Will atmospheric scientists be replaced by AI?
Updated
Atmospheric scientists score 54 out of 100 for AI exposure. AI helps with 78% of their task time, and they stay in charge. AI can do another 14%. 8% needs a person.
- 14% AI can do it
- 78% AI helps
- 8% Needs a person
High exposure
Higher than 83% of the 923 jobs we scored.
Task by task
We graded 27 tasks, most exposed first. AI can do 3 of them, helps with 21 and leaves 3 to people.
| Task | AI exposure |
|---|---|
| Develop computer programs to collect meteorological data or to present meteorological information. AI coding tools write most data collection and display scripts. | 70% AI can do it |
| Prepare weather reports or maps for analysis, distribution, or use in weather broadcasts, using computer graphics. Graphics software and AI generate maps; staff check accuracy. | 70% AI can do it |
| Gather data from sources such as surface or upper air stations, satellites, weather bureaus, or radar for use in meteorological reports or forecasts. Sensor and satellite feeds are already automated; AI handles QC. | 65% AI can do it |
| Prepare scientific atmospheric or climate reports, articles, or texts. AI drafts most report text; scientists verify facts and claims. | 65% AI helps |
| Prepare forecasts or briefings to meet the needs of industry, business, government, or other groups. AI drafts tailored briefings from model output; staff review them. | 58% AI helps |
| Create visualizations to illustrate historical or future changes in the Earth's climate, using paleoclimate or climate geographic information systems (GIS) databases. AI writes visualization code quickly; interpretation remains with scientists. | 58% AI helps |
| Analyze historical climate information, such as precipitation or temperature records, to help predict future weather or climate trends. AI scans long records and flags trends; experts judge significance. | 55% AI helps |
| Analyze climate data sets, using techniques such as geophysical fluid dynamics, data assimilation, or numerical modeling. AI assists code and data handling; dynamics expertise guides interpretation. | 50% AI helps |
| Develop or use mathematical or computer models for weather forecasting. AI speeds model code; experts still design and validate forecasts. | 45% AI helps |
| Develop and deliver training on weather topics. AI drafts training outlines and quizzes; instructors deliver sessions. | 45% AI helps |
Show the other 17 tasks
| Task | AI exposure |
|---|---|
| Conduct wind assessment, integration, or validation studies. AI speeds wind data processing; validation judgement remains with experts. | 45% AI helps |
| Conduct numerical simulations of climate conditions to understand and predict global or regional weather patterns. AI helps write simulation code; physics choices stay with experts. | 45% AI helps |
| Estimate or predict the effects of global warming over time for specific geographic regions. AI speeds scenario analysis; regional projections need expert review. | 45% AI helps |
| Interpret data, reports, maps, photographs, or charts to predict long- or short-range weather conditions, using computer models and knowledge of climate theory, physics, and mathematics. AI aids pattern reading; forecasters still make the final call. | 40% AI helps |
| Formulate predictions by interpreting environmental data, such as meteorological, atmospheric, oceanic, paleoclimate, climate, or related information. AI can summarize environmental data; scientists verify predictions. | 40% AI helps |
| Apply meteorological knowledge to issues such as global warming, pollution control, or ozone depletion. AI helps synthesize literature; policy and impact judgement stay expert. | 40% AI helps |
| Research the impact of industrial projects or pollution on climate, air quality, or weather phenomena. AI speeds literature and data analysis; causal conclusions need scientists. | 40% AI helps |
| Conduct meteorological research into the processes or determinants of atmospheric phenomena, weather, or climate. AI helps literature and analysis; research design stays with scientists. | 35% AI helps |
| Broadcast weather conditions, forecasts, or severe weather warnings to the public via television, radio, or the Internet or provide this information to the news media. AI drafts scripts; live broadcast delivery needs a person. | 35% AI helps |
| Teach college-level courses on topics such as atmospheric and space science, meteorology, or global climate change. AI drafts lectures and grading aids; teaching stays in person. | 35% AI helps |
| Consult with other offices, agencies, professionals, or researchers regarding the use and interpretation of climatological information for weather predictions and warnings. AI can summarize guidance; consultations need human expertise and trust. | 25% AI helps |
| Speak to the public to discuss weather topics or answer questions. Chatbots answer simple questions; live public discussion stays human. | 25% AI helps |
| Design or develop new equipment or methods for meteorological data collection, remote sensing, or related applications. AI helps design code; building and testing hardware stays physical. | 25% AI helps |
| Perform managerial duties, such as creating work schedules, creating or implementing staff training, matching staff expertise to situations, or analyzing performance of offices. AI can draft schedules and reports; staffing decisions stay managerial. | 22% AI helps |
| Direct forecasting services at weather stations or at radio or television broadcasting facilities. Directing operations means staffing and live decisions with people. | 18% Needs a person |
| Measure wind, temperature, and humidity in the upper atmosphere, using weather balloons. Balloon launches require physical handling and field operations. | 5% Needs a person |
| Collect air samples from planes or ships over land or sea to study atmospheric composition. Sampling air from aircraft and ships needs physical presence. | 4% Needs a person |
How atmospheric scientists compare
All jobs (923)
Life, physical and social science (60)
Pay and outlook in the US
- Median pay
- $99,070 a year, May 2025
- People employed
- 10,000 in the US, May 2025
- Projected change in jobs
- +2.6% from 2024 to 2034
Skills to build
- Verify AI-drafted forecasts against surface and radar observations
- Write and debug forecast code in Python and R
- Launch weather balloons for upper air wind and temperature measurements
- Build climate visualizations in GIS using paleoclimate databases
- Query observation databases with SQL for forecast validation
- Brief industry and government clients on tailored forecast products
Safer roles nearby
5 related jobs with lower scores.
Questions about atmospheric scientists and AI
Will AI replace atmospheric scientists?
Partly. About 4 in 10 working hours of an atmospheric scientist go to tasks AI can already do, mostly developing data collection programs and preparing weather maps. Forecasting judgment, field sampling, and live public briefings stay with people because they need trained judgment and physical presence.
Which atmospheric scientist tasks is AI taking over?
Developing data collection and display programs is the most exposed task, with an exposure score of 70, because AI coding tools write most of the scripts. Preparing weather maps with graphics software, also 70, follows because AI generates the maps and staff only check accuracy. Gathering satellite and radar feeds, scored 65, is already largely automated.
Which atmospheric scientist tasks stay human?
Collecting air samples from aircraft and ships, with an exposure score of 4, stays human because it needs physical presence in the field. Launching weather balloons for upper air measurements, scored 5, needs hands on equipment. Speaking to the public about weather, scored 25, stays with people because live questions depend on human trust.
How can atmospheric scientists stay valuable?
Move toward verification and interpretation of AI output. Review model output in Python and R, check AI-generated climate visualizations against paleoclimate GIS data, and fact-check every claim in scientific texts before publication. Keep direct control of forecasting operations at stations and hold consultations with agencies on warnings.
Related jobs
Next steps for atmospheric scientists
Check which roles in your team AI can take on
We map your team's tasks, show what AI can do today and build the automations that pay back first.
Related topics
- Jobs AI can't replaceRoustabouts score lowest, at 6.
- Jobs most at risk from AIProofreaders score highest, at 76.
- High-paying jobs with low AI exposurePediatric surgeons earn a median $559,030 and score 29.
- Entry-level jobs with low AI exposureRoustabouts need little prior training and score 6.
- Jobs AI speeds up the mostAI helps with 98% of the task time for health informatics specialists.
- AI and jobs in Malta15 iGaming roles graded task by task.
- Skills that keep work humanThe skills to build in the jobs AI changes least.
- Jobs A to ZEvery job we score, by first letter.
- AI exposure by industryAverage scores for 22 job groups.
- Write a CV screening software can readHow applicant tracking systems read a CV, and what to change.
- AI for businessesThe tasks to automate first in each team and industry.
- How we score AI exposureThe grading, the weights and every source.