrphidae adults
80
54
0
0
83
26
0
0
0
Syrphidae larvae
9
0
0
0
0
2
0
0
0
68
29
41
76
34
6
1
14
41
Total
335
288
163
76
676
80
60
57
41
Natural enemies
Collards ? alyssum
b.t. collards
Pitfall
Bands
Earwigs
On collards
b.t. collards
Collards alone pitfall
Bands
b.t. flowers
On collards
Experiment 2 (2015–2016) Spiders
54
57
150
0
279
41
43
147
0
Carabidae beetles
0
0
16
0
0
0
0
6
0
Chrysopidae adults Chrysopidae larvae
0 0
0 0
0 1
0 0
1 2
0 0
0 0
0 0
0 0
Coccinellidae adults
17
16
10
0
18
4
4
4
0
Coccinellidae larve
7
0
4
0
8
0
0
0
0
7
0
0
0
12
0
0
0
0
76
19
0
0
13
35
26
0
0
Orius sp. Syrphidae adults Syrphidae larvae Earwigs
4
1
0
0
0
0
1
0
0
14
10
33
54
6
7
2
27
31
Adult parasitoids
111
90
0
0
212
37
32
0
0
Total
290
193
214
54
551
124
108
184
31
On collards, direct count; b.t., beating tray; bands, cardboard bands
flowers to enhance biological control, none yet provide information regarding the effectiveness of this system in a tropical setting. Acknowledgements We would like to thank Eˆnio Basilio, Lucas Ferreira Neto, Marta Ribeiro and Robson da Silva for helping with the experiment setup and some of the data collection.
References Amaral DSSL, Venzon M, Duarte MVA, Sousa FF, Pallini A, Hardwood JD (2013) Non-crop vegetation associated with
123
chili pepper agroecosystems promote the abundance and survival of aphid predators. Biol Control 64:338–346 Berndt LA, Wratten SD, Scarratt SL (2006) The influence of floral resource subsidies on parasitism rates of leafrollers (Lepidoptera: Tortricidae) in New Zealand vineyards. Biol Control 37:50–55 Bone NJ, Thomson LJ, Ridland PM, Cole P, Hoffmann AA (2009) Cover crops in Victorian apple orchards: effects on production, natural enemies and pests across a season. Crop Prot 28:675–683 Brennan EB (2013) Agronomic aspects of strip intercropping lettuce with alyssum for biological control of aphids. Biol Control 65:302–311 Brennan EB (2016) Agronomic aspects of strip intercropping broccoli with alyssum for biological control of aphids. Biol Control 97:109–119
Alyssum flowers promote biological control of collard pests Chaney WE (1998) Biological control of aphids in lettuce using in-field insectaries. In: Pickett CH, Bugg RL (eds) Enhancing biological control: habitat management to promote natural enemies of arthropod pests. University of California Press, Berkeley, pp 73–83 Colfer RJ, Rosenheim JA (2001) Predation on immature parasitoids and its impact on aphid suppression. Oecologia 126:292–304 Fiedler AK, Landis DA (2007) Plant characteristics associated with natural enemy abundance at Michigan native plants. Environ Entomol 36(4):878–886 Fiedler AK, Landis DA, Wratten SD (2008) Maximizing ecosystem services from conservation biological control: the role of habitat management. Biol Control 45:254–271 Firake DM, Lytan D, Behere GT (2012) Bio-diversity and seasonal activity of arthropod fauna in brassicaceous crop ecosystems of Meghalaya, north east India. Mol Entomol 3(4):18–22 Furlong MJ, Shi Z, Liu S, Zalucki MP (2004) Evaluation of the impact of natural enemies on Plutella xylostella L. (Lepidoptera: Yponomeutidae) populations on commercial brassica farms. Agric For Entomol 6(4):311–322 Gardiner MM, Landis DA, Gratton C, DiFonzo CD, O’Neal M, Chacon JM, Wayo MT, Schmidt NP, Mueller EE, Heimpel GE (2009) Landscape diversity enhances biological control of an introduced crop pest in the north-central USA. Ecol Appl 19:143–154 Gontijo LM (2011) Integrated biological control of woolly apple aphid in Washington State. PhD dissertation, Washington State University, Pullman, USA Gontijo LM, Beers EH, Snyder WE (2013) Flowers promote aphid suppression in apple orchards. Biol Control 66:8–15 Grafton-Cardwell EE, Ouyang Y, Bugg RL (1999) Leguminous cover crops to enhance population development of Euseius tularensis (Acari: Phytoseiidae) in citrus. Biol Control 16:73–80 Gurr G, Wratten SD, Alteri MA (2004) Ecological engineering for pest management: advances in habitat manipulation for arthropods. CSIRO Publishing, Canberra Harwood JD, Desneux N, Yoo HJS, Rowley DL, Greenstone MH, Obrycki JJ, O’Neil RJ (2007) Tracking the role of alternative prey in soybean aphid predation by Orius insidiosus: a molecular approach. Mol Ecol 16:4390–4400 Hassel MP, May RM (1986) Generalist and specialist natural enemies in insect predator-prey interactions. J Anim Ecol 55:923–940 Heimpel GE, Rosenheim JA, Mangel M (1997) Predation on adult Aphytis parasitoids in the field. Oecologia 110:346–352 Hickman JM, Wratten SD (1996) Use of Phacelia tanacetifolia strips to enhance biological control of aphids by hoverfly larvae in cereal fields. J Econ Entomol 89:832–840 Hogg BN, Nelson EH, Mills NJ, Daane KM (2011) Floral resources enhance aphid suppression by a hoverfly. Entomol Exp Appl 141:138–144 Horton DR, Broers DA, Hinojosa T, Lewis TM, Miliczky ER, Lewis RR (2002) Diversity and phenology of predatory arthropods overwintering in cardboard bands placed in pear and apple orchards of Central Washington State. Ann Entomol Soc Am 95:469–480
Kevan P, Giurfa M, Chittka L (1996) Why are there so many and so few white flowers? Trends Plant Sci 1:280–284 Landis DA, Wratten SD, Gurr GM (2000) Habitat management to conserve natural enemies of arthropod pests in agriculture. Annu Rev Entomol 45:175–201 Limburg DD, Rosenheim JA (2001) Extrafloral nectar consumption and its influence on survival and development of an omnivorous predator, larval Chrysoperla plorabunda (Neuroptera: Chrysopidae). Environ Entomol 30:595–604 Naranjo SE (2001) Conservation and evaluation of natural enemies in IPM systems for Bemisia tabaci. Crop Prot 20(9):835–852 Nieto DJ, Shennan C, Settle WH, O’malley R, Bros S, Honda JY (2006) How natural enemies and cabbage aphid (Brevicoryne brassicae L.) population dynamics affect organic broccoli harvest. Environ Entomol 94:94–101 Novo MCSS, Prela-Pantano A, Trani PE, Blat SF (2010) Desenvolvimento e produc¸a˜o de geno´tipos de couve manteiga. Hortic Bras 28:321–325 Oelofse M, Hogh-Jensenb H, Abreuc LS, Almeidad GF, Huie QY, Sultanf T, Neergaarda A (2010) Certified organic agriculture in China and Brazil: market accessibility and outcomes following adoption. Ecol Econ 69:1785–1793 Oliveira MRV, Amancio E, Laumann RA, Gomes LO (2003) Natural enemies of Bemisia tabaci (Gennadius) B biotype and Trialeurodes vaporariorum (Westwood) (Hemiptera: Aleyrodidae) in Brası´lia, Brazil. Neotrop Entomol 32:151–154 Patt JM, Hamilton GC, Lashomb JH (1997) Impact of stripinsectary intercropping with flowers on conservation biological control of the Colorado potato beetle. Adv Hortic Sci 11:175–181 Perdikis D, Fantinou A, Lykouressis D (2011) Enhancing pest control in annual crops by conservation of predatory Heteroptera. Biol Control 59:13–21 Ponti L, Altieri MA, Gutierrez AP (2007) Effects of crop diversification levels and fertilization regimes on abundance of Brevicoryne brassicae (L.) and its parasitization by Diaeretiella rapae (M’Intosh) in broccoli. Agric For Entomol 9:209–214 Ramsden MW, Menendez R, Leather SR, Wackers F (2015) Optimizing field margins for biocontrol services: the relative role of aphid abundance, annual floral resources, and overwinter habitat in enhancing aphid natural enemies. Agric Ecosyst Environ 199:94–104 Rand TA, Tylianakis JM, Tscharntke T (2006) Spillover edge effects: the dispersal of agriculturally subsidized insect natural enemies into adjacent natural habitats. Ecol Lett 9:603–614 Rosenheim JA, Wilhoit LR, Armer CA (1993) Influence of intraguild predation among generalist insect predators on the suppression of an herbivore population. Oecologia 96:439–449 Schmidt JM, Peterson JA, Lundgren JG, Hardwood JD (2013) Dietary supplementation with pollen enhances survival and Collembola boosts fitness of a web-building spider. Entomol Exp Appl 149:282–291 Symondson WOC, Sunderland KD, Greenstone MH (2002) Can generalist predators be effective biocontrol agents? Annu Rev Entomol 47:561–594
123
A. L. Ribeiro, L. M. Gontijo Tixier P, Dagneaux D, Mollot G, Vinatier F, Duyck PF (2013) Weeds mediate the level of intraguild predation in arthropod food webs. J Appl Entomol 137:702–710 Traugott M, Bell JR, Raso L, Sint D (2012) Generalist predators disrupt parasitoid aphid control by direct and coincidental intraguild predation. B Entomol Res 102:239–247 Tscharntke T, Klein AM, Kruess A, Steffan-Dewenter I, Thies C (2005) Landscape perspectives on agricultural intensification and biodiversity—ecosystem service management. Ecol Lett 8:857–874 Tukahirwa EM, Coaker TH (1982) Effect of mixed cropping on some insect pests of brassicas; reduced Brevicoryne brassicae infestations and influences on epigeal predators and the disturbance of oviposition behaviour in Delia brassicae. Entomol Exp Appl 32(2):129–140 Wade MR, Zalucki MP, Wratten SD, Robinson KA (2008) Conservation biological control of arthropods using
123
artificial food sprays: current status and future challenges. Biol Control 45:185–199 White AJ, Wratten SD, Berry NA, Weigmann U (1995) Habitat manipulation to enhance biological control of brassica pests by hover flies (Diptera: Syrphidae). J Econ Entomol 88(5):1171–1176
Andre´ Ribeiro holds a Masters degree in horticulture, and works currently as a laboratory technician at the Instituto Federal Goiano, Campus Morrinhos, Brazil. Lessando M. Gontijo holds a Ph.D. degree in entomology and is currently an assistant professor at the Federal University of Vic¸osa, Campus Florestal, Brazil. His overarching research interests revolve around conservation biological control and applied insect ecology.