By Brandon Gallagher Watson
In 2007, at the onset of emerald ash borer (EAB) infestation in the Greater Chicago Metropolitan area, The Morton Arboretum, The Village of Hazel Crest, USDA APHIS/Forest Service, Davey Tree Expert Company, and Rainbow Treecare Scientific collaborated to implement an emerald ash borer research trial that would take place over several years in the Village of Hazel Crest. This study involves hundreds of ash trees and two different neighborhoods. It is looking at eleven different treatment protocols, representing three different active ingredients, applied in three different ways, and at different times of the year. The treatments, including untreated trees for comparison, were randomly assigned to each tree. The ash trees are city-owned, boulevard trees of a similar size class growing in similar soil conditions. As emerald ash borer populations are relatively high in these neighborhoods, no additional insects were needed to get pressure on the trees. This trial was slated to continue until all the untreated trees in the trial have died.
Five years later, this study is still ongoing with treatments and evaluations. UDSA-APHIS has agreed to extend this trial for at least one more year; with the possibility of extending it for up to three or four additional years to study the effects treatments have on survival rates beyond the initial EAB population wave. At the end of the trial, the data will be analyzed and the results compiled into a paper that will be submitted to a peer-reviewed scientific journal.
Research is the systematic pursuit of knowledge, and in the field of urban forestry, research can take many forms. There is a biological component to arboriculture, dealing with the growth and development of living things. This research focuses on trees, soils, insect and disease pests, physiology and ecology. As urban forestry is also the study of how trees and people interact, our industry is involved in social science research as well. Biological research frequently relies on experimentation — evaluating the effects or lack of effects of a single variable versus an untreated control group. Bioassays, nursery trials, field trials using natural insect or disease pressure, and field trials using challenge (or artificial) pest pressure are just some of the ways we investigate trees.
Social science research, on the other hand, is not as concerned with establishing an ultimate ‘correct’ answer to a question; rather, it deals with exploring the issues and details that surround a question. Although social science research can dabble in such unscientific realms such as opinions or attitudes, studies that show how citizens feel better about their communities when shaded by healthy trees are no less valuable to our industry.
The scientific method
New tools are constantly being developed that allow scientific inquiry to explore questions in increasingly deeper levels. Molecular analysis, DNA, spectroscopy, and chlorophyll florescence are all available to the modern researcher, but the basics of the scientific method haven’t changed much in the past few hundred years. All scientific investigations begin with conjectures, or predictions, about what one thinks may happen given a certain set of parameters. Better known as the hypothesis, these predictions are presented as a statement of fact, then the scientific methods attempts to disprove the hypothesis.
For example, let’s say you were interested in why success rates of soil-applied, imidacloprid insecticide treatments for emerald ash borer appear to be inversely related to the size of the ash tree. Put another way, the bigger the tree, the higher the chance the insecticide will not prevent the tree from dying from EAB, especially for trees above 15 inches in diameter. The reasons could be related to the dosage rate the treatment was applied at, where the treatment was applied (at the base of the tree or in a grid pattern within the dripline of the tree), or when the treatment was applied (spring, summer or fall). How would you test this, using the scientific method?
First, you have to ask a question. In this case, the question is “why do failure rates of soil-applied imidacloprid increase as the size of the tree increases?” Next, you need to state a hypothesis. The scientific method can only test one hypothesis at a time, so from the possibilities we mentioned above (dosage rates, treatment placement, or timing) settle in on the one you wish to test for. This is where a literature review of previously published research comes in. You would look for studies that are related to the question you are asking in an attempt understand what is currently known about the subject and eliminate the number of unknown variables.
In our example, we’ll say dosage rates are our focus, and we’ll phrase our hypothesis as a statement of fact. It can be wordsmithed a number of ways, but we’ll say “Current industry label rates of soil-applied imidacloprid under-dose trees over 15 inches in diameter.” Now we make a prediction: if current rates under-dose trees over 15 inches in diameter, we should see a lower death rate of larger trees if the dosage rate was doubled. To test this prediction, we design an experiment.
There are several ways we could design an experiment to test this hypothesis. Using nursery stock trees has many advantages for arboricultural research. For a relatively low cost, you can get any number of specimens of the same size, with similar or identical genetics, grown in nearly identical conditions. Eliminating as many variables as possible allows you to observe the effects of just the one variable we are interested in (in this case, dosage rates). Our hypothesis, however, is related to trees over 15 inches in diameter and finding a large number of ash tree at a nursery this size that would be available to use for an experiment would be nearly impossible, so we need to look for a large number of suitable trees growing in similar conditions that could be candidates. That’s where cities can become living laboratories.
Finding municipal partners for urban forestry research is an important link for the furthering of our body of tree care knowledge for many reasons. First and foremost, it provides data on real trees under real conditions. We don’t have to extrapolate that the results we saw on 2-inch-diameter trees in controlled growing conditions will translate to a mature tree in someone’s yard. Second, it forms valuable public/private relationships between municipalities, universities, product manufacturers and commercial tree care companies. Finally, municipal research trials are a great way to educate and bring awareness about trees to the people.
Another question for this experiment is where is the emerald ash borer pressure going to come from? Are we going to place adult EABs on every tree, or just let natural populations challenge the trees?
This part about “challenge pressure” is extremely important when weighing the value of any field trial. How do you know the treatment worked if the tree was never challenged? If you treat, say, 50 trees with a treatment for Dutch elm disease (DED) and the next year 25 are dead, does the treatment work 50% of the time, or does it work 0% of the time and only half the trees were unfortunate enough to have a fungus-carrying beetle land on them? Unless each tree was actually intentionally challenged with the DED fungus by, for example, injecting the fungus into the tree, one would never know for sure. Unfortunately for the ash trees in our EAB study, there are more than a few municipalities with emerald ash borer populations high enough that one can assume even pressure on all the trees that would be candidates for this trial.
Now we determine the treatments we wish to evaluate for our study and randomly assign them to the trees in the trial. Randomization is an important step as this eliminates the bias of researchers and is necessary for producing valuable statistics. Untreated trees are also one of the treatments in any study as they are the “control group” — the group against which all our treatments will be evaluated. Our trial will also include trees treated at the current industry rates and rates higher than this to satisfy our hypothesis. Depending upon the type of trial we are undertaking, these treatments may be evaluated just once or reapplied and evaluated over several years.
Evaluations can be made several ways for a trial like this. Percent-canopy-decline is the most common evaluation for field trials, where an observer estimates the percent of the canopy that appears thin or dead compared to a full canopy. We could also sample branches from the trees, remove the bark, and count the number of viable EAB larvae in the branches. This does give a more accurate assessment of the effects of the treatments, but is also more costly and time consuming.
When our data is collected, we analyze the effects the treatment had and compare the results to our untreated, control trees. This is where we revisit our hypothesis: “Current industry label rates of soil-applied imidacloprid under-dose trees over 15 inches in diameter.” Rather than asking if our hypothesis has been “proven right,” we ask, “has it been falsified?” If the data show there is no statistical significance between our trees treated at current industry rates and the higher rates we used in our study, we would reject our hypothesis that the higher death rate of larger trees was related to a dosage issue and begin the process of evaluating another hypothesis. Conversely, if our data showed the trees treated at greater dosage rates survived in significantly higher numbers than those treated with industry standard rate, our hypothesis has not been falsified, and we can begin the process of making a label rate change with the EPA.
Over the past few decades, the arboriculture industry has striven to move us away from the “spray and pray” tree care of the past by developing predictable protocols through rigorous employment of the scientific method at every step of the process. Funding for research is, more often than not, the limiting factor determining if a study is preformed or not. Product manufacturers are a major player in tree care research as data is the currency of success in this market, but they are not the only sources of research for arborists. Universities, non-profit organizations, and grant-giving foundations such as the TREE Fund are important entities for supporting the growing knowledge base of arboriculture. New breakthroughs and a better understanding of our current practices will all come from a strong foundation in research. If we continue to push for a better understanding, just think where the next few decades will bring us.
Brandon Gallagher Watson is director of communications at Rainbow Treecare Scientific Advancements, and is an ISA Certified Arborist (#MN-4086A).