Even scientists make mistakes

I went to a seminar titled “Harvesting ecosystem services from cellulosic biofuel landscapes” at Iowa State yesterday. The speaker was Michigan State Professor of Entomology Douglas Landis. His research is very practical, focusing on which types of plants should be used in biomass production for biofuels to encourage the highest biodiversity of insects. This is important because insects provide many ecosystem services, including pollination and predation of pests.

His work shows that switchgrass and mixed prairie encouraged higher numbers of some native insect species, but also encouraged some invasive insects. Corn likewise had mixed results (especially encouraging native insects that like to eat corn!). His methods of data collection are simply fun – low tech but effective – nets, sticky traps, leaf blowers turned into vacuums, and such.

It’s good work, because we really need people working on the sustainability issues associated with farming. Growing biomass for fuel, as any farming, could be beneficial or detrimental to the environment – it’s up to people like Dr. Landis to make sure it’s the former.

Unfortunately, Dr. Landis seems to be misinformed on one issue – Bt crops. I won’t go into too much detail here, but Bt crops have been shown to increase insect biodiversity because overall levels of pesticide are decreased. I would have thought that an entomologist concerned with biodiversity would at least contemplate a cost-benefit analysis of Bt.

In the introduction of his talk, Dr. Landis mentioned the 2007 PNAS paper “Toxins in transgenic crop byproducts may affect headwater stream ecosystems”. This paper is more than a little controversial, because the authors make broad claims that do not follow from their results. The authors make the typical mistake of lumping that is never appropriate in science. All Bt is not the same, all aquatic insects are not the same, all corn fields are not managed in the same way… Dr. Landis should have at a minimum mentioned that the paper has been contested before trotting it out as evidence.

I wrote about the press release of the paper when it first came out, and I continue to be frustrated by the mediocre science and the terrible way that people twist results to make the point they want. Feel free to read the paper for yourself, but don’t pass up the rebuttals. Beachy, et. al. make the point that papers like this are taken by the popular media and used to fuel debates – regardless of the strength of the science. It truly is the responsibility of every scientist and especially of every reviewer to put aside personal bias for or against the paper and make sure the conclusions are supported as much as possible by hard evidence. The reviewers of this paper should have sent it back, requesting some of the experiments explained in the rebuttals (such as dose-response measurements, as suggested by Parrott).

Streamhead with and without corn residues

Dr. Landis showed this pair of images from the Rosi-Marshall paper, implying that corn residues in streams are all the fault of genetic engineering, as if organic corn crops or other crops in general don’t clutter up streamheads. He seems to think that harvesting switchgrass or mixed prarie would somehow not leave the same sorts of residues in streams.

Of course, all crops are going to leave residues in nearby streams, plant matter that is not a natural part of the ecosystem. What I find most ironic is that, if crop residues actually do damage streams, we should start harvesting them for biofuels right now!
To make things even worse, Dr. Landis told the audience (full of impressionable sustainable ag majors) that “the Bt toxin leaches from crop residues into streams where it kills aquatic insects.” This statement is wrong for a lot of reasons, the main one being that there is zero evidence that Bt leaches from the plants that produce it. If you know of any evidence showing that it does, please let me know, but I’m not going to hold my breath.
ResearchBlogging.orgE. J. Rosi-Marshall, J. L. Tank, T. V. Royer, M. R. Whiles, M. Evans-White, C. Chambers, N. A. Griffiths, J. Pokelsek, M. L. Stephen (2007). Toxins in transgenic crop byproducts may affect headwater stream ecosystems Proceedings of the National Academy of Sciences, 104 (41), 16204-16208 DOI: 10.1073/pnas.0707177104