Loading statewide data…
If you read nothing else, here is the argument in order of what matters most here:
1. Use is large and rising. Herbicide volume nearly quadrupled from 2021 to 2024, driven by replanting after the 2021 Dixie Fire (see Data & Trends).
2. The dominant chemicals here are the water-mobile ones. The most-used are hexazinone and glyphosate, with imazapyr third — and hexazinone and imazapyr are exactly the chemicals that leach and move through soil and water. They are going onto a forested watershed above a rural community on wells and springs.
3. The biggest unknowns are what isn’t tested: the surfactants and ‘inert’ co-formulants (often more toxic than the active ingredient), real-world tank-mix synergy, and the cumulative, year-after-year load on one watershed.
4. The largest ecological effect is probably indirect: removing plants removes the base of the food web — flowers for pollinators, host plants for insects, food for birds and fish.
The contested glyphosate–cancer question (below) is one thread of this story, not the headline — it is the most dose-dependent and the least specific to Plumas. The chemical-by-chemical detail that follows is a reference appendix.
This map combines the public records that exist, and is honest about the gaps.
California Pesticide Use Reporting (PUR), filed with the County Agricultural Commissioner and DPR. Covers private timberland, ranches, golf courses and the like — with the actual product, active ingredient, amount, and location. This is the bulk of the map (the warm-colored dots).
USDA Forest Service FACTS activity records (the blue ‘Federal’ layer): treatment activity, NEPA project, acres, year and location — but not the chemical or quantity. The Forest Service has said active-ingredient and amount data require a records request, so those show ‘FOIA pending.’
Federal chemical/quantity data (FOIA to USFS Region 5), the most recent local permits/NOIs (county Ag Commissioner), and additional counties/years are being requested or added. Toggle the Streams & water layer on the map to see treatments relative to waterways.
Coverage of these Plumas map layers: county PUR records 2021–2024; federal FACTS 2020–2029 (includes planned treatments). The statewide database behind the map and Data & Trends covers 2020–2022 (complete) with a partial later tail. A historical snapshot, not real-time. Full source inventory and draft records-requests are in the project repo.
This started with a concern about glyphosate — and glyphosate is the most-used herbicide here by volume. But the data show a mix of chemicals. By active-ingredient weight it very likely leads as well. The data are a mix, though — and the chemical that stands out for local risk is hexazinone (Velpar), applied as a dry solid and the dominant soil-applied, water-mobile herbicide here, notable for leaching into groundwater. (Liquid gallons and dry pounds measure different things and aren’t added together, so we rank glyphosate first by volume and flag hexazinone as the key water-mover rather than calling either ‘largest by weight’.) The table is built from the EPA registration numbers in the dataset, so the active-ingredient identifications are reliable.
| Active ingredient | Gallons | Pounds | Applications |
|---|---|---|---|
| Hexazinone | 0 | 76,204 | 273 |
| Glyphosate | 26,912 | 0 | 759 |
| Imazapyr | 7,408 | 0 | 523 |
| Aminopyralid | 616 | 0 | 185 |
| Oxyfluorfen + penoxsulam | 410 | 0 | 7 |
| Indaziflam | 364 | 0 | 39 |
| 2,4-D | 129 | 0 | 5 |
| Sulfometuron/sulfonylurea (verify) | 94 | 0 | 6 |
| Triclopyr | 20 | 0 | 4 |
| Dicamba | 16 | 0 | 5 |
| Clopyralid | 13 | 0 | 4 |
Figures are formulated product as applied grouped by active ingredient (e.g., Velpar DF is ~75% hexazinone), not pure active-ingredient mass. Liquids (gallons) and solids (pounds) aren't summed. In addition, ~22,600 gallons of adjuvants/surfactants (methylated seed oils, spreaders) were tank-mixed in — these aren't herbicides themselves but matter for toxicity (below). One product flagged "verify" needs its active ingredient confirmed.
The dominant chemical here by weight — a soil-applied forestry herbicide. Brand: Velpar DF.
Applied in Plumas County, 2021–2024: 76,204 lb across 273 applications (formulated product, from county records)
A triazinone that blocks photosynthesis (photosystem II). Applied to soil; taken up by roots to kill competing brush and broadleaf plants around planted conifers.
Extremely water-soluble (~33,000 mg/L), poorly held by soil, and only slowly degraded — a combination that makes it highly mobile and prone to leaching into groundwater and running off into surface water. Reported soil half-lives range from about 30 to 180+ days. EPA and state health agencies flag groundwater contamination risk where soils are permeable or the water table is shallow. For a community on wells and springs in a forested watershed, this is the herbicide whose mobility deserves the most attention.
Low acute toxicity. Long-term animal feeding studies show liver effects (and some kidney/body-weight effects); the chronic reference dose is based on a 1-year dog study. Not classified as a likely carcinogen, but the chronic database is older and thinner than for glyphosate.
Practically non-toxic to birds and fish acutely and doesn't accumulate in fish, but native woody plants (oak, maple, cherry, willow, hazel) are sensitive, and it can exceed risk thresholds for small mammals at higher rates.
Sources: Drinking-water health advisory (hexazinone) · MN DNR environmental risk assessment · EPA hexazinone problem formulation
Most-used by volume. Brands here: Accord XRT II, Roundup Pro/Custom, Drexel Duplicator 6, Glystar, Cornerstone Plus.
Applied in Plumas County, 2021–2024: 26,912 gal across 759 applications
Blocks the EPSP synthase enzyme in the shikimate pathway, which plants (and many microbes) use to make essential amino acids; animals lack this pathway. Non-selective and systemic.
Binds fairly strongly to soil and is usually broken down by microbes within days to a few months, but it is water-soluble enough to run off into streams. Its main breakdown product, AMPA, is more persistent than glyphosate itself and is widely detected in surface waters.
This is the genuinely contested one. In 2015 the WHO's cancer agency (IARC) classified glyphosate as a probable human carcinogen (Group 2A), citing limited human evidence for non-Hodgkin lymphoma plus animal and genotoxicity data. The U.S. EPA instead concluded it is 'not likely' carcinogenic — but in June 2022 a federal appeals court (9th Circuit) vacated EPA's human-health finding, ruling EPA used inconsistent reasoning and ignored evidence; EPA withdrew that interim decision and is still re-evaluating, and now expects to issue its final glyphosate registration decision in 2026. One nuance often lost in this fight: IARC assesses hazard (whether a chemical can cause cancer at some dose), while EPA assesses risk (whether it does at real-world exposures) — part of why two serious bodies can land in different places. EPA and IARC reached opposite conclusions largely because EPA leaned on industry-submitted, unpublished studies while IARC weighted independent published literature. Recent reviews also report associations with endocrine effects, liver changes and gut-microbiome disruption.
Toxicity to aquatic life depends heavily on the formulation's surfactant (see 'What isn't well studied'). Kills non-target vegetation on contact/drift.
Sources: EPA – Glyphosate · 9th Circuit vacated EPA finding (2022) · EPA vs IARC genotoxicity review
Second-largest by volume. Brands here: Alligare Rotary 2SL, Polaris SP, Nufarm Polaris SP, Cavalier 2L.
Applied in Plumas County, 2021–2024: 7,408 gal across 523 applications
An imidazolinone that inhibits the ALS enzyme (acetolactate synthase), blocking branched-chain amino-acid synthesis. Animals lack this enzyme, so direct animal toxicity is low; it is broadly lethal to plants.
Highly water-soluble and weakly bound to soil. It breaks down quickly in water by sunlight (half-life a few days) but is persistent in soil — months to as long as ~1–2 years. Sources differ on leaching: its chemistry suggests high groundwater-leaching potential, though some field studies found limited downward movement. It can move off-site via runoff and eroded soil.
Damages non-target vegetation at very low residue levels — as little as ~1/50 of a normal application rate — so drift, runoff and soil persistence can harm desirable and rare plants well beyond the treated area, and can disrupt soil nutrient cycling.
Low mammalian toxicity by the standard endpoints; the main residue concern is plant damage rather than acute animal harm.
Sources: NPIC / NCAP imazapyr factsheet · Massachusetts imazapyr review
Brand here: Milestone. A small volume, but with an outsized off-site footprint.
Applied in Plumas County, 2021–2024: 616 gal across 185 applications
A pyridine-carboxylic-acid 'auxin mimic' that causes uncontrolled growth in broadleaf plants. Grasses tolerate it.
Field soil half-life is roughly 20–70 days, but it is famous for carryover: it survives passage through animals and the composting process and can remain active in hay, manure and compost for months to years. It damages sensitive broadleaf crops (tomatoes, beans, etc.) at concentrations as low as ~1 part per billion, and aquatic half-lives can exceed a year.
Low acute toxicity to mammals, birds, fish and bees; chronic reference dose 0.5 mg/kg/day. The dominant real-world risk is unintended damage to non-target broadleaf plants via persistence and movement.
Sources: OSU – aminopyralid in compost · EPA aminopyralid fact sheet
Brand here: Esplanade F. A newer, long-residual pre-emergent.
Applied in Plumas County, 2021–2024: 364 gal across 39 applications
Inhibits cellulose biosynthesis, preventing seeds/seedlings from establishing. Long-lasting in soil — residual control can persist up to ~3 years.
Marketed as low acute toxicity, but the nervous system is the main target in mammal studies (reduced motor activity, neuropathology in rats and dogs). Studies also report effects on thyroid (possible endocrine disruption), liver, kidney and reproductive organs, plus developmental effects; emerging research raises genotoxicity (DNA-damage) concerns. It was used for about a decade under an incomplete ('conditional') registration.
Highly toxic to aquatic and terrestrial plants and long-lived in soil, so impacts ripple through the base of the food web; some reviews also raise soil-microbe concerns that warrant further study.
Sources: USFS indaziflam health & ecological risk assessment · PEER indaziflam fact sheet
Brand here: Cleantraxx — a deliberate two-chemical mix.
Applied in Plumas County, 2021–2024: 410 gal across 7 applications
Oxyfluorfen inhibits the PPO enzyme (Group 14); penoxsulam is an ALS inhibitor (Group 2). The two are combined for a synergistic, broader-spectrum, residual pre-emergent.
Oxyfluorfen is poorly water-soluble, binds strongly to soil and sediment, is persistent, and is labeled toxic to aquatic organisms; penoxsulam is more mobile. Because this is a fixed mixture, its combined and synergistic effects are exactly the kind of thing single-chemical testing doesn't fully capture (see below).
Confirm the current EPA carcinogenicity/toxicology classification for oxyfluorfen against the EPA label/assessment before citing specifics publicly.
Sources: EPA Cleantraxx label (62719-702)
Smaller-volume 'growth-regulator' (auxin-type) herbicides used for brush.
Applied in Plumas County, 2021–2024: 28 gal across 5 applications
Auxin-mimic. Slightly toxic to mammals acutely; animal studies show moderate reproductive/developmental toxicity (fetal loss and skeletal changes in rabbits at higher doses) and a couple of unpublished studies suggested mammary tumors. The ester (BEE) form is more runoff-prone than the amine salt; inert ingredients such as kerosene and triethylamine in some products add their own hazards; it is an emerging aquatic contaminant.
Older auxin-type herbicides used in small amounts here. 2,4-D has its own long-running cancer-hazard debate (IARC 'possible carcinogen', Group 2B). Dicamba is prone to drift/volatilization and off-target damage. Clopyralid, like aminopyralid, is a persistent compost/manure-carryover chemical.
Sources: NPIC triclopyr general fact sheet · Triclopyr toxicological profile
Most of these herbicides are rated "practically non-toxic" to bees, fish and birds in the acute lethality tests regulators rely on. That is exactly why the real ecological story is mostly about two things those tests miss: sublethal effects and indirect, food-web effects.
Glyphosate doesn't kill bees outright, but at field-realistic doses it disrupts their gut microbiome (reducing beneficial bacteria like Snodgrassella and Lactobacillus), impairs navigation and learning, weakens immunity, and raises susceptibility to pathogens such as Nosema and Deformed Wing Virus — lowering survival. Glyphosate has been detected in pollen, nectar, honey and larvae. Standard pollinator testing measures acute death of adult honey bees and largely misses these effects, and honey bees may not represent native/solitary bees.
Glyphosate-based formulations have been shown to reduce earthworm activity and reproduction and to shift soil nutrient cycling. Indaziflam (Esplanade) has been reported to harm soil amoebae — microbial predators that help drive nutrient cycling. Imazapyr can disrupt the soil enzymes that break down plant material. These are the organisms that keep forest soils alive.
The aquatic risk often comes from the surfactant, not the named herbicide: the POEA surfactant in many glyphosate products is far more toxic to amphibian larvae and aquatic invertebrates than glyphosate itself. Oxyfluorfen (in Cleantraxx) is labeled toxic to aquatic organisms. Given hexazinone's and imazapyr's tendency to reach water, aquatic food webs in treated catchments warrant monitoring.
Herbicides work by removing plants — so fewer plants means fewer flowers for pollinators, fewer host plants for caterpillars, and fewer insects overall, which means less food for birds, amphibians and fish. Conservation scientists link herbicide-driven loss of milkweed and nectar plants to monarch-butterfly decline (documented largely in farmland, but the mechanism — simplifying habitat and removing forage — applies to sprayed forest understory too). This habitat and forage loss is probably the single largest ecological impact of broadcast herbicide use, and it is poorly captured by chemical-by-chemical toxicity testing.
Alongside the herbicides, a small number of applications (about 6) used Tombstone Helios, whose active ingredient is the pyrethroid insecticide bifenthrin. Unlike the herbicides, bifenthrin is designed to kill insects and is highly toxic to bees and extremely toxic to fish and aquatic invertebrates; it is a restricted-use, persistent compound. Even small volumes near water or flowering plants carry outsized risk to insects.
Sources: Xerces Society · Glyphosate & bee gut microbiota (PNAS) · NPIC bifenthrin · Bifenthrin toxicity to bees
The strongest scientific argument is often not "chemical X definitely causes Y," but how much remains unexamined:
Regulatory testing focuses on the active ingredient, yet the co-formulants — surfactants, solvents, "other ingredients" — are frequently more toxic than the active ingredient itself and are largely undisclosed. The classic example: the POEA surfactant in older glyphosate formulations is far more toxic to amphibians and aquatic life than glyphosate alone. Roughly 22,600 gallons of surfactants/oils were tank-mixed into these applications.
Chemicals here are sprayed in combinations (herbicide + oil + sometimes multiple herbicides). One product, Cleantraxx, is itself a deliberate two-herbicide synergy. Regulators test single active ingredients one at a time; the combined and synergistic toxicity of real-world tank mixes is largely unstudied.
Registration leans on acute and high-dose animal studies. Low-dose, long-term, endocrine-disruption, gut-microbiome and neurodevelopmental effects — signals that have appeared for glyphosate and indaziflam — are emerging areas not central to the approval framework.
These forests are treated year after year, and use here is rising. The combined load on a single watershed over time — plus breakdown products like AMPA (glyphosate), hexazinone's degradates, and imazapyr's metabolites — and real-world exposure for rural residents on wells and springs, are not well captured by single-application risk models.
Genotoxicity testing of formulated products often differs from tests on the pure active ingredient, and some chemicals (e.g., indaziflam) were used for years under conditional registrations with data still outstanding.
Sources: Surfactants in GBHs (Mesnage/Benbrook) · POEA toxicity to amphibians · Co-formulant cytotoxicity
Every claim on this page traces to the sources below — regulatory assessments (EPA, IARC, USDA Forest Service, state agencies) and peer-reviewed studies. Open links to verify; classifications can change over time.
This summary is a starting point for further reading, compiled from regulatory and peer-reviewed sources; it is not legal or medical advice. Active ingredients were identified from EPA registration numbers in the dataset. Before using any specific claim in public materials, follow the linked sources and confirm the current EPA/IARC status, which can change.
Loading ledger…