Forestry Data Science at Reed College
This is the application to participate as a research assistant in Prof. Kelly McConville's Summer 2020 Forestry Data Science Research Group at Reed College. There will be four research assistants: two students from Swarthmore College and two from Reed College. (Unfortunately, this opportunity is not open to seniors who are graduating in Spring 2020.) The application is due no later than January 31st. The ten week program will run from June 1st to August 7th. Participants will receive a competitive stipend but must arrange their own housing. Please e-mail Kelly for more details on housing options.

The group will work on problems generated by the US Forest Inventory and Analysis Program (FIA) and will collaborate directly with Research Statisticians and Foresters at FIA. FIA is tasked with providing a comprehensive inventory of the US's forests. With the rise of new data sources, such as GIS data and large scale photography, and with the explosion of new statistical learning tools, a wealth of estimation techniques are available to consider.

Students will work on projects in groups of two and will be assigned to approximately two projects. Here are two of the tentative projects for 2020:

1. FIA uses a variance estimator that assumes the data were collected via simple random sampling. They actually collect their data using a geographically systematic sample. Last year, one group found evidence of negative bias in the variance estimator when applied to estimation in Daggett County, Utah. This year, we'll explore if that bias holds over different landscapes and at various sampling intensities. We will also create an R package of variance estimators which account for the spatial correlation. Check out this video for more info on last year's project: https://www.causeweb.org/usproc/eusrc/2019/virtual-posters/5

2. The LANDFIRE program creates several maps that correlate nicely with forest attributes of interest to FIA. One map of particular interest is the Existing Vegetation Type (EVT) map. Incorporating this map into FIA's estimation engine would likely improve the precise of many of their estimators. However, the EVT has over 600 categories, making it an unwieldy variable for post-stratification, the most common estimation technique for incorporating categorical auxiliary data. Last year's group created two new versions of EVT that have 5 and 10 categories. This summer, we will explore the utility of including these new EVT variables in a post-stratified estimator. Check out this video for more info on last year's project:
https://www.causeweb.org/usproc/eusrc/2019/virtual-posters/3

The projects will vary in terms of the computational and statistical skills needed but each research assistant should have prior experience coding in R and building statistical models. For the Reed students, these requirements can be met through taking both Math 141 and Math 243. If you have not taken these two classes, you are still encouraged to apply but should address your level of proficiency in R and your experience with statistical modeling.



Happy Applying!
First Name *
Last Name *
Email address *
Class Year (for 2020-2021) *
List the statistics courses you have taken. *
List the computer science courses you have taken.
In 500 words or less, explain why you are interested in this research project. *
Empathy, ability to work on a team and good communication skills are vital to the success of this project. In 300 words or less, please discuss your strengths with respect to these traits. *
What are your professional goals? *
Name of reference *
How do you know the reference person? *
Email address of reference person *
Anything else you want me to know?
Submit
Never submit passwords through Google Forms.
This form was created inside of Reed College. Report Abuse