Decomposition analysis of LTREs may facilitate the design of short-term ecotoxicological tests

Abstract

This study compared two methods, based on re-analyzed data from a partly published life table response experiment (LTRE), to help determine the optimal approach for designing ecotoxicological assessments. The 36-day LTRE data recorded the toxic effects of cadmium (Cd) and imidacloprid, alone and in combination, on the reproduction and survivorship of aphids (Acyrthosiphon pisum Harris). We used this data to construct an age-classified matrix model (six age classes, each 6 days long) to estimate aphid population growth rate (λ) under each treatment. For each treatment, an elasticity analysis and a demographic decomposition analysis were performed, and results were compared. Despite different results expected from the two toxicants, the elasticity values were very similar. The elasticity of λ with respect to survival was highest in the first age class, and that with respect to fertility was highest in the second age class. The demographic decomposition analysis examined how changes in life-history traits contributed to differences in λ between control and treated populations (Δλ). This indicated that the most important contributors to Δλ were the differences in survival (resulting from both demographic sensitivity and toxicity) in the first and the second age classes of aphids and differences in fertility in the third and the fourth age classes. Additionally, the toxicants acted differently. Cd reduced Δλ by impairing fertility at third age class and reducing survivorship from the second to the third age class. Imidacloprid mostly reduced survivorship at the first and second age classes. The elasticity and decomposition analyses showed different results, because these methods addressed different questions about the interaction of organism life history and sensitivity to toxicants. This study indicated that the LTRE may be useful for designing individual-level ecotoxicological experiments that account for both the effects of the toxicant and the demographic sensitivity of the organism.

The compadrePlant Matrix Database: an open online repository for plant demography

“…Matrix population models describe the dynamics of populations over a discrete projection interval: where n ( t ) and n ( t +1) are population vectors containing the number of individuals in each life cycle stage at times t and t +1, respectively, and A is the population projection matrix. The projection interval can vary from days (Hamda, Jevtic & Laskowski ) to weeks (J. Metcalf, pers. comm.…”

“…where n(t) and n(t+1) are population vectors containing the number of individuals in each life cycle stage at times t and t+1, respectively, and A is the population projection matrix. The projection interval can vary from days (Hamda, Jevtic & Laskowski 2012) to weeks (J. Metcalf, pers. comm.…”