Impact of water-deficit stress on tritrophic interactions in a wheat-aphid-parasitoid system

Abstract

Increasing temperature and CO2 concentrations can alter tritrophic interactions in ecosystems, but the impact of increasingly severe drought on such interactions is not well understood. We examined the response of a wheat-aphid-parasitoid system to variation in water-deficit stress levels. Our results showed that arid area clones of the aphid, Sitobion avenae (Fabricius), tended to have longer developmental times compared to semiarid and moist area clones, and the development of S. avenae clones tended to be slower with increasing levels of water-deficit. Body sizes of S. avenae clones from all areas decreased with increasing water-deficit levels, indicating their declining adaptation potential under drought. Compared to arid area clones, moist area clones of S. avenae had a higher frequency of backing under severe water stress only, but a higher frequency of kicking under well-watered conditions only, suggesting a water-deficit level dependent pattern of resistance against the parasitoid, Aphidius gifuensis (Ashmead). The number of S. avenae individuals attacked by the parasitoid in 10 min showed a tendency to decrease with increasing water-deficit levels. Clones of S. avenae tended to have lower parasitism rates under treatments with higher water-deficit levels. The development of the parasitoid tended to be slower under higher levels of water-deficit stress. Thus, the bottom-up effects of water-deficit stressed plants were negative on S. avenae. However, the top-down effects via parasitoids were compromised by water-deficit, which could favor the growth of aphid populations. Overall, the first trophic level under water-deficit stress was shown to have an indirect and negative impact on the third trophic level parasitoid, suggesting that parasitoids could be increasingly vulnerable in future warming scenarios.

Fig 1. Sampling locations of Sitobion avenae and Aphidius gifuensis (moist area: Longting Town of Yangxian Co., 33°12'43"N, 107°38'30"E; Jinshui Town of Yangxian Co., 33°16'20" N, 107°47'45" E; semiarid area: Yanlian Co., 35°41’29” N, 109°16’13” E; Yaozhou Co., 34° 53’38” N, 108°58’18” E; arid area: Shanglang Town of Minqin Co., 38°35'48" N, 103°06'17" E; Xiaotian Town of Minqin Co., 38°36'40" N, 103°07'18" E; A. gifuensis larvae sampled in Yangling, 34°16’56” N, 108°04’28” E).

Fig 2. Comparison of body sizes of newly emerged adults of Sitobion avenae clones from arid, semiarid and moist areas under three water treatments (n = 18; different letters on bars indicate significant differences among treatments at the P < 0.05 level, nested ANOVA followed by Tukey tests).

Fig 3

Mean frequencies (/10 min) of Aphidius gifuensis contacts with plants carrying 4^th intar (A) and adult (B) individuals of Sitobion avenae under three water treatments (n = 24; different uppercase and lowercase letters on bars indicate significant differences among water treatments within an area and among areas within a water treatment, respectively; logistic regression, P < 0.05).

Fig 4. Number of Aphidius gifuensis attacks in 30 min on apterous adults of Sitobion avenae clones from arid, semiarid and moist areas under three water treatments (n = 24; different letters on bars indicate significant differences among treatments at the P < 0.05 level, nested ANOVA followed by Tukey tests).

Fig 5

Mean frequencies of behavioral resistance against Aphidius gifuensis for adults of Sitobion avenae clones from arid, semiarid and moist areas under three water treatments (A, backing; B, dropping; C, kicking; n = 24; different uppercase and lowercase letters on bars indicate significant differences among water treatments within an area and among areas within a water treatment, respectively; logistic regression, P < 0.05).