How Michael Desjardins Turns Climate Data into Pandemic‑Prevention Lessons

Hook: Imagine checking the weather forecast and instantly seeing where the next zoonotic spillover could erupt. That’s the everyday reality for students in Michael Desjardins’ classroom at the Johns Hopkins Bloomberg School of Public Health. By treating climate data as a front-line predictor rather than background noise, he equips the next generation of epidemiologists with a crystal ball for disease emergence.

Medical Disclaimer: This article is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional before making health decisions.

1. Climate-Driven Risk Modeling: Predicting the Next Pandemic Hotspots

Michael Desjardins directly answers the question of how climate change can be turned into a predictive tool for the next pandemic by layering satellite temperature readings, wildlife migration tracks, and human population density into a single risk map.

At Johns Hopkins Bloomberg School of Public Health, Desjardins and his team download daily land surface temperature data from NASA's MODIS sensor. By comparing temperature anomalies with the 2022 paper he co-authored, they identified a 15% rise in avian influenza hotspots across Southeast Asia when average temperatures exceeded 27°C for more than ten consecutive days.

Next, the team imports bird migration routes from the Movebank database. When a species known to carry H5N1 migrates northward earlier than usual, the model flags the new stop-over sites as potential spillover zones. In 2023, this approach correctly predicted a spike in poultry deaths in Bangladesh three weeks before local officials reported any unusual mortality.

To ground the model in human risk, Desjardins layers WorldPop population grids. The result is a heat-map that highlights 42 high-risk districts where climate-driven wildlife movement overlaps with densely packed settlements. These districts currently account for 12% of global COVID-19 vaccine gaps, according to a 2022 Johns Hopkins report.

Students in his epidemiology class use the same workflow. They download the latest temperature raster, run a Python script that calculates the “spillover index,” and then present a one-page briefing that names the top three emerging-risk counties. The exercise forces them to treat climate data not as background noise but as a core predictor of disease emergence.

Key Takeaways

• Satellite temperature data can be turned into a quantifiable "spillover index."
• Combining wildlife migration with human density pinpoints where climate change meets viral exposure.
• Students learn to produce policy-ready briefings directly from climate-risk models.

Having seen the power of layered data, the class now moves on to simulate real-world decision-making under climate stress.

2. Embedding Climate Scenarios into Case-Based Learning

Desjardins forces students to confront climate stressors head-on by turning real-world heatwaves and sea-level rise events into classroom case studies.

In a 2021 semester, he introduced the "Heatwave-Linked Dengue Surge" case. The scenario began with a 3-day, 38°C heatwave in Jakarta that, according to the Indonesian Ministry of Health, increased reported dengue cases by 27% compared with the previous month. Students received a spreadsheet of daily temperature, mosquito larval counts, and hospital admissions. Their task: adjust vector-control budgets, re-allocate hospital staff, and recommend community outreach - all within a 48-hour simulation.

Another case, "Sea-Level Rise and Cholera in Bangladesh," used the 2020 projection that 0.5 meters of sea-level rise could displace 13 million people by 2050. Students examined how saltwater intrusion would compromise freshwater wells, then drafted a rapid-response plan that included mobile water purification units and a targeted oral-rehydration campaign.

Results are measurable. After the dengue case, pre-test scores on climate-adaptation concepts averaged 62%; post-test scores rose to 84%, a 22-point jump documented in the course’s annual benchmark report. Moreover, 71% of students reported that the scenario changed how they view climate data - up from 38% at the start of the semester.

These case-based modules are not static PDFs. Desjardins updates them each year with the latest IPCC temperature trajectory and the most recent WHO zoonotic disease alerts, ensuring that learning stays tethered to the evolving climate-pandemic nexus.

With a toolbox of scenarios under their belts, students are ready to collaborate across disciplines, a skill that shines in the next module.

3. Interdisciplinary Collaboration with Climate Scientists

Desjardins bridges the gap between epidemiology and climate science by creating a joint research ecosystem where students learn from both worlds.

Every spring, the Johns Hopkins Climate-Epidemiology Lab hosts a week-long seminar series co-led by climate modeler Dr. Lena Ortiz and infectious disease specialist Dr. Raj Patel. Each day features a live data-sharing session: Ortiz uploads CMIP6 climate projections, while Patel provides real-time pathogen surveillance data from the Global Health Observatory.

The collaboration produced a shared data hub on the university’s cloud platform. By July 2023, the hub stored 2.4 petabytes of climate rasters, 1.1 million wildlife occurrence records, and 3.7 million human case reports. Graduate students pull from this hub to build cross-disciplinary projects. One team created a “Climate-Adjusted Reproduction Number” (R_c) that scales the classic R₀ by local temperature and humidity. Their model predicted that in West Africa, a 2°C rise could increase the basic reproduction number for yellow fever by 0.4 on average.

Joint papers reinforce the partnership. In 2022, Desjardins and Ortiz co-authored a Lancet Planetary Health article that quantified how a 1°C temperature increase could expand the suitable habitat for the Aedes aegypti mosquito by 30% across South America. The paper has been cited 112 times and is used as a reading assignment in three different public-health programs.

Beyond publications, the lab runs a monthly "Data Sprint" where students, faculty, and external climate experts spend 48 hours cleaning, merging, and visualizing datasets. The sprint’s output - often an open-source R package or a Tableau dashboard - feeds directly into the next semester’s classroom material, creating a virtuous loop of learning and research.

Armed with interdisciplinary data, learners now step into the policy arena, as shown in the upcoming workshop module.

4. Data-Driven Policy Translation Workshops

In Desjardins’ workshops, students convert raw climate-risk outputs into bite-size policy briefs that decision-makers can actually read.

During a 2023 workshop, students received a climate-risk model that highlighted 18 U.S. counties where rising temperatures overlapped with low vaccination rates for influenza. The model showed that each 1°C increase above the historical average correlated with a 5% rise in flu-related hospitalizations, a relationship documented by the CDC in its 2022 seasonal flu report.

Students were tasked with three deliverables: a two-page policy brief, a budget scenario, and a prototype of an open-source mapping tool. The brief distilled the technical findings into three actionable recommendations: (1) allocate $2.3 million for mobile vaccination units during summer heatwaves, (2) partner with local schools to integrate climate-health curricula, and (3) launch a public-awareness campaign linking heat alerts to flu vaccination reminders.

To build the budget scenario, learners used a simple spreadsheet that multiplied projected hospital cost per flu case ($7,200 on average) by the expected increase in cases derived from the model. The resulting cost-benefit analysis showed a potential $9.8 million savings over five years if the recommended interventions were implemented.

The mapping prototype, built in Leaflet.js, allowed users to toggle layers for temperature anomalies, vaccination coverage, and hospital capacity. When the prototype was presented to the state health department, officials praised its clarity and requested a pilot rollout in the next fiscal year.

Follow-up surveys indicated that 68% of workshop participants felt more confident turning data into policy, and 42% later secured internships with health agencies that cited the workshop as a key credential.

These policy-translation skills set the stage for the final, high-energy component of Desjardins’ curriculum: the hackathon.

5. Student-Led Climate-Health Hackathons

Desjardins’ hackathons turn theory into tangible tools by giving students a 48-hour sprint to build climate-responsive health solutions.

The 2022 "Climate-Pandemic Hack" attracted 120 participants from five continents. Teams were seeded with datasets: satellite-derived heat maps, WHO disease incidence reports, and open-source climate projections from the Copernicus Climate Change Service.

Team "HeatTrace" built a dashboard that cross-referenced daily temperature spikes with COVID-19 case growth in 25 U.S. counties. Their algorithm flagged a 12-day heatwave in Phoenix that preceded a 23% surge in cases, a pattern later confirmed by the Arizona Department of Health Services. The dashboard earned a $250,000 grant from the National Science Foundation to pilot the tool in three metropolitan areas.

Another team, "MigrantShield," created a mobile app that alerts climate-displaced populations about emerging vector-borne disease hotspots. Using real-time sea-level rise data, the app pushes notifications when a coastal community’s flood risk exceeds 0.3 meters, prompting users to seek prophylactic treatment for diseases like leptospirosis.

All hackathon outputs are required to be open-source. As of 2024, the hackathon’s GitHub repository hosts 37 projects, has amassed 4,200 stars, and has been forked 1,150 times. Desjardins tracks the downstream impact: three projects have been incorporated into city health dashboards, and two have been cited in WHO technical briefs on climate-adapted surveillance.

Beyond the prizes, participants report a 35% increase in confidence when presenting climate-health solutions to non-technical audiences - a skill that aligns with the Bloomberg School’s emphasis on communication.

Having proven their chops in a sprint, students close the loop by measuring what they have learned.

6. Continuous Assessment Through Climate-Resilient Metrics

Desjardins keeps student learning measurable by applying a climate-adaptation rubric, pre-/post knowledge tests, and an annual benchmark report that can be compared across institutions.

The rubric evaluates four dimensions: (1) data integration, (2) scenario analysis, (3) policy translation, and (4) communication. Each dimension is scored on a 0-5 scale, with a total possible score of 20. In 2023, the average class score rose from 11.2 to 16.7, reflecting a 49% improvement.

Pre- and post-tests focus on core concepts such as the relationship between temperature and pathogen replication rates. For example, one test item asks students to calculate the expected increase in mosquito biting frequency when nighttime temperatures rise from 22°C to 26°C, a relationship derived from a 2020 WHO entomology study that found a 12% increase per 1°C.

The annual benchmark report compiles these metrics and compares them to peer programs at universities like Harvard T.H. Chan School of Public Health and the London School of Hygiene & Tropical Medicine. In the 2024 report, Johns Hopkins ranked in the top quartile for climate-integrated curricula, a jump from the bottom half in 2020.

To ensure global comparability, Desjardins collaborates with the Global Health Education Consortium to standardize the rubric. The consortium’s 2023 guideline recommends that any climate-health course include at least one hands-on data-fusion assignment and a policy brief, criteria that are now embedded in the course syllabus.

"Students who complete the climate-risk module are 2.3 times more likely to propose actionable public-health interventions in real-world simulations," says Desjardins in a 2023 Johns Hopkins press release.

Next, we arm readers with quick definitions and a cheat-sheet of pitfalls to avoid.

Glossary

• Satellite temperature raster: A grid of temperature values captured from space, often used in climate modeling.
• Spillover index: A numeric score that estimates the likelihood of a pathogen jumping from animals to humans.
• WorldPop: An open database that maps human population distribution at high resolution.
• CMIP6: The sixth phase of the Coupled Model Intercomparison Project, a suite of climate model simulations used worldwide.
• R₀ (basic reproduction number): The average number of secondary infections produced by one infected individual in a fully susceptible population.
• R_c (climate-adjusted reproduction number): An adaptation of R₀ that incorporates local temperature and humidity effects.

Common Mistakes to Avoid

• Treating climate data as a static backdrop. Climate variables are dynamic predictors; they should be re-run each semester with the latest satellite feed.
• Skipping the human-population layer. Without overlaying density maps, risk models miss the places where spillover actually matters.
• Ignoring uncertainty ranges. Always present confidence intervals; policymakers need to know the margin of error.
• Presenting raw code instead of clear visuals. Decision-makers respond to maps and briefings, not to lines of Python.

What background does Michael Desjardins have in climate and public health?

Desjardins earned a PhD in epidemiology from Johns Hopkins and completed a post-doctoral fellowship in climate-impact modeling. He now leads the Climate-Health Integration Lab at the Bloomberg School.

How are satellite temperature data used in his teaching