Reinvigorating Computational & Quantitative undergraduate Curricula for the Earth, Ocean, Atmospheric, Environmental and Planetary Sciences at UBC

https://agu2021fallmeeting-agu.ipostersessions.com/default.aspx?s=2A-C2-3D-E0-3B-DD-9A-5E-A3-C9-44-74-4B-06-45-A9

1. Motivation and context Research & Development Question How best to adapt programs, courses & practices to meet changing needs for Quantitative Earth Sciences (QES)? ~~~ Two projects: 3 years, well-funded by UBC: OCESE: Opensource Computing for Earth Sci. Education QuEST: Quantitative Earth Sciences Transformation Why now? · Geoscience is becoming more quantitative and is growing in importance (refs 1, 2, 3, especially AGI "Vision and Change in the Geosciences", Peers & colleagues) · Meanwhile - declining ugrad enroll’s across the geosciences (source): Therefore: quantitative geoscience degree programs and courses need to be revisited and renewed to ... Enhance relevance Increase enrollments of appropriate students Showcase potential of corresponding careers Readiness - OCESE: 13 faculty co-investigators, 19+ courses - QuEST: 22 faculty co-investigators, - Five quantitatively oriented degree programs offered in EOAS: Atmospheric sciences (~14 students / yr) Geophysics (~9 students / yr) Environmental sciences (~55 students / yrr) Oceanography (~30 students / yr) Geological engineering (~45 students / yr) - UBC's Dep't Earth, Ocean and Atmospheric Sciences 45 Research/teaching faculty 12 Educational leadership faculty ~190 Ugrads in 2020 88 & 78  MSc & PhD students Earth, Ocean & Atmospheric Sciences - EOAS ~ ~ ~ At AGU Fall 2021: Via links, bottom right of this poster . . . Contribute your perspectives and priorities. Discuss with us via contacts. Take the short survey below. Contribute to the discussion Contribute a few minutes towards the conversation about quantitative B.Sc. Earth science degree programs: Undergraduate Quantitative Earth Sciences - Survey Project & Department Activities Colleagues: conversations with EOAS. Partnering with other UBC science departments: With Statistics: build a 1st year datascience course section; Earth science context, python-based. Partnering in 1st / 2nd yr courses (in progress). (Also, math department is rejuvenating Calculus I.) Peers & Alumni Discussions about the evolving QES discipline and hiring practices. Podcasts (by PME) with professionals, students, community leaders (60+). Peer institutions Discussions about evolving QES program & course offerings and programming options. Public: Contribute QES content in our Pacific Museum of Earth, schools, summer science camps. ~ ~ ~

3. Curriculum: progress QuEST project goals: · Recommend curricular changes to meet the needs of students who will work in emerging quantitative Earth science (QES) occupations. · Attract and inspire appropriate students to pursue QES degress or take QES courses. Progress towards two goals: 1. Background & current state What data will inform recommendations about how to meet emerging needs? * Colleagues, peer institutions & hiring trends * Current UBC course dependencies Online visualization helps find gaps & opportunities. * Current QES content: Number of courses (y-axis) using each math, physics or computing concept (x-axis). * Student insights: Past & present geophysics students: Likes:   >small;    >faculty support & expertise;             >emphasis on fundamentals. Recommendations:    >review relevancy of pre-requisites;    >too heavy on global, short on applied. 2. QES "marketing" What actions can make QES more visible to prospective students and the public? New 1^st yr course ("Climate Physics"?) to be developed, targeting "quantitative" students. Renovated geophysics degree requirements; More versatile prerequisite options (link). UBC's 1^st yr datascience course (GitHub) will have a newly developed, python-based section emphasizing Earth science contexts. EOAS web is increasing the visibility of QES options and opportunities. Updated co-op workterm EOAS information. Updated careers info. & advice. Podcasts, videos, spotlights on faculty, research, students, community, and more (partner: PME). New, industry-sponsored field station. ~ ~ ~

4. Courses: progress OCESE project goals: Students: Enhance computational, numerical & data science knowledge, skills and attitudes. Faculty: Support for introducing quantitative, data-rich resources and pedagogies. Resources: Sustainable infrastructure, resources, and tactics. Four areas of progress: 1. Quantitative teaching & learning Transform Intro Computing to Python (from MatLab); >pedagogy, >learning activities >text, >JupyterHubs, >autograding: All piloted Fall 2021, 100 students. Transform Earth Science timeseries analysis to JNotebooks and Python (was on paper & MatLab). Upper level coures: establish Python as the preferred language so students can grow & practice skills. Results: Data collection in progress. Evaluating impacts of: >Python, >J-Noteboks, >nbgrader, >Earth Sci. context, >cloud servers >costs, >sustainabiltiy. Open Ed Resources: Jupyter Notebooks (assigs & labs), Jupyter Books (texts, labs), Python, dashboards, infrastructure tactics, new pedagogies. 2. Technology and infrastructure Students to learn & use Python across the Department. Why? It is readily learned, opensource, widely used in QES, versatile, in-demand. ( I.E.E.E. ) Jupyterhubs: deploy to local or commercial data centers; scalable. (Piloted in 2 courses, 138 students.) Lightweight dashboards to interactively explore concepts & data in classes & assignments (12 so far). Containerizing for reliable, sustainable deployment on scalable servers. Jupyter Notebooks to facilitate and "standardize" quantitative, datascience & computational learning. Autograding: "nbgrader" is a JNB extension; a collaboration with UBC Statistics, & others. Jupyter Books: intereactive, open source, readily maintained by faculty & students. 3. Interactive QES resources; dashboard learning apps Each explores a focused concept and/or dataset. Example1, interactive climate factors. Example2: groundwater flow with recharge. 12 so far (list); more to come. Students do much of the programing. Eight piloted with >500 students. Five courses: hydrogeology, climate science, intro. Earth sci., intro oceanography, palaenontology. Results; Students They engage with concepts. Feedback is positive Student feedback (N=107). Example from an introductory oceanography course , Dec 3 (Dashboard here): I liked how the dashboard exercise applied various concepts using real data from the world's oceans. The online dashboard was quite easy to use and interesting as it allowed us to compare different nutrients/properties profiles in different oceans, synthesizing all the concepts in the course. Results; instructors: 1-on-1 support is required at this stage, including: design / build resources, discuss pedagogy, review feedback & results and iterate. The team's contributions are appreciated - eg (paraphrased): "I am so impressed ... I love how (a) sliders constrain and adjust axes, (b) data at various real stations can be chosen on a map and compared, (c) results can be saved to submit for assessment. I agree that now is the time to finalize an assignment, so thank you to the team! 4. Professional development for instructors & TAs Guidance & support: ideas ⇒ plans ⇒ build ⇒ pilot ⇒ iterate Planned workshops evolved into 1on1 collaborations. Each course is unique. Each instructor has specific needs, interests, priorities and technical / pedagogical skills. TAs supporting Profs - always effective & impactful. Hired students: developing dashboards, resources, assignments, assessments & infrastructure. Science Education Specialist (40%fte) to coordinate, advise on teaching & learning, initiate specific ideas, etc.

5. Ongoing curriculum activities Current priorities 1. Background leading to recommendations Gather learning goals, syllabi, assessments & activities Generate consistent course learning outcomes (CLOs), find commonalities, establish opportunities. Establish academic and employers' priorities. Complete a formal recommendations document. 2. Attract students; enhance visibility New 1st yr course (likely "Climate Physics") Engage directly with incoming and first-year students. Community engagement - see 2. below left. >Alumni,  >peers,  >employers,  >public,  >schools. 3. Increase QES content in elective courses Climate and data science is being introduced into electives for science or non-science students. Theory vs. applied: Revisit balance between theoretical and applied math, physics & computer science. Opportunities vs Risks Discussions are ongoing about . . . QES content: breadth vs depth (content & skills) existing vs new skills (eg, R or Python) top down vs bottom up learning frameworks QES pedagogy: Existing vs new techniques (eg using GitHub for content development and delivery) Existing vs emerging skills (eg Python or MatLab) Teaching: lecture vs activity-driven learning Assessing higher level or open-ended student products ~ ~ ~ 6. Ongoing course development Increasing computing capacity in courses 1. Enhance QES courses by improving ... Infrastructure: Chosen technologies so far: >Traefik, >Docker, >Plotly-Dash, >Conda-lock, >Jupyter Books, >Jupyter notebooks, >nbgrader, >GitHub, >Canvas, Sustainability: need to resolve who maintains (or pays for) scalable cloud computing resources? How do we respond to evolving and new technology?  New relevant, sophisticated learning tasks using dashboards to interact with data and concepts. Evaluation of effectiveness. Feedback via surveys, interviews with insrtructors, teaching assistants, students. Observations of classes, learning activities, assignments and assessments. Professional development for instructors and TAs: guidelines, best practices for support and maintenance 2. Upcoming project components: Additional faculty: 8 to engage next year. Additional courses: 8-10 to participate next year. Python & Earth Sci. context for UBC's first year Data Science Intro course (~100 students anticipated). New collaborations: faculty and grad students inspired by early activities and results. Partnering with UBC climate education grant recipients; courses at all levels. Determining resources & pedagogies with best potential to support learning across disciplines. Opportunities or Challenges Students respond eagerly to engaging with real datasets and meaningful problems while developing modern, widely useful skills (Python, open-source software, "big" data, climate, etc.) Faculty researchers are "over extended"; they need staff, student and educational support / expertise. Institutions are less agile than students. Infrastructure and personnel to supply and maintain necessary resources and capabilities are a "cost". Open source vs commercial vs "local": ongoing discussions about costs, benefits & options. Choices depend on level (course, department, institution, etc.), faculty buy-in, and curricular needs of students.

Poster at https://agu2021fallmeeting-agu.ipostersessions.com/?s=choose. See also OneNote notes from “how to” poster-system webinar, Nov 17.

References

1. Why – society, industry & student needs & changing disciplinary needs (geoscience & bioscience)

1)    Keane, Christopher M., and Carolyn E. Wilson. “The Mid-21st Century Geophysics Workforce: How Today’s Trends across Geoscience Impact Geophysics Human Resources of the Future.” In SEG Technical Program Expanded Abstracts 2018, 4829–33. SEG Technical Program Expanded Abstracts. Society of Exploration Geophysicists, 2018. https://doi.org/10.1190/segam2018-2992425.1. (Geophysics industry)

2)    Mosher, Sharon, and Christopher Keane. “Vision and Change in the Geosciences: The Future of Undergraduate Geoscience Education.” American Geosciences Institute, 2021. https://www.americangeosciences.org/change/print-edition/. (Key geoscience reference)

3)    Summa, Lori, Christopher Keane, and Sharon Mosher. “Meeting Changing Workforce Needs in Geoscience with New Thinking about Undergraduate Education.” GSA Today, September 1, 2017, 58–59. https://doi.org/10.1130/GSATG342GW.1.

4)    Dill-McFarland, Kimberly A., Stephan G. König, Florent Mazel, David Oliver, Lisa M. McEwen, Kris Y. Hong, and Steven J. Hallam. “An Integrated, Modular Approach to Data Science Education in the Life Sciences.” BioRxiv, July 27, 2020, 2020.07.25.218453. https://doi.org/10.1101/2020.07.25.218453.

5)    Batchelor, R. L., H. Ali, K. G. Gardner-Vandy, A. U. Gold, J. A. MacKinnon, and P. M. Asher (2021), “Reimagining STEM workforce development as a braided river”, Eos, 102, 19 April 2021, https://doi.org/10.1029/2021EO157277.

2. Computing and quantitative literacy, curriculum & teaching in geoscience: background, scholarship and spreading quantitative thinking across geoscience curriculum.

6)    Guzdial, Mark. “Computing Education as a Foundation for 21st Century Literacy.” In Proceedings of the 50th ACM Technical Symposium on Computer Science Education, 502–3. Minneapolis MN USA: ACM, 2019. https://doi.org/10.1145/3287324.3290953.

7)    Tenenberg, Josh, and Sally Fincher. “Opening the Door of the Computer Science Classroom: The Disciplinary Commons.” ACM SIGCSE Bulletin 39, no. 1 (March 7, 2007): 514–18. https://doi.org/10.1145/1227504.1227484.

8)    Macdonald, R. H, and C. M. Bailey. “Integrating the Teaching of Quantitative Skills across the Geology Curriculum in a Department.” Journal of Geoscience Education 48, no. 4 (2000): 482–86.

9)    Sangwin, Christopher J., and Claire O’Toole. “Computer Programming in the UK Undergraduate Mathematics Curriculum.” International Journal of Mathematical Education in Science and Technology 48, no. 8 (November 2, 2017): 1133–52. https://doi.org/10.1080/0020739X.2017.1315186.

10)    Jacobs, Christian T., Gerard J. Gorman, Huw E. Rees, and Lorraine E. Craig. “Experiences With Efficient Methodologies for Teaching Computer Programming to Geoscientists.” Journal of Geoscience Education 64, no. 3 (August 19, 2016): 183–98. https://doi.org/10.5408/15-101.1.

3. Scholarly, evidence-oriented approaches & best practices.

11)  Margulieux, Lauren, Tuba Ayer Ketenci, and Adrienne Decker. “Review of Measurements Used in Computing Education Research and Suggestions for Increasing Standardization.” Computer Science Education 29, no. 1 (January 2, 2019): 49–78. https://doi.org/10.1080/08993408.2018.1562145.

12)  Dorn, Brian, and Allison Elliott Tew. “Empirical Validation and Application of the Computing Attitudes Survey.” Computer Science Education 25, no. 1 (January 2, 2015): 1–36. https://doi.org/10.1080/08993408.2015.1014142.

13)  Green, David. “Pair Programming: Benefits, Tips & Advice for Making It Work — SitePoint,” January 21, 2020. https://www.sitepoint.com/pair-programming-guide/.

14)  Alammary, Ali. “Blended Learning Models for Introductory Programming Courses: A Systematic Review.” PLoS ONE 14, no. 9 (January 1, 2019): e0221765. https://doi.org/10.1371/journal.pone.0221765.

4. Trends towards open source for addressing “pervasive” challenges, with autograding as one example.

15)  Weiss, Charles J. “A Creative Commons Textbook for Teaching Scientific Computing to Chemistry Students with Python and Jupyter Notebooks.” Journal of Chemical Education 98, no. 2 (February 9, 2021): 489–94. https://doi.org/10.1021/acs.jchemed.0c01071.

16)  Manzoor, Hamza, Amit Naik, Clifford A. Shaffer, Chris North, and Stephen H. Edwards. “Auto-Grading Jupyter Notebooks.” In Proceedings of the 51st ACM Technical Symposium on Computer Science Education, 1139–44. SIGCSE ’20. Portland, OR, USA: Association for Computing Machinery, 2020. https://doi.org/10.1145/3328778.3366947.

17)  (Lessons from COVID) Furman, T., and M. Modwin. “Higher Education During the Pandemic: Truths and Takeaways.” Eos. Accessed July 22, 2021. https://eos.org/opinions/higher-education-during-the-pandemic-truths-and-takeaways.