This summer, I’ll be giving a seminar on “Arboriculture Myths” at the ISA conference in Portland, OR. I’ve been quizzing arborist-types for a few months now to find out what myths they would most like to see debunked during my talk. Intermixed with the suggestions of dubious products and questionable practices there was this question: “How often do the results from research with limited scope get over-extrapolated?”
I like the question a lot, because this is the fine line that we Garden Professors walk in bringing you the newest scientific information we can find. As a rule, I tend to hold back on recommending anything that has only been tested in a lab situation. I like to see field test results, where environmental variation will quickly swamp anything with marginal effects. In other words, if something can make it through an experimental, replicated field test, I can get excited about it.
Which brings me to a recent article in Arboriculture and Urban Forestry (2012, Vol. 38, Issue 1, pp. 18-23. And no, I can’t post it on the web). Briefly, the article describes an experiment where water evaporation was measured in pots filled with various substrates, which were either left uncovered or mulched with about 3” of pine bark. The results showed little difference between the mulched and unmulched containers.
As the authors point out in the discussion, it’s an artificial system that includes no trees, nor any way for water to move through the soil except from the top down. And I really don’t have a problem with the methodology, or the data generated, or even most of the discussion. What bothers me is a single sentence at the end of the abstract:
“Given the minor reduction in evaporation, and reported disadvantages of mulch application close to the trunk, landscape managers might consider changing mulch application practices for newly planted trees.”
Wow. How did we get from a series of containers with no plants in them to this recommendation?
Every gardener knows the value of mulching – a perception that’s substantiated by hundreds of publications. Since I’ve written about mulches on the blog a number of times I’m not going to belabor the point. But I will refer readers to a short Ecological Restoration article I published a few years ago that most definitively linked mulch application to plant survival in restoration sites; Bert also published an article on the benefits of mulching and lent me a few photos to illustrate. And Jeff has even more data on the topic, including some that may radically change the perception that mulch against tree trunks is a bad thing.
Mulch increases soil moisture
Which plot would you rather have in your garden?
Those of you who read scientific journals probably read the abstract first – I know I do. If it interests me, I’ll read the entire article. But sometimes the abstract is the only thing you can find online. And for this reason, the peer-review process in many of the journals asks whether the contents of the abstract are justified by the results. Honestly, I don’t think this article meets that standard.
I’ve usually thought that over-extrapolation of results happens as people other than the original researchers interpret the results. Journalists, for instance, or people with vested interests or who are looking for a way to make money on it, but sometimes other researchers who really ought to know better. It’s a huge problem in education research where, for instance, a study on bilingual children in an immersion school will get generalized to adult seniors living in a foreign country, or a small, limited study on improving reading skills gets generalized into a huge, multi-million dollar program with books, workshops, etc. There are lots of fads in education, often with limited justification. I studied teaching ESL for a while, and had a professor whose mantra was ‘Check the original study!’ But it’s very sad that the original researchers are the culprits in your example. Unless someone else is responsible for writing the abstract?
Thanks for writing on this topic. One question I often face in the arid west of Colorado is the following: “During very dry periods, organic mulch- and the soil below it- dries out. When it finally rains, only the mulch gets wet. None of it penetrates to the soil below. In areas with rock mulch, the rain actually wets the soil. Am I better off using rock mulch in parts of the country with erratic rainfall?” Any thoughts?
Could you please post the citation for the graph entitled, “Mulch increases soil moisture”?
Catherine: I don’t have any direct experience but I suspect it won’t make much difference. If things have been dry for while and you have a light rain, say, 1/10 inch or less, I suspect little of that will reach the soil in any event. One thing that will happen with organic mulch is you will have roots begin to proliferate in the soil mulch interface and in the mulch itself. I’m not sure if anyone has any data to show if that is good, bad, or indifferent.
Here’s the citation for the paper. The figure above is simplified from Fig. 2 (2005 data) in the paper.
http://www.hrt.msu.edu/assets/PagePDFs/bert-cregg/Cregg-and-Schutzki.pdf