I didn’t expect to write about insecticide use again so soon, but upon reading Christie Bahlai and colleagues’ new manuscript (preprint available online in Ecological Applications), I find myself pulled back in.
Their paper (Title: “Shifts in dynamic regime of an invasive lady beetle are linked to the invasion and insecticidal management of its prey”) presents data on regional trap captures of Harmonia axyridis, along with reports of soybean aphid infestation levels. The authors subdivided H. axyridis densities into 3 phases in time representing different population growth regimes.
The authors identified these 3 phases with an iterative model selection approach that fit to their data two alternative discrete time models (Ricker and logistic). Both the Ricker and the logistic model include two demographic parameters: carrying capacity (K) and intrinsic growth rate (r) which are used to describe density-dependent growth.
The 3 phases they identified were:
(1) a stable low K period which existed from 1994 to 2000
(2) an unstable, high r period from 2001 to 2005
(3) a return to a lower K phase from 2006 to 2013
By identifying break points in time that corresponded to the dynamical properties of the the lady beetle population, the authors next sought to identify potential extrinsic factors which might explain the different parameter values found during the 3 phases. The soybean aphid invasion is associated with the shift from phase (1) to phase (2). This association is correlative, and so causal inferences should be made cautiously. However, it seems like reasonable reasonable explanation, and there is an abundance of literature supporting a strong link between soybean aphid densities and H. axyridis population growth.
Next the authors examine the return to a lower K in the shift from phase (2) to phase (3).
The authors find an interesting correlation between the adoption of seed treatments and a decline in soybean aphid density. They suggest – contrary to the commonly held belief that seed treatments are ineffective against soybean aphids (see earlier post) – that seed treatments are suppressing soybean aphids and, therefore, indirectly suppressing H. axyridis populations.
The usual thought is that seed treatments don’t impact soybean aphid densities because the major outbreaks occur too long after planting. Bahlai and colleagues argue to the contrary. They suggest that high titers of insecticide present in young soybean plants could impact winged aphids (and their offspring) as they migrate early in the season from buckthorn to colonize young soybean plants, and this could have important impacts on aphid densities later in the season.
This is an interesting idea, and their correlative data are striking. My main criticism is that they don’t give enough consideration to the alternative hypothesis that other exotic biological control agents could be impacting soybean aphid on the same relative time scale as the increased use of seed treatments. In the discussion, they mention the possibility of Aphelinus certus being a factor, but they dismiss it as unlikely:
“… given that A. certus was detected through much of the North American range of soybean aphid as early as 2005, it is likely that the parasitoid arrived with soybean aphid or soon after (Desneux et al. 2009, Heimpel et al. 2010). Thus, if the decline of soybean aphid and shift in Harmonia’s dynamics were primarily driven by the arrival of this parasitoid, we would have expected to see shifts in dynamics prior to 2005, or a slow shift in dynamics, rather than discrete break points.”
First, I think they overstate the evidence that A. certus was widespread prior to 2005. For example, experiments in 2004 and 2005 by Costamagna et al. (2008) indicated low parasitism of soybean aphid by resident parasitoids, and those that were recorded were mostly Braconidae (A. certus is in the family Aphelinidae). Second, the authors report a “discrete break point” after the arrival of the soybean aphid, so why would they expect to see a “slow shift in dynamics” with the arrival of a biological control agent? I don’t understand why we should expect this kind of fundamentally different consequence from the arrival of a resource population versus the arrival of a competitor population.
Criticisms aside, I thought this was a fascinating paper. It introduced me to some cool methods of model fitting. And if their hypothesis about the role of seed treatments in soybean aphid control is correct, this may substantially alter the debate about the role of prophylactic neoncicontinoid seed treatments in soybean fields.