Simulating the Great Dying: Student Projects that Model Volcanic CO2, Ocean Anoxia and Recovery

Why Model the Great Dying in the Classroom?

The Permian–Triassic extinction event, often called the Great Dying, is a powerful case study because it sits at the intersection of volcanism, climate, ocean chemistry, and biological recovery. The scientific consensus points to the Siberian Traps as the main trigger: huge flood-basalt eruptions released carbon dioxide and sulfur gases, warming the planet and helping drive ocean deoxygenation and euxinia, or toxic, sulfide-rich anoxic waters. For students, this event is ideal for project-based learning because it is dramatic enough to motivate inquiry but complex enough to teach the limits of any single model. That makes it an excellent fit for data-driven investigation and classroom experimentation.

One of the best ways to learn scientific thinking is to compare a simplified simulation against real-world evidence. In this article, students will build notebook-based models of volcanic CO2 release, ocean oxygen decline, and recovery trajectories, then compare outputs to proxy datasets such as carbon-isotope excursions, extinction timing, and sedimentary indicators of low oxygen. This approach helps learners practice the same habits used by working researchers: make assumptions explicit, test sensitivity, and ask where the model breaks down. If you want a broader framework for organizing inquiry, see our guide on mapping data and collaborations like a product team, which translates neatly into team-based science projects.

Because many classrooms now rely on notebooks, reproducible workflows, and shared datasets, these projects also introduce valuable digital skills. Students can learn how to structure inputs, document parameters, and interpret uncertainty rather than treating the output as “the answer.” That mindset mirrors the reproducible approach used in workflow-based digital projects and in technical fields where reproducibility matters. In the sections below, you will find project outlines, notebook templates, comparison tables, and an FAQ designed for teachers, students, and lifelong learners.

What the Science Says About Siberian Traps, Anoxia, and Recovery

The volcanic trigger: carbon and sulfur on a planetary scale

The Siberian Traps were one of the largest volcanic provinces in Earth history, and they matter because the emissions were not just large, but prolonged. Volcanism released carbon dioxide that accumulated in the ocean-atmosphere system, and the warming effect increased weathering, altered circulation, and reduced oxygen solubility in seawater. Sulfur dioxide also played a major role by promoting acid rain and short-term cooling pulses, creating a climate system that swung between stressors rather than stabilizing. For students, this is a useful reminder that Earth systems often respond to multiple forcings at once, not one neat cause.

The proxy record suggests that atmospheric CO2 rose from roughly 400 ppm to around 2,500 ppm during the crisis, although the exact value depends on dataset and reconstruction method. That range is large enough to matter for model design: if the input is uncertain, the output should be treated as a plausible scenario, not a precise forecast. A strong classroom lesson can use this uncertainty to teach students how scientists think about evidence quality, much like careful consumers compare options in evidence-based decision frameworks. The goal is not to eliminate uncertainty, but to quantify it.

Ocean anoxia and euxinia: why marine life collapsed

Ocean anoxia means seawater with little or no dissolved oxygen; euxinia is a more severe state in which anoxic waters also become rich in hydrogen sulfide. During the end-Permian crisis, warming likely reduced oxygen solubility and altered ocean circulation, making low-oxygen conditions more widespread. In some basins, euxinic conditions may have become especially dangerous because sulfide is toxic to most complex life and can inhibit ecosystems even after temperature stress improves. Students can visualize this by simulating oxygen decline alongside a threshold for sulfide onset, then comparing that threshold to proxy indicators such as pyrite formation and sediment geochemistry.

This is also a good place to teach that marine chemistry is not a simple switch. Oxygen loss can happen gradually, but biological consequences may become nonlinear once a tipping point is crossed. That makes the Great Dying a useful entry point to systems thinking, similar to how observers learn that atmospheric seeing, telescope aperture, and mount stability interact in complex coordination problems. A student model can therefore show that a small increase in temperature or nutrient loading may produce outsized ecological effects when the system is already close to a threshold.

Recovery: why ecosystems did not bounce back quickly

Recovery after the extinction was slow because the world that emerged was not the same world that had existed before. Elevated CO2, altered weathering, ocean chemistry changes, and repeated environmental stressors made it difficult for diverse ecosystems to re-establish. In many regions, survival was possible, but ecological richness took millions of years to rebuild. Students should understand that mass extinction is not just a “die-off” followed by a reset; it is a prolonged reorganization of the biosphere.

When comparing simulations to proxy data, recovery is one of the most informative parts of the lesson. Proxy records often show that carbon-cycle perturbations outlast the initial extinction pulse, which helps explain why ecosystem rebuilding lagged behind the extinction event itself. Teachers can connect this idea to long-term planning in other domains as well, such as case studies in adaptation and recovery. In both science and society, resilience depends on how quickly systems can adapt after a shock.

Project 1: A Simple Siberian Traps CO2 Release Simulation

Learning goal and scientific question

This project asks: if volcanic carbon enters the atmosphere-ocean system in pulses, how quickly might CO2 rise, and how sensitive is the result to assumptions about emission rate and ocean uptake? Students will build a simple carbon-budget model using a time loop or differential equation approximation. The model does not need to reproduce every detail of the end-Permian world; instead, it should show how a large carbon injection can push the climate system toward a new state. That makes it an ideal first project for students practicing quantitative literacy.

Recommended outputs include a CO2 time series, a warming proxy, and a “stress index” representing environmental pressure. Students can then compare the shape of their curves to published proxy estimates. A good extension is to run multiple scenarios: one with a single rapid pulse, one with several smaller pulses, and one with slower degassing over a longer period. This helps students see that the timing of emissions can matter as much as the total amount.

Notebook outline: from assumptions to plots

A useful notebook starts with inputs: initial CO2 concentration, volcanic carbon injected per time step, ocean uptake fraction, and a simple climate sensitivity coefficient. The model can be as straightforward as:

CO2[t+1] = CO2[t] + volcanic_input[t] - uptake_rate * (CO2[t] - preindustrial_baseline)

Next, students can translate CO2 into a relative temperature anomaly using a logarithmic approximation or a linear classroom simplification. Then they can add an ecological stress metric based on temperature and oxygen decline. If your class is new to notebooks, it may help to borrow the logic of readable step-by-step systems used in technical transition guides: keep the code modular, comment every assumption, and save each run with a scenario name.

How to judge whether the model is “good”

The point is not to make the simulation perfect. The point is to ask whether it captures the broad pattern seen in proxy data: a sharp carbon rise, environmental instability, and delayed recovery. Students should compare their results to available estimates for CO2 increase and extinction timing, then note which assumptions most strongly change the curve. This is where scientific literacy becomes data literacy: learners begin to see that models are maps, not territory. For a practical lesson in comparing options without overtrusting a result, our guide to error mitigation and uncertainty control offers a helpful mindset.

Project 2: Ocean Oxygen Loss and Euxinia Thresholds

Learning goal and scientific question

This project focuses on a more biologically relevant question: under what conditions does an ocean become anoxic, and when does it cross into euxinia? Students can model oxygen supply and demand using a simple reservoir approach. The ocean gains oxygen from mixing and exchange with the atmosphere, but loses oxygen through respiration and thermal stratification. As temperature rises, oxygen solubility drops, and the system becomes more vulnerable to low-oxygen states. That interaction is the heart of the end-Permian marine crisis.

Students can simulate a shallow ocean box and test how nutrient loading or warmer temperatures affect oxygen concentration. Then they can add a sulfide threshold: once oxygen falls below a chosen level, euxinia can occur, triggering a separate ecological response. This creates a clear “before and after” structure that students can analyze without needing advanced oceanography. The framework is similar to watching how a platform behaves under load in multi-system architectures: once one component is stressed, the entire system can shift rapidly.

Notebook outline: variables and pseudo-code

A classroom-friendly notebook might include temperature, oxygen solubility, mixing rate, and organic carbon flux as inputs. A simple loop can update oxygen each year or century:

O2[t+1] = O2[t] + mixing_gain - respiration_loss - warming_penalty

Then define a threshold where, if O2 falls below a critical value, the model flags anoxia. A second threshold can indicate euxinia. Students can plot oxygen over time and shade the regions where the ocean becomes unsafe for marine life. They can also run side-by-side scenarios with different nutrient inputs to show how productivity can worsen oxygen stress even without extra CO2. This kind of controlled comparison resembles the logic of cost pattern analysis: isolate the variable, then interpret the result carefully.

Connecting code output to proxy records

Proxy data for anoxia and euxinia often come from geochemical indicators in sediments, such as redox-sensitive elements, sulfur isotopes, and organic biomarkers. Students should not expect their simplified model to recreate those proxies exactly, but they can compare trends: Does oxygen stay higher longer in the simulation than in the record? Does a faster warming run better match evidence for widespread marine collapse? When students see that a model can match one pattern but fail on another, they learn a crucial scientific lesson. Good models are constrained by multiple lines of evidence, not one graph.

Pro Tip: Ask students to write a “model limitation statement” after every run. Require them to name at least two assumptions, one proxy they matched well, and one place where the model likely oversimplified the science.

Project 3: Recovery After the Extinction Pulse

Learning goal and scientific question

Recovery projects shift the focus from catastrophe to resilience. A strong classroom question is: how long does it take for biodiversity to recover if environmental stress gradually declines, but not all at once? Students can create a simple recovery model where biodiversity rebounds only when temperature, oxygen, and carbon levels move back below stress thresholds. This lets them explore why the biosphere may remain unstable even after the worst extinction pulse ends. For educators, this is a good opportunity to connect long-term Earth change to broader systems thinking, as in data-based monitoring of complex human systems.

Students can use one curve for abiotic recovery, such as CO2 decline, and a second for biotic recovery, such as genus richness. The key insight is that the two curves need not move together. Environmental improvement can begin earlier than ecological recovery, and that lag is one of the most important lessons from the Great Dying. A simple model can show this clearly by assigning a regeneration rate that accelerates only after stress drops below a threshold.

Recovery notebook outline

The notebook can define an environmental stress index and a biodiversity index. If stress is high, biodiversity declines or remains low; if stress falls, biodiversity grows slowly. This creates a delayed rebound that students can compare to proxy evidence from the Early Triassic. Teachers can ask: Which parameter most delays recovery, and what does that imply about real-world ecosystems under climate stress? Because the output is visual and intuitive, students can learn the difference between a rapid physical change and a slow biological response. That distinction is also important in planning for real-world uncertainty, much like forecasters interpreting outliers.

Why recovery modeling matters for climate literacy

Recovery models help students understand that ecosystems have memory. Once habitats are simplified, species interactions are lost, and recovery may be slower than expected even if the original stressor weakens. This is a key idea in modern Earth science and in contemporary climate education. It teaches that resilience depends on the speed, magnitude, and duration of disruption, not simply whether the disruption eventually stops. Students who grasp this are better prepared to think about today’s environmental changes with nuance rather than alarmism or false reassurance.

Comparing Simulations to Proxy Data

What proxy data can and cannot tell us

Proxy data are indirect clues about past environments, such as isotope ratios, fossil assemblages, mineral signatures, and sediment textures. For the Great Dying, proxies help reconstruct carbon-cycle disturbance, ocean oxygen loss, and extinction timing. But proxies do not give a perfect movie of the past; they provide partial windows that must be interpreted carefully. That is why a student project should emphasize evidence triangulation rather than simple curve fitting. A helpful analogy comes from measuring impact beyond one metric: a single number rarely tells the whole story.

Students should be encouraged to ask what each proxy is actually measuring. For example, a carbon isotope excursion tells you something about carbon sources and sinks, but not directly about oxygen levels. A sulfide-related geochemical signal can support euxinia, but it may vary by basin. This is a perfect moment to teach “proxy literacy,” the skill of reading the evidence with caution. It also reinforces the idea that science advances through multiple converging lines of evidence, not one decisive record.

Suggested comparison table for classrooms

Teaching model validation as a habit

Validation should be framed as an iterative habit, not a final stamp of approval. Students can run a model, compare it to one proxy, revise an assumption, and test again. This cycle mirrors real research, where models evolve as new data arrive. It also helps students see why disagreement between model and proxy is productive: it reveals where the science is still being refined. In classrooms, that “productive mismatch” is often where the best learning happens.

Classroom Implementation: Lesson Plans, Roles, and Differentiation

A three-day project sequence

A simple classroom sequence could begin with a background lesson on the Great Dying, followed by a guided notebook activity, and end with student presentations. Day one should focus on the Earth-system story: volcanism, carbon cycle disruption, ocean anoxia, and recovery. Day two can be the coding and simulation day, where students run base-case and alternative scenarios. Day three should emphasize comparison, interpretation, and uncertainty statements. This sequence works well in science classes, data science clubs, and interdisciplinary environmental units.

To make the project manageable, divide students into roles: one group handles volcanic forcing, another handles ocean oxygen, and a third handles proxy comparison. That structure keeps the workload balanced while reinforcing collaboration. If you want an organizational model for roles, versioning, and responsibilities, the logic in team mapping and case-study style iteration can be adapted to science classrooms. Roles also make it easier for teachers to differentiate instruction.

Differentiating for beginners and advanced students

For beginners, provide a partially completed notebook with editable parameters and prewritten plotting code. Let them change one variable at a time and describe the result in plain language. For advanced students, ask them to modify the model structure itself, perhaps by adding multiple volcanic pulses or a delayed recovery function. They can also compare linear versus nonlinear responses to make the concept of thresholds more concrete. This layered approach helps all learners succeed without reducing rigor.

Teachers may also want to connect the project to digital citizenship and reproducibility. Students should cite their sources, label graphs, and save their code with clear filenames. Those practices echo the habits used in structured digital workflows, such as repeatable templates and expert-informed methods. Good scientific communication is part of the skill set, not an extra.

Notebook Design: What to Include in a Student Repository

Recommended files and structure

A clean repository helps students focus on the science rather than the file maze. At minimum, include a README, a data folder, a notebook for each simulation, and a short reflection worksheet. If possible, add a simplified dataset of proxy values in CSV format so students can compare output without hunting for sources. This is also a good opportunity to practice version control habits and clear file organization, which are foundational in research and industry alike. The same logic appears in practical guides to moving from general skills to specialized workflows.

Suggested notebook sections

Each notebook should include an introduction, assumptions, code cells, plots, and a conclusion. Students should annotate each step in plain language: what the variable means, why they chose a value, and how changing it might alter the result. End with a “What this model does not include” section. That limitation section is vital because it prevents the notebook from being mistaken for a full reconstruction of the end-Permian world. It also models honest scientific writing.

Assessment ideas

Grade the project on scientific reasoning, not just code correctness. A strong student submission should explain the relationship between the simulation and proxy evidence, identify at least one sensitivity, and discuss one limitation in detail. Teachers can use a rubric that rewards clarity, evidence use, and interpretation over flashy graphics. If you want to borrow an assessment mindset from applied fields, the structured comparison habits in data-driven analysis and impact measurement are surprisingly transferable.

Common Mistakes Students Make When Simulating Deep-Time Events

Confusing correlation with causation

One common mistake is assuming that because volcanic eruptions and extinction happen near the same time, the eruption must explain every effect at the same scale and pace. In reality, the causal chain includes emissions, warming, ocean circulation changes, and ecological thresholds. Students should be encouraged to build causal diagrams before they code. This helps them distinguish direct forcing from downstream consequences. It also keeps the model conceptually honest.

Using too much precision

Another mistake is reporting outputs with unrealistic certainty, such as CO2 = 2,487.3 ppm in a model that is built on coarse assumptions. Students should round appropriately and use ranges when possible. Teachers can emphasize that paleoclimate models are best interpreted as scenarios with uncertainty bands, not exact historical reconstructions. That habit mirrors how professionals manage complexity in fields from weather forecasting to systems engineering. The same caution is helpful when interpreting outliers in forecasting contexts.

Ignoring the role of proxies

Students sometimes stop after generating a pretty graph. The deeper lesson, however, is the comparison to proxy evidence. Without that comparison, the model stays abstract and disconnected from the scientific record. Encourage students to ask what sediment, isotope, or fossil evidence would support or challenge their result. That habit turns the activity from a coding exercise into a real science investigation.

Pro Tip: Ask students to identify one parameter they believe is most uncertain, then run a sensitivity test with low, medium, and high values. This single activity often reveals more about scientific reasoning than a long lecture.

Conclusion: Turning the Great Dying Into a Modern Science Lab

The Great Dying is a devastating event in Earth history, but it is also one of the best teaching tools we have for systems thinking, modeling, and evidence-based reasoning. A classroom simulation of Siberian Traps CO2 release, ocean anoxia, and euxinia lets students do more than memorize a mass extinction story. It invites them to test assumptions, compare models to proxy data, and understand why uncertainty is not a weakness but a feature of real science. That is a powerful lesson for any student of environment and space science.

When learners model the Great Dying, they practice the exact habits scientists use to study Earth under stress: build a hypothesis, define variables, compare with evidence, and revise. They also see that recovery can be slow, nonlinear, and incomplete. In a world dealing with contemporary climate change, those insights matter far beyond paleontology. For more on building a practical, evidence-first learning habit, explore our guide to authenticity and audience trust and our perspective on iterative learning from case studies.

Related Reading

• How Airports Coordinate With Space Agencies During Reentries and Rocket Launch Windows - A systems-thinking look at coordinating complex operations under uncertainty.
• Interview With Innovators: How Top Experts Are Adapting to AI - Useful for thinking about reproducible workflows and expert methods.
• From IT Generalist to Cloud Specialist: A Practical Roadmap for Platform Engineers - A helpful analogy for moving from simple to more advanced model structures.
• Why Great Forecasters Care About Outliers—and Why Outdoor Adventurers Should Too - A strong companion piece on uncertainty, extremes, and interpretation.
• How to Use Branded Links to Measure SEO Impact Beyond Rankings - A practical reminder that multiple metrics matter when judging success.