Gene‐to‐ecosystem impacts of a catastrophic pesticide spill: testing a multilevel bioassessment approach in a river ecosystem

Abstract

SUMMARY1. Pesticides can have strong deleterious impacts in fresh waters, but understanding how these effects cascade through natural ecosystems, from microbes to apex predators, is limited because research that spans multiple levels of biological organisation is rare. 2. We report how an accidental insecticide spill altered the structure and functioning of a river across levels ranging from genes to ecosystems. We quantified the impacts on assemblages of microbes, diatoms, macroinvertebrates and fish and measured leaf-litter decomposition rates and microbial functional potential at upstream control and downstream impacted sites 2 months after the spill. 3. Both direct and indirect impacts were evident across multiple levels of organisation and taxa, from the base of the food web to higher trophic levels. At the molecular level, differences in functional gene abundance within the impacted sites reflected a combination of direct and indirect effects of the pesticide, via elevated abundances of microbial populations capable of using chlorpyrifos as a resource (i.e. direct effect) and oxidising ammonia released by decaying macroinvertebrate carcasses (i.e. indirect effect). 4. At the base of the food chains, diatom taxa found only in the impacted sites were an order-ofmagnitude larger in cell size than the largest comparable taxa in control communities, following the near extirpation of their consumers. Population biomass of the key detritivore Gammarus pulex was markedly lower, as was the rate of litter decomposition in the impacted sites. This was partially compensated for, however, by elevated microbial breakdown, suggesting another indirect food-web effect of the toxic spill. 5. Although many species exhibited population crashes or local extirpation, total macroinvertebrate biomass and abundance were largely unaffected due to a compensatory elevation in small tolerant taxa such as oligochaetes, and/or taxa which were in their adult aerial life stage at the time of the spill (e.g. chironomids) and thus avoided contact with the polluted waters and were able to repopulate the river quickly. Mass-abundance scaling of trophic links between consumers and resources revealed extensive restructuring within the food web. 6. This case study shows that pesticides can affect food-web structure and ecosystem functioning, both directly and indirectly across levels of biological organisation. It also demonstrates how an integrated assessment approach, as adopted here, can elucidate links between microbiota,

Large woody debris “rewilding” rapidly restores biodiversity in riverine food webs

“…Abundance (individuals per m 2 ) was estimated using iterative maximum weighted likelihood statistics (Carle & Strub, ). Dry‐mass estimates were made for each species using length‐mass regression equations and wet‐to‐dry mass conversions (Thompson et al., ). A large Hess sampler (0.14 m 2 ) was used at the Wensum impact reach due to the relatively low sampling efficiency of bullhead ( Cottus gobio L.), as catches did not reduce on consecutive runs during electrofishing (Lauridsen et al., ).…”

“…This approach has been widely applied (e.g. Mulder & Elser, ; Pocock, Evans, & Memmott, ; Strong & Leroux, ) , especially in the construction of river food webs (Gray et al., ; Layer, Hildrew, & Woodward, ; Layer, Riede, Hildrew, & Woodward, ; Thompson et al., ). The percentage of directly observed links that were also reported in the literature was 99%: i.e., only 1% of the 4,535 observations from gut contents analysis were new records.…”

Recommendations for the Next Generation of Global Freshwater Biological Monitoring Tools

“…At present though, these more holistic and information-rich approaches are still in the early stages of development as novel monitoring tools (e.g. Thompson et al 2016;Gray et al, THIS VOLUME) and still some way behind the other more measures we have highlighted here as approaches that could realistically be applied globally in the near-future. MiniSASS (http://www.minisass.org/) in South Africa uses a similar approach and both have large and growing online data repositories (Box 2).…”

“…Yet despite the functional importance of these groups, and established NGS and qPCR approaches for quantifying them (e.g. Papaspyrou et al 2014;Li et al 2015;Lansdown et al 2016;Thompson et al 2016), they are still rarely included in monitoring of diversity change. Combining information on functionally important groups of microbes with eDNA based assessments of biodiversity could provide a new paradigm in monitoring, as it provides direct insights into the functional capacity and health of ecosystems.…”

“…For example, a small spill of the organophosphate chlorpyrifos, which wiped out invertebrate life over a 15km stretch of a UK river, triggered a greater than three orders of magnitude increase in the abundance of microbial populations containing the organophosphate hydrolase gene (opd; Thompson et al 2016). In addition, collection of the samples required for all molecular approaches is straightforward and therefore there is potential for convergence with with citizen science approaches (Box 2) and global coverage is feasible.…”