← Risk register SOC 19-1031 · reviewed 2026-08-11

Conservation Scientists

25,950 US workers · median $73,010/yr · Science

EXPOSED

The modal conservation scientist — a soil or water conservationist or rangeland manager, often at NRCS or a state agency — splits time between on-site inspection of farms, pastures and watersheds and desk work writing conservation plans, grant paperwork and GIS-based erosion or grazing analyses. That desk half is squarely in reach of AI: remote-sensing classification, plan templating from field data, cost-share application drafting and technical report writing. What holds is walking the ground with a landowner, reading actual soil profiles and vegetation, and persuading a skeptical producer to change practices on their own land.

10-year outlook: Headcount holds roughly flat as agencies lean on AI for plan drafting and satellite monitoring, but field verification, landowner persuasion and credentialed determinations keep the job intact — with less time at the desk and more in the truck.

US employment, 2019–2025+17.6%
22,07025,950 workers

Headcount grew steadily across the period.

Median pay $62,660 → $73,010 -6.8% in real terms (nominal +16.5%, less ~25% US inflation over the period)

The job count is not the verdict

This line is counted by the Bureau of Labor Statistics — the one figure on this page that isn't a judgement of ours. Headcount moves on demand, offshoring, demographics and the business cycle, and automation is one term among several, often not the loudest.

So a falling line is not evidence that AI did it, and a rising one is not evidence that it won't. Both happen in this register: some occupations resist automation and shrink anyway, others are highly automatable and keep growing. The marked year is 2020.

BLS projection, 2024–2034

+3.4% 28,500 → 29,500 on the projections basis

Growing, and only partly exposed

The BLS expects +3.4% more of these jobs by 2034, and at 56/100 the work is only partly exposed — some tasks are automatable, the core of the job is not. Nothing here is in tension.

Different clocks. The score is what current AI could do to this work today. The projection is how many of these jobs will exist in 2034. Everything between the two — how fast employers actually adopt, whether demand grows in the meantime — is why they can point opposite ways without either being wrong.

~2,500 openings a year on average, including replacing people who leave.

One email if this score changes. Watch as many occupations as you like from the same address — no account, and nothing is sent on a schedule, only when a verdict actually moves.

Also known as — 24 job titles this covers

Titles reported by people doing this work, from the US Department of Labor's O*NET survey. If your job title is here, this page is about your work even though the name doesn't match.

RangerPark AideNaturalistPark GuideCamp RangerPark RangerPark WorkerFarm PlannerRange OfficerSoil SurveyorPark AttendantRefuge ManagerConservationistPark NaturalistPark SpecialistPreservationistRange EcologistRange ScientistField AgronomistLand Use PlannerRange TechnicianResource ManagerAquatic EcologistBotany Technician

Score — 56/100 resistance

Holding it up: embodiment (14/20). Weakest point: liability shield (6/20).

Five dimensions, 0–20 each, summed. Higher means more protected. The arithmetic is shown so you can check it: 12 + 14 + 6 + 12 + 12 = 56. · Scored 2026-08-11, and re-examined when evidence accumulates rather than on a schedule.

Task resistance 12/20

Mixed — a routine tier and a judgment tier A 12 rather than a 16 because half the job — drafting EQIP and CSP cost-share applications, running RUSLE2 and NRCS Field Office Technical Guide calculations, classifying NAIP or Sentinel imagery into cover maps, writing the plan narrative — is templated work AI already does faster, while the auger-hole soil texture calls, rangeland similarity-index transects and pasture walk-throughs still need someone standing on the site.

Embodiment 14/20

Hands-on in uncontrolled environments Field days are spent on unfenced rangeland, streambanks and cropland in whatever weather the season brings — digging pits, running line-point intercept, checking gully headcuts after a storm — but a meaningful share of the week is back in a field office at a screen, which keeps it at 14 instead of the high-teens territory of a wildland fire or logging crew.

Liability shield 6/20

Certification preferred, not legally required Nothing legally bars an unlicensed person from writing a conservation plan; the Society for Range Management CPRM and SSSA Certified Professional Soil Scientist credentials are resume assets, and the agency or the engineer who stamps a structural practice — not you — carries the sign-off, which is why this sits at 6 rather than in licensed-professional range.

Trust premium 12/20

Some relationship component Getting a fourth-generation producer to accept prescribed grazing or a riparian buffer depends on years of showing up on that specific operation and being remembered as the person who did not waste their time — but you are also the interchangeable agency contact for whoever walks into the county office, so the relationship carries the work without being the whole product.

Judgment & accountability 12/20

Meaningful discretion You set stocking rates, decide whether a wetland determination is farmed wetland or prior converted, and pick which practice standard fits an odd field — real calls with money and resource consequences attached, but bounded by NRCS practice standards, ecological site descriptions and state technical guides that narrow the answer before you get to it.

Confidence: medium · reviewed 2026-08-11 · how scoring works

What this job involves — and which parts are yours

The verdict above describes this occupation as a whole. Almost nobody does the typical version of a job — tick what's actually in your week and see how your own mix sits.

AI already does these at usable quality

These still need a person

Active moats on the surviving side: embodiment, trust, judgment

How to future-proof this job

Training paths for your skill gaps: MIT OpenCourseWare — finance and accounting free · edX — systems thinking and evaluation methods free to audit · MIT OpenCourseWare — systems analysis and engineering free · Coursera — decision making under uncertainty free to audit

All 35 skills ranked by how many jobs they open →

Where this experience transfers — occupations you could move toward

Computed from U.S. Dept. of Labor O*NET skill and knowledge profiles: high overlap with what you already do, a materially higher resistance score, no large jump in required training, and no licence you would have to start a new pipeline to get. Targets that pay meaningfully less, that are themselves COOKED, or whose own headcount is falling are excluded — a move into a shrinking trade is not an escape.

Chief Executives SAFE · 69/100 · you already have ~67% of the skill profile

Skills to close: Management of Financial Resources, Systems Evaluation, Systems Analysis, Judgment and Decision Making

What would move this back up — beyond any one person

The moves above are yours to make. This is the other half: what would have to change in the world for the occupation itself to score higher. None of it is in any one person's gift, but it is where the floor actually comes from. Scores here are not a one-way ratchet. Only two of the five dimensions — task resistance and embodiment — track what machines can do. The other three track law, what buyers will pay for, and who is answerable, and those move in both directions, often in response to the same pressure AI creates. If every lever below landed, this occupation would score around 72/100 — SAFE.

5 specific changes that would raise this score
  • already happening liability shield +3

    Carbon and water-quality credit markets are moving toward mandatory third-party verification with named, personally accountable verifiers (Verra, Climate Action Reserve, and state nutrient trading programs). If verification bodies bar model-only attestation and require a credentialed individual to certify field baselines and additionality — as Verra's VM0042 soil carbon methodology's sampling requirements point toward — signature demand attaches to this occupation.

  • already happening task resistance +3

    Genuine two-tier structure: if remote-sensing classification, plan templating and cost-share drafting are automated, what remains is site-specific ground truthing, resolving conflicts between model output and observed soil profiles or vegetation state, and negotiating practice adoption. The residual job is the judgment tier, which scores higher than the current blended average.

  • plausible liability shield +5

    NRCS conservation practice standards already require a certified Conservation Planner or a Professional Engineer to sign engineering job approvals (Job Approval Authority) for structural practices like waterways, terraces and waste storage facilities. If NRCS extends JAA-style personal certification requirements to AI-generated resource management system plans — i.e. a named certified planner must sign that they walked the site and concur with the plan — the shield rises. Similarly, state-level Certified Professional in Rangeland Management (SRM) or Certified Professional Soil Scientist licensure being written into state cost-share eligibility (as Texas and some Midwest states have debated for nutrient management plan writers) would do it.

  • plausible judgment accountability +3

    If audit findings on cost-share fraud (USDA OIG has repeatedly flagged EQIP and CSP payment errors) push agencies to name an individual planner as accountable for eligibility determinations and practice-failure calls, the role owns consequential contested decisions rather than routing them upward.

  • plausible trust premium +2

    Narrow route only: landowner-facing trust is real but rarely priced, since most work is free federal technical assistance. A raise would require private conservation planning and carbon-project origination to grow enough that producers pay a fee specifically for a person who walks their ground — visible in TSP (Technical Service Provider) contracting volume.

The limit. Federal employment is the dominant employer and a hiring or budget contraction at NRCS lowers headcount regardless of how any dimension scores; a signature requirement protects the task, not the job count. Trust premium is structurally capped because the service is mostly given away free.

These are conditions, not forecasts — what would have to happen, not what will. Specific rules, cases and bills are named so you can go and check whether they exist and where they stand; verify before relying on any of them. Nothing here is legal or financial advice.

Where this work is, and what it pays there

BLS metro figures for 128 areas. The verdict above does not change by city — the rubric judges what the work involves, not where it happens — but pay and headcount do, and the national median hides a very wide range.

Most of these jobs

Fort Collins-Loveland, CO 650 $84,820 +16%
New York-Newark-Jersey City, NY-NJ 590 $64,020 -12%
Washington-Arlington-Alexandria, DC-VA-MD-WV 490 $99,360 +36%
Boston-Cambridge-Newton, MA-NH 480 $78,580 +8%
Portland-Vancouver-Hillsboro, OR-WA 450 $83,580 +14%
Austin-Round Rock-San Marcos, TX 430 $68,620 -6%
San Francisco-Oakland-Fremont, CA 410 $97,810 +34%
Chicago-Naperville-Elgin, IL-IN 370 $62,710 -14%

Best paid

San Jose-Sunnyvale-Santa Clara, CA 70 $110,200 +51%
Springfield, IL 40 $107,990 +48%
Boulder, CO 70 $101,380 +39%

Percentages are against this occupation's national median of $73,010. Counts are jobs in that metro, not vacancies. Metros where the BLS suppressed the cell are absent rather than shown as zero.

Who is actually doing this — nobody, on the record

We have no reported case of a named organisation automating this occupation. Not one deployment, not one announcement.

That is worth saying out loud next to a score of 56. The verdict above is about what the work exposes — what current AI could do to these tasks. It is not a claim that anyone has done it. For this occupation those two things have come apart completely: the capability argument is on this page, and the evidence column is empty.

Read that as a gap in the reporting we can see, not proof of absence — the dispatch runs on English-language feeds and misses plenty. If you know of a case, tell us, or add a field report from inside the job.

Quick take — do you do this job?

Has AI actually changed your work? One tap, anonymous, and the running tally is public. Nothing else is asked of you.

Self-reported and unverified — a sentiment signal, not a survey. One response per person per occupation; you can change your answer.

Field reports — what people say has changed

No field reports yet. A written account takes a paragraph rather than a tap, goes to an editor before it appears, and is the one thing on this page the rubric cannot produce on its own.

File a field report

Concrete beats general: a tool that arrived, a task that moved, a headcount decision you watched happen. Don't include anything that identifies you or your employer if that would put you at risk.

Watch this verdict
Kept current

Rather than check back: get the digest and we'll tell you what changed — or watch a single occupation from its own page.