Graduate Paper Presentation & Best Practices Guide

Graduate Paper Presentation & Best Practices Guide#

ESS 512 - Introduction to Seismology Graduate Component

Overview#

Graduate students (ESS 512) are required to present one research paper that demonstrates the application of computational methods learned in this course to real seismological research. This assignment develops critical skills in:

  • Reading and understanding research literature

  • Evaluating methodology and reproducibility

  • Scientific communication

  • Connecting theory to practice

Weight: 15% of final grade Format: 15-minute presentation + 5 minutes Q&A Distributed throughout: Weeks 5-9 (one student per week)

Learning Objectives#

By completing this assignment, you will:

  1. Connect coursework to research: Understand how methods from our computational labs are used in published research

  2. Critical evaluation: Assess the strengths and limitations of computational methods in real applications

  3. Reproducibility: Evaluate whether the research could be reproduced from the information provided

  4. Communication: Present complex technical material clearly to peers

  5. Best practices: Learn standards for publication-quality figures, code documentation, and method description

Paper Selection Guidelines#

Approved Topics#

Your paper should connect to at least one of our computational modules:

Module

Example Research Topics

Ray Tracing & Travel Times

Earthquake location algorithms, seismic tomography methods, phase identification

Surface Waves

Regional dispersion analysis, ambient noise tomography, crustal thickness studies

Fourier Analysis

Spectral analysis methods, attenuation (Q) measurement, time-frequency analysis

Data Processing

Network processing, data quality metrics, instrument response

Noise Cross-Correlation

Green’s function extraction, velocity monitoring, noise sources

Selection Criteria#

Your paper should:

  • Be published in a peer-reviewed journal (last 10 years preferred)

  • Include computational seismology methods

  • Have clear methodology section

  • Include code/data availability statement (bonus)

  • Be accessible through UW Libraries

Good Journals:

  • Seismological Research Letters (SRL)

  • Bulletin of the Seismological Society of America (BSSA)

  • Geophysical Journal International (GJI)

  • Journal of Geophysical Research: Solid Earth (JGR)

  • Geophysical Research Letters (GRL)

  • Earth and Planetary Science Letters (EPSL)

Paper Approval Process#

  1. Week 3: Submit 3 paper candidates to instructor via canvas

    • Include full citations

    • Brief (2-3 sentence) description of why each interests you

    • Which computational module(s) it connects to

  2. Week 4: Instructor approves paper and assigns presentation week

  3. One week before presentation: Send slides to instructor for feedback

Presentation Structure#

Time Allocation (15 minutes total)#

  • Introduction (2 min): Scientific motivation and research question

  • Data & Methods (5 min): FOCUS HERE - Computational approach, algorithms, processing

  • Results (4 min): Key findings

  • Discussion & Critical Evaluation (3 min): Strengths, limitations, reproducibility

  • Conclusion (1 min): Main takeaway and connection to course

Required Content#

Your presentation must include:

1. Research Context (Brief)#

  • What is the scientific question?

  • Why does it matter?

  • What was previously unknown?

2. Data Description#

  • What seismic data was used? (Network, stations, events, time period)

  • Data quality considerations

  • How was data accessed/processed?

3. Computational Methods (CORE - spend most time here)#

  • Algorithm(s) used - explain in detail

  • Processing workflow (flowchart helpful)

  • Key parameters and their selection

  • How does this connect to methods we learned in class?

  • What software/tools were used?

4. Results#

  • Main findings (select 2-3 key figures)

  • Uncertainty/error analysis

  • Validation approach

5. Critical Evaluation#

  • Reproducibility Assessment:

    • Is code/data available?

    • Could you reproduce the results from the paper?

    • What information is missing?

  • Methodological Strengths:

    • What did they do well?

    • Novel aspects?

  • Limitations:

    • Assumptions that may not hold?

    • Alternative approaches?

    • How could methods be improved?

6. Connection to Course#

  • Which lectures relate to this work?

  • What extensions could we implement in class?

  • What did you learn that enhances your understanding of course material?

Publication Best Practices Analysis#

As part of your presentation, evaluate the paper against modern publication standards:

1. Reproducibility Checklist#

Assess whether the paper includes:

  • [ ] Code Availability: Link to GitHub, Zenodo, or supplementary materials

  • [ ] Data Availability: DOI or permanent archive link

  • [ ] Software Versions: Specific versions of tools used (e.g., ObsPy 1.4.0)

  • [ ] Parameter Documentation: All processing parameters clearly stated

  • [ ] Workflow Description: Step-by-step processing flow

  • [ ] Random Seed: If applicable (for MC methods, inversions)

Rate: Poor / Fair / Good / Excellent

2. Figure Quality Assessment#

Evaluate the main computational figure (choose one):

  • [ ] Axes Labels: Clear, with units

  • [ ] Colorbar: If applicable, with label and units

  • [ ] Font Size: Readable when printed

  • [ ] Caption: Self-contained (can understand without reading text)

  • [ ] Error Bars: Shown where appropriate

  • [ ] Legend: Clear identification of all elements

  • [ ] File Format: Vector (PDF/SVG) for line plots?

Example of Excellent Practice: Include a screenshot showing good vs bad

3. Method Documentation#

Evaluate the methods section:

  • [ ] Algorithm Description: Mathematical formulation included

  • [ ] Pseudocode/Flowchart: If complex algorithm

  • [ ] Trade-offs Discussed: Why this method over alternatives?

  • [ ] Failure Cases: When does the method fail?

  • [ ] Computational Cost: Runtime, memory requirements mentioned

  • [ ] Validation: Synthetic tests or comparison with known results

Rate: Poor / Fair / Good / Excellent

4. Open Science Practices#

Does the paper follow open science principles?

  • Open Access: Is the paper freely available?

  • Open Data: Are datasets in public archives (IRIS, etc.)?

  • Open Source: Is code freely available?

  • Permissive License: MIT, BSD, GPL, or similar?

  • Community Standards: Uses standard formats (SAC, miniSEED)?

Discuss: How does this paper compare to ideal open science standards?

Deliverables#

1. Presentation Slides#

Submit one week before: PDF of slides to instructor for feedback

Format Requirements:

  • 12-16 slides maximum (excluding title/references)

  • Title slide: Paper citation, your name, date

  • Clear slide titles

  • Not text-heavy (use figures, diagrams, bullets)

  • Readable fonts (≥18pt for body text)

  • Final slide: 2-3 discussion questions for class

Include:

  • At least one slide showing code snippet or algorithm pseudocode

  • At least one slide with reproducibility checklist results

  • References slide (cite paper + any additional sources)

2. One-Page Summary#

Due day of presentation

Content:

  • Paper citation

  • 3-4 sentence summary of paper

  • Key computational method(s) used

  • Reproducibility score (1-5) with justification

  • Connection to course module(s)

  • One limitation or improvement you would suggest

  • One question for further investigation

Format: PDF, single page

Rubric (100 points total)#

Content (60 points)#

Component

Points

Criteria

Methods Explanation

20

Clear description of algorithms; connection to course; technical accuracy

Critical Evaluation

15

Thoughtful assessment of reproducibility, strengths, and limitations

Scientific Context

10

Motivation clear; results summarized appropriately

Best Practices Analysis

15

Thorough evaluation of code/data availability, figures, documentation

Presentation Skills (25 points)#

Component

Points

Criteria

Clarity

10

Logical flow; concepts explained clearly; appropriate level for audience

Time Management

5

Stays within 15 minutes; appropriate pacing

Slides

5

Readable, well-organized, good visuals

Q&A Response

5

Thoughtful answers; admits uncertainty appropriately

Written Summary (15 points)#

Component

Points

Criteria

Completeness

10

All required elements included; accurate summary

Writing Quality

5

Clear, concise, professional

Example Papers (as inspiration)#

Surface Waves & Ambient Noise#

  • Bensen et al. (2007). “Processing seismic ambient noise data to obtain reliable broad-band surface wave dispersion measurements.” GJI 169(3), 1239-1260.

    • Why good: Extremely detailed methods; became community standard

  • Lin et al. (2008). “Surface wave tomography of the western United States from ambient seismic noise: Rayleigh and Love wave phase velocity maps.” GJI 173(1), 281-298.

    • Why good: Clear workflow; connects noise to structure

Ray Tracing & Tomography#

  • VanDecar & Crosson (1990). “Determination of teleseismic relative phase arrival times using multi-channel cross-correlation and least squares.” BSSA 80(1), 150-169.

    • Why good: Algorithm details; widely used method

  • Rawlinson & Sambridge (2004). “Wave front evolution in strongly heterogeneous layered media using the fast marching method.” GJI 156(3), 631-647.

    • Why good: Clear algorithm description; synthetic tests

Fourier Analysis & Attenuation#

  • Lawrence & Prieto (2011). “Attenuation tomography of the western United States from ambient seismic noise.” JGR 116(B6).

    • Why good: Combines several methods; good validation

Data Quality#

  • Ringler et al. (2015). “Seismic station installation orientation errors at ANSS and IRIS/USGS stations.” SRL 86(3), 926-931.

    • Why good: Practical problem; straightforward analysis

Tips for Success#

Reading the Paper#

  1. First pass: Skim for main idea (10 min)

    • Abstract, figures, conclusions

    • Identify computational methods

  2. Second pass: Read in detail (1-2 hours)

    • Methods section carefully

    • Try to understand each figure

    • Look up unfamiliar terms/methods

  3. Third pass: Critical reading (1 hour)

    • Could you reproduce this?

    • What’s not explained?

    • Are conclusions justified?

Preparing Presentation#

  1. Start with methods: Build presentation around computational approach

  2. Use paper’s figures: Okay to screenshot (with citation) - don’t recreate

  3. Practice timing: Rehearse at least twice

  4. Anticipate questions: Prepare for what classmates might ask

  5. Connect explicitly: Make course connections obvious

Common Pitfalls to Avoid#

  • ❌ Spending too much time on background/introduction

  • ❌ Just summarizing results without explaining methods

  • ❌ Not evaluating reproducibility

  • ❌ Reading dense text from slides

  • ❌ Skipping connection to course material

  • ❌ Exceeding time limit

What Makes an Excellent Presentation#

  • ✅ Clear explanation of algorithms (could we implement it?)

  • ✅ Critical but fair evaluation

  • ✅ Engaging visuals (flowcharts, diagrams, not just text)

  • ✅ Specific examples from the paper

  • ✅ Thoughtful discussion questions

  • ✅ Explicit course connections

Resources#

Finding Papers#

Evaluating Reproducibility#

Presentation Skills#

Presentation Schedule#

Presentations will occur in Weeks 5-9 (one student per week). Schedule determined after paper approval.

Week

Date

Student

Paper Topic

Connection

5

TBD

Student 1

TBD

TBD

6

TBD

Student 2

TBD

TBD

7

TBD

Student 3

TBD

TBD

8

TBD

Student 4

TBD

TBD

9

TBD

Student 5

TBD

TBD

Questions?#

Contact instructor during office hours or via email. Start early - finding the right paper and understanding it deeply takes time!

Last Updated: January 2026

Contents