I'm compiling some examples of good bad aquatic biomonitoring studies for a commentary on monitoring practices. I have a couple published bad examples , but am on the lookout for more good, bad examples.
For example, in a study of power plant effluents discharged near the transition of a flowing river to a standing water reservoir, the reference sites were located in the flowing river and the assessment sites were located in the reservoir. The investigators concluded that benthic communities are different in rivers and reservoirs.
In another reference example, the investigators dispensed with reference sites all together because they were only interested in whether there were changes year to year in assessment sites, not in whether the reference conditions were different.
Another bonehead design is to use chemical detection limits that were higher than the water quality criteria that are being assessed.
I'm after citable works that can be specially discussed in open. Appreciate any leads, which may of course be sent privately.
Confoundedly yours,
Chris
I coauthored a biomonitoring report in 1984 on the Spruce Budworm Project in Oregon on my Researchgate. The map quality, scanning and report writing tools and technology back then was not comparable to today and the results had to be produced quickly, so you would have to decide if you think good or bad. As with many writers, one doesnt always get the necessary time and support to produce a super document before moving on to other work.
Thanks, but I'm not a critic of production quality. I'm after the truly atrocious, logically conflicted monitoring, some of which I'm sure has found its way into the published literature. Working on a small-scale outing of ill conceived monitoring and pseudo-assessment literature in the footsteps of Hurlburt's critique of pseudoreplication.
And thanks to those who have responded privately! I appreciate that this could be a sensitive request when it involves current practitioners.
Well, almost all of the Canadian Metal Mining Environmental Effects Monitoring program reports are pseudo-replicated when used to understand whether there has been an effect of effluent discharge on downstream waters.
Not sure if the book of Indian colleagues is in the sence you are meaning - esp chapter 14.
Andrey
Thanks much, Andrey. Scanned through the excerpt from "Biomonitoring of Water and Waste Water" and it looks to be very well done and useful. I was looking for bad examples, but I'll happily take a good example. Appreciate you're pointing it out. I have to admit to provincialism and have not followed Indian literature much. I will try to have a broader view.
Dave, the pseudo-replication problem is thorny if dealing with flowing waters with upstream reference and downstream exposed sites. That's probably the most common setting at the EEM sites. It's all the same stream, but sampling amidst other streams has its own problems, not the least of which is logistics. Thanks!
Dear Christopher,
All monitoring can be qualified for my opinion as bad if it isn't shown an effect on biota as result. Monitoring is not only passive following some time series or periodical sampling but finding relations and - the most important - action-reaction. So, should be provided by perfect statistical analysis, as well modeling (diff equations) as a result!
This last is the ideal variant enough rare in practice. But I'm a bachelor in the theory of monitoring, just field hydrobiologist. Correct me if I'm not right.
Andrey
Belatedly, I'll weigh into this. I'm leary of judging studies that may have different objectives than my own. By biomonitoring, are you referring just to abiotic pollution? I recently published a long-term study that's relevant to reed canarygrass invasion (biotic pollution):
Vadas, R.L. Jr., H.A. Beecher, S.N. Boessow, and J.H. Kohr. 2016. Coastal Cutthroat Trout redd counts impacted by natural water supply variations. North American Journal of Fisheries Management 36: 900-912 (cf. http://www.tandfonline.com/doi/full/10.1080/02755947.2016.1173138).
It’s an agency study, something that’s harder to accomplish as a grad. student or consultant. The latter provides some of the worst cases of biomonitoring, based on my past experience (Vadas 2015).
-Bob Vadas, Jr.
Vadas, R. Jr. 2015. Bias in aquatic-habitat science and potential solutions, with a focus on instream-flow issues (abstract). Page 21423 in American Fisheries Society (ed.). 145th Annual Meeting. Portland, OR (https://afs.confex.com/afs/2015/webprogram/Paper21423.html).
Thanks much. A number of good examples, but It's hard to get people to volunteer knowing of the bad examples (no reference, sketchy stats or no stats, can't really tell what was done, confounded measurements, ....)
In general, I have problems w/ the IBI (index of biotic integrity) for its ad hoc nature (re: metric formulation) in assessing aquatic ecosystems. Notably, it has under-represented the importance of omnivory in forested streams like scenic-designated Goose Cr., which I studied as a grad. student:
Vadas, R.L. Jr. 1990. The importance of omnivory and predator regulation of prey in freshwater fish assemblages of North America. Environmental Biology of Fishes 27: 285-302 (cf. http://link.springer.com/article/10.1007/BF00002747).
Vadas, R.L. Jr. 1991. Feeding habits and omnivory of forage fishes in Goose Creek, Virginia (abstract). Bulletin of the North American Benthological Society 8: 104.
-Bob Vadas, Jr.
Hi Chris, are you preparing something based on this? If so, I would really like to read it. Anyway, interesting to see the comment "almost all of the Canadian Metal Mining Environmental Effects Monitoring program reports are pseudo-replicated." I am sort of glad to hear this. The pseudoreplication is often hard to overcome (e.g., appropriate sites) and picking up several more (reference) sites may increases the unwanted data variation (so there should be a tradeoff).
Here's one where we kept adding replicates to go from bad (low N) to good (high N). We used 16 metrics showing that some were very useful and others have little discriminatory power. We also cite a paper by Vince Resh that highlights how few replicates people typically use in biomonitoring studies: Article Measuring standardized effect size improves interpretation o...
As you said, no one wants to publicly disrespect a colleague's work. But there are objective measures: how many replicates? appropriate controls? randomization? statistics? If you are doing this with students...make it their problem. Give criteria and send them out to find 2 or 3 good and bad examples. They'll certainly learn more and will internalize the elements of good design better than from a lecture. And you'll have lots of examples for next year.
I do apologize in advance for not providing examples and I don't wish to start a trend here. I am just curious as what is your take on upstream, downstream reference sites. That is for biomonitoring freshwaters. I am somehow against such approach as habitats are significantly differ even before occurrence of any disturbance. Therefore, whatever data we may obtain (both chemical and biological) does not account for natural difference among the sites. If that's the case there are plenty of bad examples in literature. Finding studies that account for natural gradients of species distribution and abundance or even chemical may provide good examples. Monitoring two (or more) hydrologically comparable freshwater one with disturbance and other without. Just a thought!
Dr. Namayandeh, a paper from the 1990s called "Beyond BACI: the detection of environmental impacts on populations in the real, but variable, world" by Underwood started quite a conversation in the literature. Good place to start.
- Badges
- Science topic
- Similar topics
- Solid State Physics
- Devices