Wildfire Data Science Challenge

Competition Setting

As described on the website of the World Health Organization, a wildfire is an unplanned fire that burns in a natural area such as a forest, grassland, or prairie. Wildfires are often caused by human activity or a natural phenomenon such as lightning, and they can happen at any time or anywhere. Wildfires have a number of negative impacts on communities and individuals. They can cause serious health problems, loss of life, and long-term air pollution. In addition, they can have a significant economic impact, destroying properties and businesses and harming tourism and recreation. These events can be devastating for affected communities.

According to the National Interagency Fire Center (NIFC), in 2021 there were a total of 58,733 wildfires in the United States, which burned a total of 7.1 million acres of land. The average number of wildfires per year in the US over the past decade has been around 70,000, with an average of about 7 million acres burned per year. In Texas, wildfire activity can vary significantly from year to year. In 2021, there were around 3,700 wildfires in Texas that burned a total of about 200,000 acres of land. In 2020, there were about 8,000 wildfires in Texas that burned a total of about 2.5 million acres of land.

Each state in the United States has a forestry agency, but Texas was the first in the nation to establish its state forestry agency as part of a land-grant college. Texas A&M Forest Service (TFS) is one of seven Texas state agencies headquartered in College Station, Texas. In collaboration with TFS, the Texas A&M Institute of Data Science (TAMIDS) is organizing the 2023 Data Science competition to seek innovative ideas for the application of data-driven approaches in wildfire research and education across multiple levels.

Competition Challenge

Data-driven wildfire research involves using data and statistical analysis to understand and predict wildfire behavior. This can include analyzing historical data on wildfires to identify patterns and trends, using remote sensing data to map and monitor fires in real time, and using weather and climate data to forecast fire conditions and potential fire spread. The goal of data-driven wildfire research is to improve our ability to predict and manage wildfires and ultimately reduce the impact of fires on human communities and natural resources.

With this competition, we are looking for innovative and sustainable solutions that can help better predict and manage wildfires and ultimately mitigate the effects of fires on both people and the environment. Particularly, the contestants are expected to address one or both of the following challenges and provide actionable advice to decision-makers based on their analysis.

• Predicting wildfire behavior: Wildfire behavior is influenced by a complex interplay of factors, and it is difficult to predict how a wildfire will behave in a specific location. This makes it challenging to develop accurate models and forecasting tools.
• Communication with end-users: Researchers need to effectively communicate their findings to end-users such as land managers, policymakers, and the public, which can be challenging to achieve in a timely and effective way.

Data & Resources

Data resources will be shared and posted here at the competition period opening on February 28, 2023.

Competition Organization & Schedule

Registration Deadline: Sunday, February 26th, 2023

• Registration: Competitors register using a Google Form (TAMU NetID login required) and acknowledge their understanding and intent to follow the rules of the competition. All registrations will be acknowledged. Registrants should list their teammates. Registrations can be updated to include further teammates. All team members must individually register by 11:59 pm February 26, 2023; otherwise, they will not be included in the competition.
• Teams, Divisions & Mentors: Students will work in teams of up to 5 members. The competition is split into two divisions: graduate and undergraduate. A team that contains at least one graduate student will be assigned to the graduate division. Teams have the option to have a faculty mentor to provide guidance. Competitors must obtain the agreement of the mentor to serve in this role before listing them in the registration. Competition judges and organizers cannot serve as team mentors.
• Eligibility: The competition is open to graduate and undergraduate students from all majors at Texas A&M University, including Galveston and Qatar campuses. Competitors must be enrolled as a student at Texas A&M University during the Spring 2023 semester.  
• Find Teammates through the Competition Slack Channel.
  • Quick start guide: (1) Click on the. Slack Channel; (2)  Click on “create an account” in top right corner; (3) Under “OR”, enter TAMU NetID and click “Continue”; (4) Click “Confirm Email” in the confirmation email; (5) In the browser window that opens, enter your name and chosen password then click on “Create Account”; (6) Open workspace in browser or app.
  • Introduce yourself in the slack channel!
  • Problems? Email Angie Rollins to be invited to the channel.
• Competition Operation: Data release and competition submission will be conducted through the competition’s Canvas community, into which the organizers will enroll registrants and configure teams.
• Questions or Concerns? Email Angie Rollins.

February 28, 6-8 pm: Technical Orientation Session (Hybrid); Data Release; Competition Opens

• Technical Orientation Session: will be held on Feb 28th, 2023, 6-8 pm, in person in the ILSB with remote participation via zoom. The session will present information on conference organization, the competition datasets, the competition context of bibliometrics analysis, and some orientation around approaches to statistical analysis, visualization, and data science project management.
• Data Release: datasets will be released through Canvas and registered teams may commence their analyses.

March and Early April: Office Hours (Online)

• Office Hours: Members of the organizing committee will be on hand to advise competitors on access and use of the competition data resources, and technical issues concerning analysis.
  • 4pm – 5pm on March 17th, March 21st, and April 7th
  • Zoom link: https://tx.ag/tamidsdsc2022

March 23, 6-7 pm: Midpoint Event (Online): Best Progress Graphic Prizes

• Best Progress Graphic Prize: Teams may optionally submit a one-page summary graphic of their initial work via Canvas by Tuesday, March 21, 2023, 11:59 pm. The graphic must include the team name. The submissions will be displayed and prizes of $250 will be awarded to each of the top three team entries at the online midpoint event on Thursday, March 23, 2023, 6:00 pm-7:00 pm. Zoom details to be announced.

April 4: Report Submission Deadline

• Report Submission & Format: teams will submit their report through Canvas as a PDF file, maximum 10 pages using 10 pt Arial font with 1-inch margins all around.  Teams may submit supplementary materials such as (but not limited to): a code or data repository, a Jupyter notebook, a dashboard, or an app. A rubric will be supplied to registered competitors through Canvas. Reports must be submitted through Canvas by the deadline of April 4, 2023, at 11:59 pm.

April 11: Finalists Announced

• Finalist Selection and Preparation: After the close of the submission period, judges will review all entries and select participants to advance to the final round of the competition. Finalist teams will be notified through email by April 11, 2023. Finalist teams will prepare a 10-minute presentation for in-person delivery of their findings and solutions at the finalist event.

April 18, 6-9 pm: Final Event (Hybrid): Presentations and Prize Awards

• Event Format: Finalist teams will deliver to the judging panel at the final event on April 18, 2023, 6-8 pm, held in-person in the ILSB with remote participation via zoom. Judges will review and select the winning teams based on their written reports and presentation. The competition winners will be announced at the final event, along with special team prizes. Detailed program and zoom details are to be announced.

---

Prizes

Organizing Team

• Nick Duffield (Electrical & Computer Engineering, TAMIDS)
• Jian Tao (Performance, Visualization & Fine Arts, TAMIDS)
• Diego Rodriguez (TAMIDS)
• Angie Rollins (TAMIDS)
• Jennifer South (TAMIDS)

Judges

Sponsors

TAMIDS gratefully acknowledges support for the 2023 Data Science Competition: Chevron, the Texas A&M Departments of Electrical & Computer Engineering, and Statistics, and the School of Performance, Visualization & Fine Arts.