Christy Rollinson & Rufus Nicoll

Title: Modeling of hemlock (Tsuga canadensis) mortality as a result of hemlock woolly adelgid (Adelges tsugae Annand)

Background: Hemlock woolly adelgid (HWA) is an invasive aphid from Asia that was first detected in Richmond, VA in the 1950s. It has since spread throughout most of the hemlock’s range from Tennessee up into Massachusetts and has already caused the death of roughly 80% of the hemlocks in Virginia. Although methods of chemical control do exist, they are not feasible on a large enough scale to stop the invasion and so far no successful biological control has been found.

Although HWA has caused complete mortality of hemlock stands throughout it’s range, the time until mortality[not clear what you mean by “time unit mortality” – do you mean mortality rate?] is not universal. McClure (1991) has observed complete hemlock stand mortality in 4 years of initial infestation, but stands that have been infested for 10+ years have also been observed. The research by McClure has shown HWA infestation density to be highest within the first year of infestation and then steadily declining with a peak just before total stand death [how is it possible to steadily decline to a peak?]. This fits patterns observed in the southern range of HWA, but not in the northern most extent. Currently, this is believed to be attributed to winter temperatures that increase HWA mortality.

Research Objective: We hope to look at one or more factors that play a role in the infestation dynamics and hope to be able to create a model that would help determine an expected length of HWA infestation before most of the hemlocks in the stand die. Aspects Independent variables examined will include stand composition and hemlock density and cold temperatures. If enough we have enough time, we might look at the effects of nutrients (mainly nitrogen) in the soil as it has been linked to increased severity of HWA infestation.

Our model will be based off of data available from southern New England (Connecticut and Massachusetts), but if it works, should be applicable throughout the it’s range [of hemlocks or HWA?].

If we encounter a lot of difficulty with this model, it can easily be modified to model HWA spread, which although would contribute less to the current body of research, could still be useful creating management strategies designed at slowing the spread of HWA.

Key State variables, flows, forcing functions etc: The two primary state variables would be HWA and hemlocks [strikes me that you might want to consider hemlock abundance as forcing (initial hemlock density). If you include hemlocks as a stock, perhaps you model this as %]. The model would be for a single population of hemlocks and would include separate stocks for different levels of tree health (ranging from uninfested to infested but healthy to dead) [yes, good]. Factors such as stand composition and low temperatures would be involved in the rate of infestation and deterioration of health by controlling the HWA population. This might require a the creation of a parallel stock of new growth, which is the largest factor affecting HWA population (according to McClure). Data on HWA mortality is available from McClure and others and Orwig has done long term research tracking HWA infestation in southern New England.

[Describe the feedback. Seems to me that it might only be in growth of HWA – but not clear how this relates to the decline in HWA abundance you describe over the course of a stand infestation]

Responsibilities: We will break this model into two sections: one for the HWA population dynamics and one for forest response. One person will be responsible for the primary research and model-building for each section, but we will work closely throughout the whole process.

Biobiography:

McClure, M.S. 1991. Density dependent feedback and population-cycles in Adelges tsugae (hoptera, Adelgidae) on Tsuga-canadensis. Environmental Entomology 42 (1): 258-264.\

  • This paper explains population dynamics of HWA that would be necessary for creating the basic population cycles. Also has average rate of spread. [sounds important]

Koch, F.H., H.M. Ceshire, and H.A. Devine. 2006. Landscape-scale prediction of hemlock woolly adelgid, Adelges tsugae (hoptera: Adelgidae), infestations in the Southern Appalachian Mountains. Environmental Entomology 35 (5): 1313-1323.

  • This describes observed patterns of HWA infestations in the Great Smoky Mountains National Park and identifies observed risk factors for HWA infestation.

Orwig, D.A., D.R. Foster, and D.L. Mausel. 2002. Landscape patterns of hemlock decline in New England due to the introduced hemlock woolly adelgid. Journal of Biogeography 29: 1475-1487.

  • This shows patterns of HWA infestation in New England and can be compared to Koch et al. Includes information on rate of spread of HWA infestation in New England.

Reviews:

#1

First off, the title of this project is excellent and provides a clear indication of what you intend to study. The background gives a solid context for your project, as well as providing a good, quick summary of previous pertinent research. However, I think you need to more clearly outline the “gap in knowledge” you’re addressing [good point]; I gather that you are modeling the northern part of the range because this has not yet been successfully modeled, but this is not clearly spelled out. Also, who’s the audience and why is this important to them? [good point. Is this primarily heuristic or research? My sense is that this is heuristic] What can you do once you’ve determined the expected length of HWA infestation before most of the hemlocks in the stand die?

This is a great, informative project proposal. It also seems quite feasible, and you clearly have a good understanding of the background and where you will need to go to find pertinent data. I would only recommend further elaboration of these aforementioned points. [Nice evaluation]

#2

This project proposal is clear and well-written. Good job guys! Your research objectives seem feasible and very useful to the field. I can definitely see scientists being able to use and even add to your model as time goes on and more is learned about HWA. It does sound pretty complicated, so I liked that you had a back-up plan to model HWA spread, which would also be interesting.

I’m a little confused about how the cohorts will be set up and how you can model degrees of health or infestation [yes, this does need clarification]. This could be due to my lack of knowledge in the area, but I think it could use some clarification.

#3

Project proposal critique: Modeling of hemlock (Tsuga canadensis) mortality as a result of hemlock woolly adelgid (Adelges tsugae Annand)

The title, background information, problem statement and objectives are expressed clearly in this project proposal. I believe their objectives are feasible within the scope of this course.

They mentioned that they would look at one or two factors that help infest Hemlocks with HWA. I am concerned this process may simplify the situation by ignoring the multiplicative factors that “play a role in infestation.” I suggest looking into this multiplicative aspect to improve the model. [ideally a model is built with a minimum level of complexity and then additional complexity can be added]

Their learning objectives are clear, however the audience of this project seems vague. [good point]

The key state variables, flows and forcing functions are well thought out and researched. It appears that they have a good plan on this project.

However, the labor distribution of this project seems a little unfair at this point. Perhaps, the cooperative work plan includes fair responsibilities.

Overall, I think it is a good final project with a good start.