Risky Business: Invasive species management on National Forests - A review and summary of needed changes in current plans, policies and programs

Chapter 3 - Adverse effects

Report Published by Kettle Range Conservation Group (www.kettlerange.org/weeds/)
February, 2001
Return to the Main Document Table of Contents

Section A. Adverse impacts of chemicals on human health must be quantified and eliminated.

The Council on Environmental Quality (CEQ) regulations (40 CFR § 1508.14) state that decisions will analyze effects on the “human environment”, which is defined to include “the natural and physical environment and the relationship of people with that environment.” NEPA (§ 1500.2(f)) states that,

Federal agencies shall to the fullest extent possible…avoid or minimize any possible adverse effects of their actions upon the quality of the human environment.

Despite such direction, National Forests have radically increased their use, and hence reliance, of herbicides over the last several years in projects that cover thousands of acres of public lands (Okanogan NF, 1997; Colville NF, 1998). In the 5-year period from 1994 to 1999, the Forest Service’s use of herbicides on National Forests in Washington and Oregon increased by 1600% (personal communication, Gary Smith, Region 6 Forest Service Noxious Weed Coordinator, December 4, 2000). Yet projects involving the use of herbicides were and continue to be designed and analyzed without due consideration for actual incidents acute poisonings (Wooten, 2000c), known adverse health effects or for the health and safety of vulnerable groups of society (McCampbell, 2000; Voorhees, 1999). The Forest Service continues to disregard their fundamental legal requirements to protect human health and safety.

Vulnerable populations include children, the developing fetus, the elderly, the ill and immunocompromised, and those with asthma, allergies, and other medical conditions. Herbicides pose significant public health risks, particularly for cancer, infertility, miscarriage, birth defects, and effects on the brain and nervous system (Pesticide Education Center, date unknown).

Furthermore, the presence of herbicides discriminates against people disabled with multiple chemical sensitivities by restricting their accessibility to public lands, in violation of the Americans with Disabilities Act (42 USC § 1201 et seq.).

Some of the potential and likely effects, which have been omitted from Forest Service planning documents are briefly described in this section. NEPA (§ 1508.8) specifies that the effects analyzed must address the following:

(a) Direct effects, which are caused by the action and occur at the same time and place.

(b) Indirect effects, which are caused by the action and are later in time or farther removed in distance, but are still reasonably foreseeable.

Indirect effects may include growth inducing effects and other effects related to induced changes in the pattern of land use, population density or growth rate, and related effects on air and water and other natural systems, including ecosystems.

Effects and impacts as used in these regulations are synonymous. Effects includes ecological (such as the effects on natural resources and on the components, structures, and functioning of affected ecosystems), aesthetic, historic, cultural, economic, social, or health, whether direct, indirect, or cumulative. Effects may also include those resulting from actions which may have both beneficial and detrimental effects, even if on balance the agency believes that the effect will be beneficial.

The amount of details to be described about detrimental health effects are given in the NEPA (§ 1508.27):

‘Significantly’ as used in NEPA requires considerations of both context and intensity:

(a) Context. This means that the significance of an action must be analyzed in several contexts such as society as a whole (human, national), the affected region, the affected interests, and the locality. Significance varies with the setting of the proposed action. For instance, in the case of a site-specific action, significance would usually depend upon the effects in the locale rather than in the world as a whole. Both short- and long-term effects are relevant.

(b) Intensity. This refers to the severity of impact. Responsible officials must bear in mind that more than one agency may make decisions about partial aspects of a major action. The following should be considered in evaluating intensity: (1) Impacts that may be both beneficial and adverse. A significant effect may exist even if the Federal agency believes that on balance the effect will be beneficial. (2) The degree to which the proposed action affects public health or safety.

The Forest Service has evaded their responsibility to analyze detrimental human health effects by claiming that such analyses are already available under the Environmental Protection Agency (EPA) analyses included with pesticide registrations. Even so, the Forest Service is still bound to refer to those analyses and apply mitigation for known effects as necessary. However, EPA registration data for herbicides is far from complete and Forest Service Districts that prepare project-planning documents have little technical experience that would allow them to access this technical information or to necessarily understand it.

Actually, according to NEPA (§ 1504.1 (c)), it is the EPA that should review Forest Service actions:

Under Section 102(2)(C) of the [NEPA] Act other Federal agencies may make similar reviews of environmental impact statements, including judgments on the acceptability of anticipated environmental impacts. These reviews must be made available to the President, the Council and the public.

Exposure risks. Exposure risks refer to the amount of a chemical taken up by a body through various routes such as ingestion, inhalation, or dermal absorption. False and misleading claims about the safety of pesticides, lack of government disclosure, and flaws in the federal registration process all raise serious concerns about increased exposure to environmental chemicals when there is lack of information on their reproductive and endocrinological effects, synergy, bioaccumulation and continual low-dose exposure.

Forest Service documents have made claims that exposure risks from herbicides are very low because they primarily occur via dermal routes and that preferred pesticides have low skin permeability. These shallow arguments pale under scrutiny as explained by the Pesticide Action Network (Puvaneswary, 1999):

Scientific principles, particularly toxicokinetics, must apply. The exposed person will be subjected to risks of adverse effects, known or unknown. Even if the chemical has low vapor pressure, appreciable inhalation exposure can occur since micro-droplets can form and particulates can be carried by movement of air. Oral intake can also occur through contaminated food or water. The fact that glyphosate is a systemic herbicide and persists in the environment for a relatively long period of time (as long as 3 years in soil) makes it likely to enter the body through residues (contamination) in food and water. Residues are unlikely to be removed from plant tissues and use of glyphosate in animal feed can result in residues in animal food products such as meat, milk and eggs. Residues are stable to up to one year in plant materials and water and up to two years in animal products in storage.

Clement and Colborn (1992) examined the issue of increased exposure to pesticides, herbicides and fungicides, and the difficulties encountered in measuring exposure rates for women, children, and embryos in vivo, in contrast to typical standards based on adult males. Exposures include routes from both active as well as so-called “inert” ingredients, through food, water, rainwater, snow, household dust, yard soil, and indoor air. The timing of exposures exerts a profound teratologic effect on embryos.

Differences in diet can cause increased susceptibility to pesticide effects. Children with vitamin A deficiencies are more susceptible to the effects of DDT, hydrocarbon carcinogens and polychlorinated biphenyls (Mekdeci, date unknown). The fact remains that there is great uncertainty regarding the public’s exposure, both in terms of risk and frequency, to herbicides used on National Forests. Without such information, it is highly questionable as to whether agencies should continue with the practice of prescribing such chemicals on public lands.

Inhalation exposures. Effects through inhalation of volatile herbicides, spray mists and semi-combusted by-products formed during controlled fires are not addressed in most government documents because research on inhalation exposure routes is seldom provided in pesticide registrations. Nonetheless, this is a method of human exposure which can occur and needs to be addressed.

The herbicide dicamba is highly volatile. Its use in one area can damage crops or native plants another area by vapor migration. Thus, it can be expected to exhibit adverse effects on people and wildlife removed from the application. When applying volatile organic compounds during hot days in narrow valleys, their concentration can build up to very high amounts where airflow is constrained, for instance in narrow valleys, or under tree canopies. This may expose a large number of people to high concentrations of chemicals. For herbicides, which contain hydrocarbon carriers, the volume of volatile ingredients released from the carriers may be much higher than that of the herbicide alone.

In the case of a fire involving previously treated areas, extremely large numbers of the public could be exposed to herbicide-contaminated smoke fumes.

Vulnerable groups. Although the adverse effects of herbicides and their associated surfactants and carriers are profound, the Forest Service prefers to characterize these risks as minimal (Okanogan NF, 1997, p. 127):

It is unlikely that any members of the general public would receive sufficient exposure to develop any adverse effects from the treatment.

This statement indicates that concerns about health risks from herbicides were dismissed from consideration. Such statements are irresponsible and insulting to those who are subsequently harmed by or concerned about harm by the chemicals used. In addition to contributing to a biased decision, the statement is also a denial of human diversity. Even beyond the obviously vulnerable groups of children, fetuses, the elderly, those with impaired nervous, respiratory or immune systems, and sensitive individuals, chemicals can impact the health and well-being of even “the general public”, for example during periods of stress or within predisposed cross-sections of the public.

For instance, the New Mexico Department of Health determined in a 1997 random survey of the general public that 16% of New Mexicans report being sensitive to everyday chemicals like pesticides. Among women and Native Americans the prevalence is 21% and 27%, respectively (Voorhees, 1999).

Unfounded statements, like the one given by the Forest Service above, should not be the basis for a lack of analysis of health hazards. People are commonly made sick by low levels of the pesticides described in this report. In 1998, approximately 50 residents of Tierra Amarilla, New Mexico became ill after a farmer in the center of town sprayed his field with a combination of 2,4-D and Roundup© , at a level that was falsely assumed to be safe. It has been found that the general population of agricultural regions has a higher incidence of birth defects than elsewhere (Garry et al., 1996).

The average citizen is at risk from ambient pesticides. To claim otherwise, or to conceal such evidence from environmental effects documents, is illegal and a direct violation of the NEPA.

It is a fallacy to present herbicide exposures as unlikely. Pesticide poisonings are underreported, according to Ann McCampbell, Chair, Multiple Chemical Sensitivities Task Force of New Mexico:

Most physicians are not sufficiently trained in recognizing pesticide poisonings and hence many cases are misdiagnosed as the flu or some other ailment. Long term effects from pesticide exposures, such as peripheral neuropathy following organophosphate exposures, are also not usually connected with the earlier exposure. Individuals often do not make the connection themselves. The end result is that there is a vast underestimate of the number of pesticide poisonings each year which contributes to a false reassurance about their safety.

Although the increased risks associated with certain public groups is generally ignored in Forest Service documentation, in some instances the Forest Service has provided for special protection for applicators and its own personnel. The acknowledgment of enhanced risk in certain segments of society is a first step toward better recognition of actual health hazards. A Minnesota study indicated an association between paternal employment as a pesticide applicator and a variety of birth defects in offspring, including abnormalities of the lungs, heart, musculoskeletal system, and urogenital system.

Forest Service planning documents often make claims that there is a low likelihood that chemically sensitive people will actually be exposed to herbicides during project implementation, but as noted above, this group of vulnerable people actually represent a significant portion of the population. In addition, their likelihood of exposure is higher than in the general public, according to Ann McCampbell, Chair of the Multiple Chemical Sensitivities Task Force of New Mexico (2000):

In addition, chemically sensitive people frequently seek refuge in the National Forests, sometimes camping for months to years, because they are one of the few remaining refuges from our ever increasingly polluted world. Thus the chances that chemically sensitive people will be exposed to forest herbicides is many orders of magnitude greater than estimated in this report. Another factor that needs to be considered when deciding whether to use herbicides is that it can be a violation of the Americans with Disabilities Act when the presence of herbicides makes forest facilities inaccessible to people disabled with multiple chemical sensitivities ...

When exposed to pesticides, chemically sensitive people can become extremely ill and may suffer severe relapses for months. Some reactions are life threatening. Symptoms can include, but are not limited to, headache, nausea, diarrhea, vomiting, aphasia, trouble thinking and concentrating, weakness, incoordination, numbness and tingling, fatigue, difficulty breathing, irregular heartbeat, seizures, and joint and muscle pain. In addition, many people report that they developed chemical sensitivities after a pesticide exposure, such as after a home, office, or school treatment or exposure to aerially or ground sprayed agricultural pesticides.

The Forest Service has no basis to claim that the general public is unlikely to be unaffected because most studies on human susceptibility to toxic substances are performed on average, healthy, adult males, which do not account for effects on underweight or overweight persons, women, children or different races. Yet, the magnitude of effects on humans can vary by 2 to 3 orders of magnitude (Santa Fe NF, 2000, p. III-E-42).

Children. In 1989, the National Cancer Institute reported that children develop leukemia six times more often when pesticides are used around their homes (American Defender Network, 1989). The American Journal of Epidemiology found that more children with brain tumors and other cancers had been exposed to insecticides than children without (ibid.).

The increased use of pesticides and herbicides in industrial countries may be an important contributing factor to the 50% rise in non-Hodgkin’s lymphoma (NHL) over the past ten years in the American population. Studies of farmers who once used these pesticides found alarmingly high numbers of NHL, particularly in those who didn’t wear protective clothing. This latest finding also proves the theory that most danger from pesticides comes through dermal absorption, not ingestion (Zahm, 1992).

Teratogenic effects. Teratogenic effects refer to birth defects resulting from gene mutations acquired during fetal exposures. Numerous birth defects, particularly limb-reduction defects, have been associated with pesticide exposures in human studies (Restrepo et al., 1990; Schwartz and LoGerfo, 1988; Lin et al., 1994). Exposure of the fetus to pesticides more than doubles the risk of stillbirth due to congenital anomalies (Pastore, 1997).

According to the Executive Director of the Association of Birth Defect Children (Mekdeci, date unknown):

An analysis of current research on immunotoxins also suggests that prenatal exposure to xenobiotics can result in a fourth type of adverse outcome—teratogenesis. New research in developmental immunotoxicology is exploring the possibility that one teratogenic outcome of prenatal exposure to immunotoxins may be impairment of the developing fetal immune system [National Toxicology Program, 1988]. Children born with dysfunctional immune systems are at increased risk of allergies, chronic infections, autoimmune disease, learning problems and/or childhood cancer.

Fat-soluble pesticides accumulate over time in our bodies, then are released at potentially toxic levels when illness or stress results in our fat reserves being metabolized. A large portion of a woman’s lifetime exposure to such pesticides is released in the breast milk for her firstborn child (International Joint Commission on the Great Lakes, 1990).

As a result of chemical exposures, reproductive sterility has resulted in human females, reduced sperm counts in human males (Sharpe and Skakkebaek, 1993) and birth defects have occurred in children (Kurzel and Cetrulo, 1981; Wilson, 1977). A California study reported a statistically significant increase in limb-reduction deformities in the children of mothers who lived in areas of high pesticide exposure. (Schwartz, 1988). Two large chemical companies paid an out-of-court settlement to the family of a child born without any arms or legs after the mother was exposed to pesticides while working in the grape fields during pregnancy (Moses, 1988).

The critical need for further research on the teratogenic effects of pesticides is underscored by Betty Mekdeci, Executive Director, Association of Birth Defect Children (date unknown):

The prenatal and neonatal periods are characterized by immunoincompetence. Any toxic interference with the delicately balanced immune system during this period may have major consequences, much more so than in the adult. (Shoham, 1986) Current research confirms that many immunotoxic agents also have teratogenic potential. One possible teratogenic outcome from prenatal exposure to immunotoxins may be impairment in the development of the immune system. This possible teratogenic outcome has not been addressed to any extent in current research nor has such an outcome been measured in any epidemiological studies of suspected immunotoxins to date. Since the consequences of immune incompetence include such serious outcomes as cancer, chronic illness, severe allergies and learning disabilities, it is critical that the new field of developmental immunotoxicology addresses these important issues as quickly as possible.

Cancer. According to Dr. Lynn Goldman of the U.S. EPA, over 100 pesticides in current use are probable or possible human carcinogens. (Goldman, 1998). One in every three Americans will develop cancer in their lifetime (Mekdeci, date unknown).

A University of Iowa study of golf course superintendents found abnormally high rates of death due to cancer of the brain, large intestine, and prostate (Davidson, 1994), while other experts are beginning to link golfers and non-golfers who live near fairways with these same problems (New York State Attorney General’s Office, 1990).

A case-controlled study (404 cases and 741 controls) linked non-Hodgkin’s lymphoma (NHL) with exposures to herbicides (Hardell and Eriksson, 1999). NHL is a cancer of the white blood cells, which is increasing rapidly in industrialized countries. In the U.S., NHL has the third highest increase in incidence rate at 3.3% per year (Harras et al., 1996, p. 17). The Hardell study observed a positive association between exposure to glyphosate and NHL, in which any chance error could be ruled out with reasonable confidence.

One of the herbicides linked to NHL in the Hardell study is glyphosate, sold by Monsanto under several trade formulations, including Roundup©. Roundup© has also been implicated in hairy cell leukemia (cancer of the blood-forming organs), a rare kind of NHL (Nordstrom et al., 1998). Animal studies have also shown that Roundup© causes gene mutations and chromosomal aberrations.

These studies contradicted previous evaluations conducted by the EPA and the World Health Organization (WHO) that suggested that glyphosate was not mutagenic or carcinogenic. The older investigations were inconclusive and limited to tests of only active ingredients on healthy individuals. In 1995 in the UK, glyphosate was the most frequently reported cause of complaints and incidents from pesticide exposures recorded by the Health and Safety Executive, according to the National (UK) Poisons Centre, which also reported an increase of glyphosate poisonings that year in Malaysia. Monsanto sells over 200 tons of glyphosate each year (Puvaneswary, 1999).

Numerous epidemiologic investigations have also linked the phenoxyacetic acid herbicides 2,4-D with non-Hodgkin’s lymphoma (Hardell et al., 1981; Persson et al., 1989; Hoar et al., 1986; Zahm et al., 1990) and with soft-tissue sarcomas in Sweden (Lynge, 1985). Studies by the National Cancer Society have discovered a link between NHL exposure to triazine herbicides like atrazine.

The Environmental Protection Agency (EPA) categorizes both picloram and atrazine as a “possible human carcinogen.” Picloram is a preferred pesticide for Forest Service use. Atrazine is a long-lived herbicide found in much of the drinking water in the midwestern U.S. and is measurable in corn, milk, beef and other foods. In female rats, it causes tumors of the mammary glands, uterus, and ovaries.

Documented cases of pesticides in groundwater wells are suspect for the incidence of cancer clusters in many towns. In 1989, drinking water in at least 38 states was known to be contaminated (American Defender Network, 1989). After the herbicide Dacthal was applied to Long Island golf courses, it was detected in drinking water wells at levels twenty times the State’s safety limits. The water also contained a dioxin that is a highly toxic by-product of Dacthal (New York State Attorney General’s Office, 1990; Sayan, 1990). The New York State Attorney General sued the manufacturer in 1989 to investigate the contamination and develop a treatment program, since ground water is the main source of drinking water for Long Island. Twenty-two other pesticides have been found in the water so far. However, there is still no requirement or systematic program designed to test for drinking water contamination (American Defender Network, 1989).

Acute effects. Acute effects refer to physical symptoms which are experienced within a short time after a chemical exposure. Acute effects from herbicide exposures are almost completely ignored in Forest Service documents. The Washington Office of the Forest Service has information about acute effects, which is seldom presented in planning documents (Syracuse Environmental Research Associates, 1996):

As indicated in Appendix 1-1, the signs and symptoms of glyphosate or glyphosate/surfactant toxicity in humans generally include gastrointestinal effects (vomiting, abdominal pain, diarrhea), irritation, congestion, or other forms of damage to the respiratory tract, pulmonary edema, decreased urinary output sometimes accompanied by acute renal tubular necrosis, hypotension, metabolic acidosis, and electrolyte imbalances, probably secondary to the gastrointestinal and renal effects. In some cases, elevated temperatures have been noted (Tominack et al. 1991).

Changes in blood enzymes have been observed and attributed to hemolysis (Sawada et al. 1988).

In experimental mammals, signs of acute toxicity after oral or intraperitoneal dosing include increased respiratory rates, elevated rectal temperature, and in some instances asphyxia convulsion. The primary pathological lesion is lung hyperemia (Bababunmi et al. 1978; Olorunsogo et al. 1977; Olorunsogo and Bababunmi, 1980). Hemolysis was not noted in sheep with an inherently low erythrocyte glucose-6-phosphate activity (Geiger and Calabrese, 1985).

The mechanism by which glyphosate exerts its acute toxic effects is not clear. As discussed below, the surfactant in Roundup may be a factor in some of the acute effects associated with exposure to this herbicide.

Based on a series of experiments using rat liver mitochondria exposed to the isopropanolamine salt of glyphosate without any surfactant (summarized in detail by U.S. EPA 1992a), glyphosate appears to be an uncoupler of oxidative phosphorylation (Bababunmi et al. 1979; Olorunsogo 1982; Olorunsogo and Bababunmi, 1980; Olorunsogo et al. 1977; Olorunsogo et al. 1979a,b).

Immune system effects. According to Betty Mekdeci, Executive Director of the Association of Birth Defect Children (date unknown),

The most immediately noticeable immune reaction to pesticide exposure is an increase in allergic reactivity often including multiple chemical hypersensitivity. People whose immunity is suppressed by pesticides may also be unable to fight off viral infections or may experience a reactivation of one or more of the herpes viruses. Immunological studies reveal that pesticide-exposure can cause a decrease in the number of B and T cells. The ratio of T-4 to T-8 helper cells is often reversed similar to the immune abnormalities found in AIDS patients. (Legro, 1988).

Endocrine effects. Endocrine effects refer to the disruption of glandular systems (such as the pituitary, the pancreas, the adrenals, and the testes) which control maturation, development, growth, and regulation within the body through the release of natural chemical transmitters. Atrazine, one of the triazine herbicides frequently used for its resistance to breakdown, can disrupt ovarian function, cause mammary (breast) tumors in animals, and interferes with the binding of steroid hormones and the breakdown pathway of estrogen (Bradlow et al., 1995; Cooper et al., 1996; Danzo, 1997).

Behavioral effects. Pesticide exposures have been experimentally linked to decreased mental abilities and increased aggression among children (Guillette, 1998), as summarized by Montague (1999):

Elizabeth A. Guillette and colleagues studied two groups of Yaqui Indian children living in the Yaqui Valley in northern Sonora, Mexico. One group of children lives in the lowlands dominated by pesticide-intensive agriculture (45 or more sprayings each year) and the other group lives in the nearby upland foothills where their parents make a living by ranching without the use of pesticides. The pesticide-exposed children had far less physical endurance in a test to see how long they could keep jumping up and down; they had inferior hand-eye coordination; and they could not draw a simple stick figure of a human being, which the upland children could readily do.

Synergistic effects. Synergistic effects refer to the combined action of two ore more chemicals that are greater than the sum of the effects of each chemical taken individually. The Forest Service relies on pesticide registrations for individual chemicals based on “acceptable risks” at levels “typically” used in applications. However, the combined effects of multiple chemicals can present much higher risks to the public. In a 5-year experiment using low levels of mixtures of pesticides in the drinking water of male mice, it was found that when combined, levels of chemicals similar to those found in U.S. groundwater have measurable detrimental effects on the nervous, immune and endocrine (hormone) systems (Porter, 1999). Effects found included lowered body weight, decreased immune responses and increased aggressive behavior. This research has a direct bearing on human safety and health because the nervous system, the immune system, and the endocrine (hormone) system are all closely related. If any one of the three systems is damaged or degraded, the other two may then also be adversely affected. The research team notes,

Of particular significance in the collective work of Boyd and others, [1990] Porter and others, [1993, 1984] and our current study, is that thyroid hormone concentration change was consistently a response due to mixtures, but not usually to individual chemicals.

The research team noted that proper levels of thyroid hormone are essential for brain development of humans prior to birth and other studies have shown that attention deficit, hyperactivity and/or aggressive behavior disorders in children are linked to levels of thyroid hormones.

Certain chemicals in the environment are estrogenic (Arnold et al.,1996). When studied singly they exhibit little effect on biological systems, however combinations of two or more weak estrogen-mimicking chemicals can be up to 1000 times as potent. This synergistic interaction of chemical mixtures with the estrogen receptor has profound environmental implications for the National Forests, whose managers generally haven’t considered the synergistic or cumulative effects of herbicides that use surfactants to increase effectiveness.

Cumulative effects. Forest Service documents often present faulty assumptions that herbicides always degrade relatively rapidly and that treatments represent only a single spraying. This assumption is incorrect if one considers that downstream users add to the burden of treatments in an aquatic system and that many noxious weed sites on National Forest lands are “treated” more than just once.

Forest Service planning documents can hardly be expected to accurately portray the manifold routes of potential human exposure risks presented by herbicides. If they could, the public would probably not tolerate any further pollution by herbicides. Instead, the Forest Service tries to portray their use of agricultural chemicals as posing small risk due to rapid breakdown of the chemicals. In fact, research on the breakdown products of herbicide products is scant and toxicological effects analyses are seldom performed on the breakdown products. The assumption that the half-life of pesticide disappearance is a measure of its safety may not be warranted.

Many herbicides are persistent in the soil. According to a Forest Service fact sheet, prometon has a half-life in the soil of up to 6 years and glyphosate has been found in crops harvested over a year after the latest application. Dicamba, triclopyr and picloram have been found in the soil 1, 2, and 3 years, respectively, after the last application. Thus, the commonly misstated assumption in Forest Service documents that herbicides are only used infrequently and do not persist in the environment is unfounded.

Many herbicides are resistant to breakdown, and when they do break down, the secondary byproducts can also have toxicity. For organochlorine pesticides, the chlorine-carbon bond resists breakdown by normal biochemical and physical processes and remains in the environment. Since the majority of organochlorines are foreign to nature, living organisms have developed few methods to detoxify them (Reinecke and Knackmuss, 1988; Nielsen 1990).

For instance, trichloroethane in groundwater may degrade to highly toxic vinyl chloride (Oldenhuis et al., 1989). There are a large number of unidentified organochlorine and other breakdown products accumulating in the environment. In the sediments of the Great Lakes, for example, some highly toxic organochlorines, such as chlorinated dioxins and dibenzofurans, have steadily increased from zero up to 3200 parts per trillion since the chlorine industry started production there in the late 1920s (Czuczwa and Hites, 1984, 1985).

“Inert” ingredients. “Inert” ingredients refer to the contents of a pesticide which are not directly involved in the killing of the intended pest, but which may be quite hazardous nonetheless. According to Knight and Cox (1998), over 2,500 substances in pesticides are not named on product labels. The report shows that over 25% of the chemicals used as “inerts” actually have been identified as hazardous.

Most effects studies are only performed on pesticide active ingredients, which comprise only a small fraction of pesticide products. The “inert” ingredients, which are only inert in a legal sense, can account for a significant portion of a pesticide’s toxicity. In fact, these effects are the basis for manufacturer’s claims of “trade secrets”, which are used to try and block public access to formulation contents. Thus, the true toxicity of herbicide products proposed for use on National Forests is undetermined.

For example, in a presentation to the Forest Service, O’Brien (1997) described the lack of information by officials:

A so-called “inert” ingredient in Banvel CST (active ingredient: dicamba), which is used in Region 6, is ethylene glycol, which has caused birth defects and a decrease in male fertility in laboratory animals. The decrease in male fertility was not reported in the Regional information profile on dicamba formulations, including the inert ingredient, ethylene glycol. Ethylene glycol appears to be an endocrine disruptor.

Chemicals that differ widely in molecular structure are involved in endocrine disruption, such that any given component of an herbicide formulation may be an endocrine disruptor and you could not know that unless it has been tested for various mechanisms of endocrine disruption such as mimicking estrogen or blocking testosterone. Most herbicide formulations have not been tested for any mechanisms of endocrine disruption and likely will never be tested.

Pesticide registration. In response to an application to register a pesticide, the EPA is required to analyze studies on the pesticide and conduct risk analyses that can be used to set limits on pesticide concentrations that will provide an acceptable safety factor during its use. The intent is to produce a balance between risks and benefits, since it is acknowledged that risks are always present. Unfortunately, most studies are commissioned by the chemical manufacturers, which may not be impartial, using animal testing models, which may be inapplicable to humans. Other faults with the process are that it has a narrow focus that allow a number of significant effects to escape documentation, including cumulative effects, synergism, environmental fate in specific environments, phytotoxicity (potential harm to plants), and analysis of “inert” ingredients.

The re-registration of many older pesticides mandated by law in 1988 is still incomplete. For many pesticides, exposure risks are unknown. By 1999, the EPA had been unable to determine risk factors for 9,700 pesticides which had outdated registrations, and limits had still not been set for 60% of the 5,500 pesticides that are supposed to get priority attention (Eisler, 1999). The decision by Congress to require EPA to update the registrations came after it was learned that children have higher risk factors than those given in the old registrations.

In fact, recent evidence indicates the studies used in registrations are inadequate at portraying toxicological risks. A study by Porter et al. (1999) found that current methods used by the EPA and others for studying the toxic effects of low-levels of pesticides may be flawed. Speaking in a press release from the University of Wisconsin (Devitt, 1999), Porter stated,

...Herbicides can have neurological impacts and hormonal impacts and immune impacts ... They are not the harmless chemicals they are sometimes portrayed to be. They can be every bit as biologically active as insecticides or fungicides ... Neurological, immune and endocrine tests for pesticides have been mandated by federal law for almost three years, but there has been no enforcement of these laws ... Toxicological testing so far has been extremely limited in scope and focused on mechanisms that require extensive mutations or cell damage to show any effects. They do not adequately assess the potential for biological effects under real world exposure scenarios.

Pesticide profiles analyze health effects by extrapolating from animal experiments, which include tests of the LD50 (the dose that kills half the test animals) and the NOEL (no-observed-effect level/dose). These endpoints only indicate gross symptoms in test animals. Such gross measures do not give any indication of the effects on impairments of memory, learning, and other more subtle areas of functioning that would be significant to humans. It is inappropriate to conclude that there is “no effect” just because no effect is observed and there is still a great deal of uncertainty about which animal studies are applicable to humans, or even appropriate.

Cancer risks are calculated for chemicals for which the EPA had established a cancer potency value at the time of analysis. Thus, carcinogenic contaminants of “inert” ingredients are not considered in recent pesticide profiles. And, even if a chemical is carcinogenic, the EPA risk / benefit models allow for an acceptable amount of cancer effects in registered pesticides. Picloram was noted to be a possible human carcinogen by the EPA but was approved for use nonetheless.

The Forest Service health protection measures are based on information contained in pesticide registration profiles stored at the Supervisor’s office. However, registration profiles made with the above assumptions may be inapplicable in specific circumstances.

Health risks described in Forest Service documents (Syracuse Environmental Associates, 1996) rapidly become outdated as new research becomes available. The Forest Service uses such documents as a shield against having to address new research, by claiming that new information can only be analyzed during periodic reviews, which occur infrequently. At that time, the Forest Service may very well switch preferences to another herbicide which has less information available on its harmful effects, and start the misinformation process over again.

Case example: Okanogan NF Integrated Weed Management Environmental Assessment (EA) (1997, 1999)

The Okanogan NF Integrated Weed Management EA for 1997 received many comments from the public asking for documentation and analysis of the risks of herbicides to human health and safety, yet all of these concerns for safety were lumped into a single issue on p. 15-16:

Noxious weed populations can degrade recreational experiences by decreasing the desirability of campsites, replacing native plant populations in developed and dispersed areas and changing the scenery. Herbicide contact could pose risks to human health through skin exposure, inhalation, or ingestion. Some noxious weeds also pose risks to human health.

The marginalization of human health as mere “issues” rather than actual hazards suggests that there was never any intention of questioning the safety or use of herbicides, except in a very limited fashion, and this is borne out in the analysis section.

Two years later the Okanogan NF prepared a second EA (1999) and through another public comment process, the issues identified through public comments were exactly the same.

Why are the issues of public health ignored? According to the rationalization given in the EA (Okanogan NF, 1997, p. 17), public comments were addressed in a “higher level document”. In other words, concerns about human health and safety were not considered in the EA. By its limited scope, the agency effectively avoids having to consider issues that it doesn't want to.

The purpose of an EA is to assess a problem, propose and evaluate alternatives and select the most effective remedy, which should be the least harmful to the environment. In this case, the alternative to use herbicides had been selected prior to doing an analysis. The EA was only used to justify a predetermined decision rather than truly explore alternatives.

Case example: McFarland Creek spraying of sensitive individuals results in acute effects In 1999, McFarland Creek on the Okanogan National Forest was sprayed repeatedly by County Weed trucks under a “Coordinated Weed Management Area” agreement. The trucks used herbicide mixtures and procedures that would have been illegal for the Forest Service to use, and they treated areas without posting signs or notifying either the Forest Service or the public. The event is described from a personal experience (Wooten, 2000c):

In the summer of 1999, the entire McFarland Creek watershed, an area of about 15 square miles, was sprayed with pesticide along most of the roads. The intense summer heat raised a cloud of petrochemical vapor, which settled in the valley bottom for several weeks. During the 5-10 days that I worked in the watershed, I experienced several bloody noses, constant headaches and occasional dizziness. These symptoms began immediately after being passed by a County spray truck bound for National Forest Land (cover illustration of this paper) and whenever I worked in the sprayed area. The symptoms were partially relieved by leaving the sprayed area and working the higher ridges a mile or more from the sprayed roads. At that time, in response to another herbicide incident, I requested that the Okanogan National Forest notify me when and where they would be using herbicides, as I was a contractor working in the treated areas. However, the Forest never notified me until the fall of 2000.

I am registered as a sensitive individual in Washington, which means that agencies must try to contact me when using herbicides in my area. During scoping for the Okanogan National Forest Environmental Assessment (EA) for spraying noxious weeds (1997), public comments were submitted reminding the Forest Service that the herbicides might harm sensitive individuals, or even casual visitors to the Forest. However, mitigation measures to protect human health were not incorporated by the EA in these subsequent applications. It is worth noting that the magnitude of this problem involves approximately 50 other sites treated this same way.

The EA written in 1997 specified that the Forest Service would use the herbicides Tordon (picloram) and glyphosate, but the effects I experienced were more akin to acute effects that would be expected from volatile hydrocarbons, rather than the systemic reactions I would expect from herbicides. The onset of symptoms began immediately upon detection of a strong hydrocarbon vapor smell, and were made worse because there was nowhere in the watershed where the smell could be avoided. The smell alone was overwhelming, and made any work in the area an unpleasant experience.

From a FOIA response of notes of Forest Service Contract Inspector, Bauman, taken in McFarland Creek on July 7, 2000, it was apparent that the Forest Service was also unaware of when this spraying occurred, so of course, no warning signs could have been posted in time to warn anyone. The EA stated that special precautions would be necessary for Forest workers, yet these were hollow claims.

Solutions

Section B. Adverse impacts of chemicals on the environment must be quantified and eliminated.

The NEPA regulations (40 CFR § 1500-1508) are specific about limiting negative environmental impacts. In fact, Section 1500.2(f) states,

Federal agencies shall to the fullest extent possible…(u)se all practicable means, consistent with the requirements of the Act and other essential considerations of national policy, to restore and enhance the quality of the human environment and avoid or minimize any possible adverse effects of their actions upon the quality of the human environment.

In attempting to minimize “any possible adverse effects”, mitigation is implied, which is defined in NEPA (§ 1508.20) as:

(a) Avoiding the impact altogether by not taking a certain action or parts of an action. (b) Minimizing impacts by limiting the degree or magnitude of the action and its implementation. (c) Rectifying the impact by repairing, rehabilitating, or restoring the affected environment. (d) Reducing or eliminating the impact over time by preservation and maintenance operations during the life of the action. (e) Compensating for the impact by replacing or providing substitute resources or environments.

The use of herbicides is both unwarranted and illegal if their impacts on the environment are not disclosed. This Section of the NEPA presents details of potential and likely impacts, many of which have been intentionally excluded from Forest Service planning documents.

One such impact from the use of herbicides, is the removal of desirable native plant species as an unintended consequence of the lack of host-specificity by herbicides. The harm this causes to biological diversity and ecosystem integrity is seldom disclosed publicly, in disregard for the Code of Federal Regulations (36 CFR 219 § 27 (G)), which state that management prescriptions,

where appropriate and to the extent practicable, shall preserve and enhance the diversity of plant and animal communities, including endemics and desirable naturalized plant and animal species, so that it is at least as great as that which would be expected in a natural forest and the diversity of tree species similar to that existing in the planning area. Reductions in diversity of plant and animal species from that which would be expected in a natural forest, or from that similar to the existing diversity in the planning area, may be prescribed only where needed to meet overall multiple-use objectives. Planned site conversion shall be justified by an analysis showing biological, economic, social, and environmental design consequences, and the relation of such conversions to the process of natural change.

Rather than provide actual data, the Forest Service prefers to characterize herbicides they intend to use in vague, general terms that underexaggerate any undesirable effects and make them appear benign, e.g., Okanogan NF (1997):

Picloram is relatively toxic to invertebrates. However, the medium lethal concentrations of (LC50s) picloram are one or two orders of magnitude less toxic to aquatic organisms than most insecticides and would probably have little impact on food resources of fish (Driver, 1991).

Such simplification of the known and potential effects from herbicides that are used on public lands greatly misrepresents the information that agencies are required to disclose to the public and decision-maker prior to undertaking such actions. Perhaps even more importantly, such simplification also misrepresents the effects of herbicides which agencies, such as the Forest Service, are required to “avoid or minimize”, according to NEPA.

Below, is a brief summary of some of the information that currently exists on the effects of herbicides and invasive species on the environment. This type information should be more readily recognized and included in planning documents and decisions. Where information is lacking, land managers should use extreme caution in assigning such situations as low risk.

Soils.Evidence is readily available to show that soils can be impacted both by invasive species as well as herbicide treatments. Research in shrub-steppe habitats showed that invasive species, which are usually non-mycorrhizal, disrupted succession by native plant species, 99% of which were mycorrhizae-dependent (Wicklow-Howard, 1994). The authors suggested that long-term impacts to mycorrhizae may result from invasive species because without host plants to support the mycorrhizae, the fungal propagules may not be able to survive.

Available research indicates that herbicides alter soil ecosystems through direct effects on soil microflora, such as plant pathogens, antagonists, or mycorrhizae, resulting in increased or decreased incidence of plant disease (Levesque and Rahe, 1992). This study also found that herbicides predispose pathogens to fungicide susceptibility, e.g., they act as synergists. Persistence of herbicides through soil and humus binding is unaccounted for in most quantitative measurements of toxicity used to determine safe exposure levels (Bordeleau and Bartha, 1971) and the possibility exists that they may be released at unexpected times in the future (Pramer and Bartha, 1980).

Herbicides can lead to alteration of soil microclimate (Evans and Young, 1984) by causing destruction of beneficial macro- and microorganisms in the soil, including earthworms, fungi and bacteria (Pimentel, 1992). Soil organisms are vital to the proper functioning of soil ecosystems and their loss leads to nutrient deficiencies. Earthworms and soil microorganisms break down organic matter and make nitrogen and other nutrients accessible to plants, yet earthworms are vulnerable to pesticides (Bugg, 1994).

The negative effects of herbicides on the living components of soils initiates a pernicious cycle of decline in forest health and a wide range of deleterious effects. The loss of soil microflora as a result of using herbicides has led to the conversion of productive forestland to unforested openings (Perry and Amaranthus, 1994; Amaranthus and Perry, 1987; Perry, 1984). Herbicides kill a broad range of non-target vegetation, which can lead to altered ecosystems, beginning with raised site temperatures as a result of loss of cover (Holtby and Baillie, 1987). The effect of vegetation removal on test plots resulted in increased sediment yields of 216% and 126% on bunchgrass and knapweed sites respectively (Lacey et al., 1989).

The soil crust and vegetative cover is important for increased soil stability, water infiltration, and soil fertility (Harper and Marble 1988; Johansen, 1993; Belnap and Gardner, 1993) and reduces the susceptibility of the soil to wind and water erosion (Iverson et al. 1981; Wilshire and Nakata, 1976). Increased erosion can result in a decline in water quality due to an increase in sediment and dissolved matter (Miller, 1970). In addition, a reduction in soil water content influences soil biota activity, nitrogen cycle dynamics (Torbert and Wood, 1992), vascular plant vigor and reproduction (Crawford 1979; Skujins, 1984), and decomposition rates of soil organic matter (West, 1981). Changes in uptake and cycling of soil nutrients have resulted from elimination of cryptobiotic crusts, which accompany species changes resulting from soil disturbance (Bolton et al., 1993; Anderson et al., 1982; Kleiner and Harper, 1972).

Belnap (1995) found that concentrations of nitrogen and macronutrients in annual, biennial, and perennial plants were significantly higher when grown on undisturbed crusted surfaces than on trampled areas. The disruption of nutrient cycles and availability can adversely impact vegetation productivity and abundance and ultimately the ecology of an area. She found that disturbed arid soils at her study site in Utah had lowered nitrogen and carbon inputs and slower decomposition of soil organic matter, resulting in lower nutrient levels in vascular plants. Additional time, ranging from 35 to 250 years is required for the recovery of cyanobacterial biomass, lichen cover, and moss cover, respectively (Belnap, 1993). As a consequence of the fragility, sensitivity, and slow recovery of desert soils, these areas are particularly susceptible to desertification (Belnap, 1995).

Reeves et al. (1979) documented a negative correlation between disturbance and mycorrhizal fungi in their study of a western Colorado sage assemblage. Reductions in survival and growth of Pinus lambertiana (sugar pine) seedlings were correlated with reductions in the formation of beneficial ectomycorrhizal fungi following seeding of the non-mycorrhizal grass Lolium multiflorum (annual, or Italian rye) (Amaranthus and Perry, 1994).

Aquatic resources. Forest Service documents often claim that herbicide impacts on water quality will be negligible, yet they consistently fail to substantiate these claims. Even when statistics are given that show a potential for lowered water quality from chemical applications, documents quickly dismiss such possibilities as insignificant. A careful interpretation of potential effects would reveal that not only ecosystems, but human industries, agriculture, and society at large suffer each time water quality is diminished by chemicals used in herbicides applications.

The Forest Service uses herbicides, such as picloram and the sulfonylureas, which have extremely high phytotoxicity and high potential for leaching and drift. Unless the herbicide breaks down before reaching groundwater, it will contribute to the rising levels of chemical mixtures already found in downstream groundwater. In recommending against the registration eligibility of picloram, the EPA had this to say about the likelihood of its effects (Abramovitch, date unknown):

The use pattern of picloram is highly specialized, but it is almost certain to eventually reach ground water in areas where it persists in the overlying soil. In submitted terrestrial field and forestry studies, picloram exhibited calculated half-lives of up to 278 days and was detected up to the limits of sampling depth (up to 1.8m). Even under the most constrained soil conditions in the submitted field studies (e.g., 1/2 the maximum application rate, high soil organic matter, minimum rainfall) the compound moves through the soil profile to the deepest sampling depth. In addition, in soils of low permeability, picloram residues may be transported by dissolved run-off during rainfall events and potentially reach non-target vegetation.

In 1989, drinking water in at least 38 states was found to be contaminated. (American Defender Network, 1989). Such conditions could potentially be made worse by herbicide applications on National Forest lands. The herbicide Dacthal, a chemical similar to the picloram which the Forest Service uses, was applied to Long Island golf courses after which it was detected in drinking water wells at levels twenty times the State’s safety limits.

The EPA eligibility registration of picloram continues:

Furthermore, incident data indicate that 15,880 pounds of fish died from symptoms of chemical poisoning at a fish hatchery in Sheridan, Montana on July 21, 1989. Picloram (Tordon 22K) was detected at the scene and the chemical had been sprayed one-quarter mile upstream from the fish hatchery by Montana State highway personnel. Rain on the day of the fish kill had washed Picloram into the hatcheries water source. Although the LC50 data indicates that the risk does not exceed the LOC [EPA determined levels of concern], the latest EPA paradigm states that an incident itself is sufficient to exceed the LOC for acute risk.

The Forest Service has this to say about the risk of picloram to fish (Okanogan NF, 1997, p. 108):

In areas adjacent to identified fish populations, buffer areas as described above [hand wicking and hand spraying within 50 feet of visible water] would be used to minimize impacts to fisheries.

Rashin and Graber (1993) examined the use of pesticides on forested sites, in accordance with Best Management Practices (BMPs) established in the Washington Forest Practices Rules and Regulations. Pesticides examined included 2,4-D, triclopyr, glyphosate, imazapyr, metasystox-R, and chlorothalonil. Following their application, pesticides were detected in streams and runoff at all seven sites, with peak levels ranging from 0.02 to 7.55 mg/L. The majority of pesticide introduction to streams was attributed to off-target swath displacement and drift from spray areas near streams. The overall distribution of pesticide levels indicated that overspray occurred in small headwater streams because the applicator had incorrectly assessed them as not having surface flow. The BMPs were judged ineffective because water quality standards were exceeded, drift of herbicide spray into surface waters was not prevented and compliance with pesticide label restrictions regarding entry to surface waters and avoidance of off-target drift was questionable.

Despite evidence of changes in streamside habitats resulting from herbicides, the effects have been largely ignored by Forest Service managers (O’Brien, 1997):

If stream or wetland temperature is raised upon the removal of vegetation, or if cover is lost upon which butterflies, nesting birds, or other wildlife depend, effects that are not even considered or tested for in the registration of herbicides may be caused. EPA states, for instance, that, ‘a number of terrestrial and aquatic plant species are listed as being at jeopardy from the use of herbicides.’ I would guess that none of the registration documents for any of those herbicides predicted or even discussed the demise of rare plants from the use of the herbicides.

Austin et al., (1991) found that glyphosate negatively affects the aquatic food chain through stimulation of eutrophication. Buhl and Faerber (1989) found that Roundup© caused an 89% decline in the numbers of the midge, Chironomas riparius, an important food resource in the food chain. Goldsborough and Brown (1988) found that the photosynthetic rates of algal communities in six forest ponds were affected by Roundup©, with an EC50 value (glyphosate level resulting in 50% inhibition of carbon fixation) between 8.9 and 89mg/L.

Increased evapotranspiration caused by invasive plants can lower water tables. (Kerpez and Smith, 1987; Horton, 1977). Herbicides applied to halt weed encroachment add to the severity of this effect by decreasing the amount of available shade and increasing solar exposure to the soil (Parendes and Jones, 2000).

The paucity of published research on the action of glyphosate on aquatic species composition, bioaccumulation and food chain relationships further recommends caution in the application of this herbicide, which unfortunately has gone unheeded by the management of National Forests.

Vegetation. Native plant loss occurs in at least three ways during invasive species management. First they may be competitively displaced by invasive species. Secondly, they may be killed outright by herbicides. Thirdly, they may be displaced by so-called beneficial seed mixtures applied to mitigate herbicides.

The loss of biological diversity attributed to invasive plants is well-documented (Randall, 1996; Rosentreter, 1994), and includes native plant displacement occurring through competitive exclusion (Harris, 1967). Other examples include interference by Cirsium vulgare (bull thistle) resulting in lowered growth rate and survival of Pinus ponderosa in forest plantations (Randall and Rejmánek, 1993). Displacement of native plants and reduced plant diversity resulted following entry of Centaurea maculosa (spotted knapweed) (Tyser and Key, 1988) and the displacement of native bunchgrasses by Bromus tectorum (cheatgrass) was noted following fire (Melgoza et al., 1990). Bromus tectorum (cheatgrass) dominance caused permanent increased frequency and severity of fires (Billings, 1983; Peters and Bunting, 1994).

Loss of species diversity occurred in timberline vegetation with exotic invasion by Kentucky bluegrass, Poa pratensis, and timothy, Phleum pratense, (Weaver et al., 1989). Destruction of nontarget plants resulted in lowered species richness and replacement by introduced species following 2,4-D treatment of native Veratrum californicum in an alpine plant community (Anderson and Thompson, 1993).

However bad the effects of invasive species are on ecosystems, in almost all cases, the effects of herbicides are often worse. When herbicides are used on a site, they may leave the area devoid of all vegetation, and ripe for future invasion. The loss of native plants from herbicides needs no explanation—it is an unavoidable impact whenever non-specific herbicides are chosen for treatment measures. However, the initial loss of species leads to further ecosystem disruption, which is seldom documented or taken into consideration. Planning documents produced by the Forest Service rarely analyze ecosystem effects to native vegetation, despite the fact that plants are primary producers in an ecosystem.

Following plant removal, soil temperatures and water retention may be negatively affected, and often severe disruptions to plant successional and nutrient cycling processes may occur as a result of destruction of important soil microflora (Evans and Young, 1984; Perry and Amaranthus, 1994; Amaranthus and Perry, 1987; Perry, 1984).

Results of herbicide applications include reduced plant cover and biomass, fewer and less vigorous plants (Jeffery et al., 1981), lowered plant diversity (Anderson and Thompson, 1993), increases in density of exotic species (Barber, 1999),

Food web disruption may be caused by elimination of important native primary producers by invasive plants (Orians and Solbrig, 1977; Marks and Bormann, 1972). Habitat itself is often altered. For example, habitat selection by birds is influenced by vegetation structure, diversity, composition, and habitat patchiness (James and Wamer, 1982; Rotenberry and Wiens, 1978), all of which are affected by changes in vegetation structure caused by herbicide applications.

The role of added surfactants is seldom accounted for in herbicide applications and effects. However, there is an extensive amount of literature on herbicides indicating that the addition of surfactants can greatly enhance their phytotoxicity (Green et al. 1992; Clay and Lawrie, 1988; Sherrick et al. 1986; Turner, 1985), which thus magnifies the effects on native vegetation. Glyphosate, as the formulation Roundup©, contains a polyethoxylated tallow amine surfactant at a level of 15% (150 g/L) and Roundup© Pro contains a phosphate ester neutralized polyethoxylated tallow amine surfactant at a level of 14.5%. Other formulations of glyphosate recommend the use of a surfactant to improve its efficacy.

As a broad spectrum herbicide, glyphosate has documented phytotoxicity to a wide array of organisms, including lichens (Brown, 1995), nitrogen-fixing bacteria (Tu, 1994; Carlisle et al., 1986; Moorman et al., 1992; Martensson, 1992) and beneficial mycorrhizal fungi (Estok et al., 1989; Chakravarty and Chatarpaul, 1990; Sidhu and Chakravarty, 1990; Chakravarty and Sidhu, 1987). These species are all integral components of the ecosystem, which are negatively affected by herbicides. The Carlisle study found that the rate of glyphosate degradation correlates with the soil respiration rate, an estimate of microbial activity. Glyphosate has been found to inhibit growth (at 50 ppm) of 59% of randomly selected soil bacterial, fungal, actinomycete, and yeast isolates; of nine herbicides tested, glyphosate was the second most toxic.

Picloram is another herbicide often touted as being of low toxicity, however its extremely high phytotoxicity, combined with its high potential to leach, have caused the EPA to recommend withdrawing its registration. Because of its broad applicability and persistence, picloram is a potent phytotoxic compound. Picloram is readily adsorbed by plant roots and is readily translocated throughout plants, where it remains intact and stable. The US Fish and Wildlife Service has determined that the compound picloram, because of its persistence, mobility and toxicity to plants, may pose a threat to endangered plant species. According to EPA’s report, Reregistration Eligibility Decision (R.E.D.), “Picloram is nearly recalcitrant to all degradation processes.”

Revegetation attempts following herbicide applications also often result in a loss of ecosystem integrity, unless restoration plant species are carefully chosen. The regular seeding of strongly competitive and aggressive alien species following National Forest management causes dramatic displacement of native vegetation (Ralphs and Busby, 1979). The use of inappropriate seed mixtures following wildland herbicide applications leads to further degradation as grazers respond to the changes. Food web disruption by maladaptive herbivores has been documented (Edwards and Gillman, 1987; Daubenmire, 1940). In the case of livestock use of reseeded wildlands, the result is conversion of native ecosystems to agricultural ones. This in turn leads to a cycle whereby livestock selectively graze beneficials, which leads again to weed invasions ( Photo 1, p. 3). The inescapable conclusion is that increased livestock use is a result of the maladaptive restoration plantings, which then becomes a source of further spread of invasive species, the need for additional herbicides, and further seedings.

Fish and wildlife. Wildlife habitat reduction by invasive plants is frequently used to justify a “need” for herbicides (Bedunah, 1992), however wildlife may be directly affected by herbicides, or indirectly through changes in habitat. Habitat for native organisms may be reduced or eliminated by invasive plants (Nee and May, 1992; Brothers and Spingarn, 1992).

Herbicide application is implicated as one of the causes in the global decline of amphibian populations (Blaustein and Wake, 1995). A summary of amphibian effects from herbicides indicates that these species are a very sensitive indicator of environmental effects that should be included in any environmental monitoring scheme intended to mirror effects (Schuytema and Nebeker, 1996).

Surfactants in different commercial preparations of the herbicide glyphosate can result in 400-fold greater toxicity to sockeye salmon fry (Monroe, 1988). In a study of the effects of glyphosate on fish, Servizi et al. (1987) found that the combined effect of glyphosate and the surfactant POEA found in the commercial produce sold as Roundup© is more than additive, and some surfactants used alone are more toxic to fish than the pesticide. Martinez and Brown (1991) found that the surfactant POEA (in doses of 1.03g/kg) has serious pulmonary toxicity, but not quite as serious as the full formulation, Roundup©, which produced 100% death in rat subjects within 24 hours. Folmar (1979) found that Roundup© is four times more toxic to rainbow trout fry and fingerlings than to larger fish.

Glyphosate formulations are acutely toxic to fish (Servizi et al., 1987). Acute toxicities of Rodeo© , with X-77 Spreader© per label recommendations, vary from 120 to 290 ppm (Mitchell et al., 1987), and can result in effects to pink, chum, coho, and chinook salmon (Wan et al., 1989). Sublethal effects of glyphosate on fish include erratic swimming, labored breathing, altered feeding, migration and reproduction and increased likelihood of being eaten (Morgan et al., 1991; Liong et al., 1988). Studies also show that salmonids may alter their migration patterns in response to avoidance of herbicides (Folmar, 1976).

Applications of glyphosate to ditchbanks near aquatic ecosystems may be hazardous to resident fauna if the water temperatures are elevated because glyphosate causes water temperatures to increase for several years following treatment (Holtby and Baillie, 1987).

A number of studies show detrimental effects from glyphosate on birds (Cox, 1991, 1995, 1995b). MacKinnon and Freedman (1993) examined the effects of glyphosate use on breeding birds and found densities of most common breeding species decreased significantly on all treatment plots.

A memorandum from Akiva Abramovitch, Ph.D., Chief of EPA Review Section #3, to Walter Waldrop, Product Manager #71, EPA Special Review and Registration Division, reminds the department of potential to fish that can occur, despite label precautions.

The above table characterizes the Picloram acid as moderately toxic to freshwater fish with a LC50 of 5.5 mg/l (ppm) and slightly toxic to freshwater invertebrates (LC50 of 34.4 mg/l). Field runoff studies conducted with cutthroat trout concluded that concentrations as low as 290 eeg/l and 610 eeg/l will affect survival & growth of cutthroat trout ...

...The preliminary aquatic risk assessment indicates that the Picloram TIPA and Potassium Salts are not likely to affect nontarget aquatic organisms from ground and aerial applications on an acute toxicity basis. However, for endangered species the Potassium salt is likely to adversely affect fish for ground applications. To complete the aquatic risk EEB will require the acute LC50s for a coldwater fish (rainbow trout), a warmwater fish (bluegill), a freshwater and marine invertebrate, and a marine oyster shell deposition study for the IOE, and a marine fish study for the Potassium and TIPA salts ...

Connor and McMillan (1990) compared moose forage resources on control and on herbicided cutovers. On control areas, available moose browse was four times greater, and utilized browse was 32 times greater, than in treated areas after one growing season post-spray. Winter moose presence was almost two times greater on untreated than treated areas after one growing season.

Fire. Invasive species can lead to increased fire frequency and severity (D’Antonio and Vitousek, 1992; Whisenant, 1990), such as in the case of cheatgrass (Bromus tectorum) (Young and Evans, 1978). Unfortunately, noxious weed managers have essentially given up controlling this pest, to the point where it is even an allowed contaminant in “native” seed mixtures.

Fires have become more common and extensive in pinyon-juniper woodlands and sagebrush ecosystems invaded by cheatgrass (Billings, 1994). Ponderosa pine forests have also shown an increase in incidence of fire following the invasion of cheatgrass (Monsen, 1994).

Cumulative, indirect and non-target effects.Cumulative effects are the incremental accumulation of effects over time and space that may not be significant individually, but which may be significant when added together. Much of our native fauna is threatened by the synergistic effects of synthetic compounds on living estrogenic activity. These estrogenic compounds are associated with many herbicides and pesticides (Fox, 1992).

According to O’Brien (1977):

The removal of microbiotic crusts, depletion of mycorrhizal fungi, erosion, soil compaction, replacement of native vegetation or wildlife with exotic vegetation or wildlife, removal of old growth trees or riparian vegetation, isolation from floodplain functioning, and other stresses may be cumulative with herbicides on wildlife and vegetation. For instance, if livestock grazing has reduced riparian vegetation, and the stream temperature has been raised somewhat, will the toxicological effects of an herbicide be enhanced by the temperature increase?

Again, the registration of the active ingredients of herbicide formulations do not, and cannot, take the cumulative impacts of site-specific stresses into account. The Forest Service is neither funded nor inclined toward detecting cumulative impacts when herbicides are used, and none of the herbicide information profiles consider these impacts.

The environmental fate of herbicides used on Forest Service lands is wrongly ascribed to be “insignificant” in Forest Service documents that do not consider that U.S. groundwater is already significantly contaminated with herbicides and other pesticides. In Washington alone, 6% of public wells were found to be contaminated with measurable herbicides and other pesticides (US Geological Survey, 1996).

National Forest disclosure documents have been remiss about documenting the detrimental effects of herbicide on non-target vegetation (Wenatchee NF, 1998; Okanogan NF, 1997). Herbicides not only destroy the target weed, but often reduce a number of non-target plant species as well. According to the Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida (Rao et al., 1998),

In addition to the pesticide solubility and soil permeability it is important that the pesticide’s toxicity to nontarget species be considered. Some of the pesticides listed in Tables 1 and 2 have severely restricted use due to acute toxicity or long half-life. An important purpose of the pesticide container’s label is to instruct users to apply the pesticide safely and with minimum threat to nontarget species, both on and off the application site.

During a denied appeal of a plan to use herbicides, the Regional 6 NF Noxious Weed Coordinator in Portland assured appellants that analysis files available at the Washington Office disclosing the effects of glyphosate application were incorporated into the final documentation. Appellants claimed that the Forest Service had not documented indirect effects described in a national survey which led to permitting the use of the chosen herbicide (Syracuse Environmental Research Associates, Inc., 1999):

Non-target plants could be damaged by unintentional application or drift ... The primary hazard to non-target terrestrial plants is from unintended direct deposition or spray drift. Unintended direct spray will result in exposure equivalent to the application rate. As discussed in the dose-response assessment for terrestrial plants (section 4.3.3), such exposures are likely to result in adverse effects to a number of plant species.

Nonetheless, an independent site visit and documenting photographic taken after the application (Wooten, 1999d) clearly show that the treatment primarily affected native plants. Noxious weeds alongside the road were missed completely while spraying over them onto native plants on streambanks as far as 30 feet beyond the road.

Herbicides may weaken native plants to the point where they are harmed. Pimentel (1999) notes that when herbicides did not kill non-target plants, plant pathogens increased in abundance up to 5-fold and attacks on plants increased up to 3-fold.

Indirect effects of herbicides include those effects that follow, like ripples, from the removal of both target and non-target vegetation (O’Brien, 1977). Transport of pesticides up food chains and concentration in lipid tissues of secondary consumers can result in exposures to fish 49,000 times higher than to target organisms (Reinert, 1967).

Many herbicides are resistant to breakdown, and when they do break down, the secondary byproducts can also have toxicity. For organochlorine pesticides, the chlorine-carbon bond resists breakdown by normal biochemical and physical processes and remains in the environment. Since the majority of organochlorines are foreign to nature, living organisms have developed few methods to detoxify them (Reinecke and Knackmuss, 1988; Nielsen, 1990).

So called “inert ingredients” are strongly implicated in the decline of Atlantic salmon. According to Montague (1999),

A study published in May in Environmental Health Perspectives, a U.S. government science journal, makes the case that insecticides sprayed on forests in eastern Canada in the mid-1970s led to a dramatic decline in the population of Atlantic Salmon (45% reduction in small salmon, 77% reduction in large salmon) (Fairchild, 1999). Salmon are born in fresh water but after 2 or 3 years they undergo physical changes called “smoltification,” after which they move downstream into salt water. Smoltification is controlled by hormones. Researchers believe the pesticide interfered with the hormones of the salmon, somehow disrupting smoltification, leading to the loss of large numbers of fish. The pesticide in question was called Matacil 1.8D. The “active ingredient” in Matacil 1.8D is aminocarb, which makes up about 25% of the insecticide by weight. The other 75% of Matacil 1.8D is an “inert ingredient” called 4-nonylphenol (4-NP for short). In laboratory tests, 4-NP is anything but inert. It is a powerful hormone disrupter. The authors of the study point out that many U.S. streams contain levels of hormone- disrupting chemicals comparable to the levels that they believe wiped out so many Atlantic salmon.

National Forests often exhibit weather patterns, which leads to serious drift problems. In the case of aerial applications, even under ideal weather conditions only approximately 25% of the herbicide reaches the target area, and it is estimated that less than 0.1% of pesticides ever reach their target pests, resulting in more than 99% of applied pesticides impacting the surrounding environment (Pimentel, 1999). Research has shown that less than one percent of the pesticides that are applied to crops actually reaches their target organism (Office of Technology Assessment, 1990, p. 104).

Finally, yet another unaccounted effect of herbicides is that they rarely solve the problem and require additional applications. Repeated chemical treatments can lead to acquiring herbicide resistance in weeds. Pesticide resistance has already been acquired by nearly 200 different species of plant pathogens and invasive plants, according to the National Research Council (1996, p.26). These studies indicate that the increase in pesticide-induced resistance suggests that dependence on pesticides as the dominant means of controlling pests is not a sustainable solution. The spiraling costs of treating resistant pests are estimated to account 10 percent of United States pesticide use (Pimentel, 1992).

Worldwide, there are over 216 herbicide-resistant weed species (Barber, 1999). Resistant species include flannel mullein (Verbascum thapsus), hoary white-top (Cardaria draba), Russian tumbleweed (Salsola kali) and diffuse knapweed ( Photo 3). These super-plants are being inadvertently bred through the excessive and regular use of too many herbicides along our highways. This leads to increasingly rapid spread along roads, and ultimately, abandonment of hopes for control. In Washington State, the Noxious Weed Control Board has had to delist a number of species because they became ubiquitous.

Case example: TES plant surveys The Forest Service Region 6 Forester has given direction to the Forests to protect threatened, endangered and sensitive (TES) species (FSM 2600). Plant survey guidelines for TES plant surveys insure that TES plants will be searched for prior to project implementation. In 1999 the Okanogan NF treated 5,956 acres with herbicides, under an Environmental Assessment (Okanogan NF, 1997) which claimed to have surveyed for sensitive plants prior to the project. Yet responses obtained through Freedom of Information Act (FOIA) requests by Kettle Range Conservation Group for these “surveys” revealed that this was not so. Many of the “surveys” were merely lists of plants for areas which were never visited; one of the “surveys” was a list of “cultural plant information” that listed edible plants (e.g., “Bailout”, 1998), some “surveys” were grazing allotment reviews or timber sale evaluations conducted at earlier dates, and several of the “surveys” were only performed through examining aerial photos, rather than actually sending a botanist to visit the site (e.g., “Cayuse”, 1993; “Redmill”, 1997). To conduct sensitive plant surveys using aerial photos in lieu of field surveys is a repudiation of the Regional Forester’s directive.

Any threatened or endangered plants existing on the sites would have been exterminated without the Forest Service or anyone else ever having known about it. In fact, valid TES plant surveys did find rare species in several of the treated areas, but these were not documented in the EA (e.g., TES surveys in McFarland Creek and Fawn Creek in 1998.

Case example: Herbicide persistence The Forest Service Region 6 currently has two available herbicides for use in the Pacific Northwest, glyphosate and picloram. These were the first chosen following lifting of the five-year injunction against their use in the Region (NCAP et al. v. Clayton Yeutter, et al., 1989). Picloram, however, has an extremely high ratio of toxicity to effective concentration for plants, and it has been recommended by the EPA to have its registration rescinded (Abramovitch, undated).

‘Beware,’ warns noted Colorado State University fisheries biologist Dr. Harold Hagen, ‘anytime you spray Tordon [picloram] it’ll come back to haunt you. It may be eight or ten years, but it will come back.’

In July 29, 1989, a weed-killing crew near Sheridan, Wyoming applied Tordon 22K a quarter mile from Hagen’s fish hatchery. The day afterward rain carried the herbicide into the trout ponds and within hours, more than 8,000 pounds of trout were dead, eventually killing all 15,000 pounds of the hatchery’s fish, and leaving Hagen out of business. ‘What they did was destroy the best trout hatchery in the country,’ said Hagen of the incident.

‘I wouldn’t let that stuff within 50 miles of my place,’ says Lew Grant, owner of the Fort Collins, Colorado based Piedmont Farms, when asked about Tordon. ‘The Soil Conservation District came out and treated circular patches of Canadian thistle on our place with Tordon. For at least seven or eight years we grew corn and other crops on that land with no problems. Then, nine years later, we planted sugar beets on it. They came up just fine but then they started dying in these big circle areas. I called Great Western Sugar, and they came out and analyzed them. They told me they were dying from Tordon. I couldn’t believe it. After nine years, it [Tordon] was still in the soil.’

Case example: cumulative effects O’Brien (1977) described how cumulative effects might be multiplied in an ecosystem:

A field study of glufosinate, for instance, found that it reduced the number of fungi in forest soils by 20 percent. Plant disease-causing fungi were among those species least impacted, while Trichoderma species, considered beneficial because they parasitize disease-causing species, were among the most sensitive to glufosinate. The researchers noted that use of glufosinate has, “important microbiological consequences”.

While glufosinate is not an herbicide Region 6 is currently using, I mention this study for two reasons. The first is that a soil whose cover and rooting vegetation have been reduced by logging or livestock or heavy recreation use might already have compromised biological functioning. The use of an herbicide that further reduces biological functioning is a cumulative impact.

The second is that this type of effect could be happening with the Region’s current use of any of its herbicides, but the agency is not looking for cumulative impacts, and would most likely not know they were occurring.

Solutions

Invasive species management projects must follow NEPA mitigation measures in 40 CFR § 1508.20, which require:

(a) Avoiding the impact altogether by not taking a certain action or parts of an action. (b) Minimizing impacts by limiting the degree or magnitude of the action and its implementation. (c) Rectifying the impact by repairing, rehabilitating, or restoring the affected environment. (d) Reducing or eliminating the impact over time by preservation and maintenance operations during the life of the action. (e) Compensating for the impact by replacing or providing substitute resources or environments.

In addition:

Section C. Stringent safety precautions for handling chemicals should be followed and applications should strictly adhere to established procedures.

In implementing vegetation management projects, all land management agencies must follow recognized safe chemical handling and spill procedures, accompanied by standardized documentation of accidents. In addition, safety precautions, established procedures and label directions must be strictly followed and chemical applications must comply with planning documents.

Chemical safety is regulated under the Occupational Safety and Health Administration (OSHA). The Forest Service is required to satisfied OSHA requirements (29 CFR 1910.1200(g)(8)) for maintaining Standard Material Safety Data Sheets:

The employer shall maintain copies of the required Material Safety Data Sheets for each hazardous chemical in the workplace and shall ensure that they are readily accessible during each work shift to employees when they are in their work area(s).

Under 29 CFR 1910.1200(e), the Forest Service is required to:

(1) ...develop, implement, and maintain at each workplace, a written hazard communication program which at least describes how the criteria specified in paragraphs (f), (g), and (h) of this section for labels and other forms of warning, material safety data sheets, and employee information and training will be met, and which also includes the following:

(1)(i) A list of the hazardous chemicals known to be present using an identity that is referenced on the appropriate material safety data sheet (the list may be compiled for the workplace as a whole or for individual work areas); and,

(1)(ii) The methods the employer will use to inform employees of the hazards of non-routine tasks (for example, the cleaning of reactor vessels), and the hazards associated with chemicals contained in unlabeled pipes in their work areas.

(2) Multi-employer workplaces. Employers who produce, use, or store hazardous chemicals at a workplace in such a way that the employees of other employer(s) may be exposed (for example, employees of a construction contractor working on-site) shall additionally ensure that the hazard communication programs developed and implemented under this paragraph (e) include the following:

(2)(i) The methods the employer will use to provide the other employer(s) on-site access to material safety data sheets for each hazardous chemical the other employer(s)’ employees may be exposed to while working ...

...(h)(1) Employers shall provide employees with effective information and training on hazardous chemicals in their work area at the time of their initial assignment, and whenever a new physical or health hazard the employees have not previously been trained about is introduced into their work area. Information and training may be designed to cover categories of hazards (e.g., flammability, carcinogenicity) or specific chemicals. Chemical-specific information must always be available through labels and material safety data sheets.

Under 29 CFR 1910.1200(d)(3)(ii), the Forest Service is required to evaluate one of the following sources for chemical hazard evaluation: Subpart Z, Toxic and Hazardous Substances, Occupational Safety and Health Administration (OSHA); or, “Threshold Limit Values for Chemical Substances and Physical Agents in the Work Environment,” American Conference of Governmental Industrial Hygienists (ACGIH) (latest edition).

Under 29 CFR 1910.1200(d)(4), the Forest Service is required to evaluate one of the following sources for establishing that a chemical is a carcinogen or potential carcinogen for hazard communication purposes: National Toxicology Program (NTP), “Annual Report on Carcinogens” (latest edition); International Agency for Research on Cancer (IARC) “Monographs” (latest editions); or 29 CFR Part 1910, Subpart Z, Toxic and Hazardous Substances, Occupational Safety and Health Administration.

The Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA), Section 12(a)2 states, “It shall be unlawful for any person to use any registered pesticide in a manner inconsistent with its labeling.” The EPA has the authority to register, restrict, or prohibit the use of pesticides, while States may offer additional protection. Pesticide registration decisions balance the risks involved with the benefits, after consideration of the nature of the chemicals, their toxicity and their environmental fate. The Washington Pesticide Control Act (RCW 15.58.150(2)(c)) states, “It shall be unlawful for any person to use ...any pesticide contrary to label directions.”

Recent Forest Service planning documents have failed to refer to available EPA information or even acknowledge the role of the Forest Service in weighing the benefits and risks of applications. There appears to be a fundamental rift in Forest Service compliance with NEPA regulations, which are primarily designed to provide disclosure, and FIFRA, which provides risk /benefit analyses. There is an explicit requirement in the use of registered herbicides that a beneficial use should be weighed against risks, a process that requires a quantified risk assessment completed by qualified individuals. Risk assessments and the valuable information they provide in quantifying harm, have been entirely lacking from recent Forest Service documents, even when risk assessments are already available.

Forest Service planning documents often make statements about herbicides that attempt to make health risks appear “unlikely” (Okanogan NF, 2000, p. 132). While it may be true that a minority of people would be affected, this is discriminatory to those who are affected.

The available literature for human toxicology is seldom accessed in Forest Service planning documents. Instead, gross generalizations are taken out of context and made to appear as if they are facts. For instance, in describing the human effects of the “most toxic surfactants” used with the herbicide glyphosate, the Okanogan NF (2000, p. 132), summarized a summary of an unpublished reference based on aquatic studies on a limited number of surfactants.

Had the Forest Service been genuinely committed to determining and disclosing effects, the Okanogan NF EA (2000) could have referenced numerous studies demonstrating harmful effects to wildlife, humans and the environment. But the Forest did not review toxicity tests of glyphosate, such as those which showed effects on salmonids (Mitchell et al., 1987; Wan et al., 1989), including sublethal effects of erratic swimming, labored breathing, altered feeding, migration and reproduction (Morgan et al., 1991; Liong et al., 1988), or those that found the combined effect of glyphosate and surfactants to be synergistic (Servizi et al., 1987) and involved pulmonary toxicity (Martinez and Brown, 1991). Despite their willingness to include unpublished anecdotes as references, the Forest Service rebuffs any efforts to provide published studies that show harm from herbicides (McDougle, 1999).

When available, it is certainly more appropriate to extrapolate human effects from studies on humans, rather than from animal toxicity models. While the EPA acknowledges that studies of glyphosate toxicity on humans are rare, those available portray this herbicide as far more toxic than the Forest Service reveals. Glyphosate exposures are associated with numerous deleterious effects including blurred vision, skin problems, heart palpitations, nausea, increased risk of miscarriages, premature birth, and non-Hodgkins lymphoma, or NHL (Cox, 1998). A case-controlled study linked NHL with exposures to pesticides including glyphosate (Hardell and Eriksson, 1999) and glyphosate has a death rate in humans of 10-20% during attempted suicides (Martinez and Brown, 1991).

Forest Service hand-waving that health risks are “unlikely” from some of their proposed herbicide treatments are not only lacking in credibility, but border on being false claims. When Monsanto Corporation was challenged by the New York State Attorney General for making for making false safety claims about their product glyphosate in 1996, Monsanto agreed out-of-court to stop advertising the product as “safe, non-toxic, harmless or free from risk.” (Cox, 1998).

Once Forest Service planning documents “determine” that health and environmental risks are “unlikely”, this becomes a handy excuse for sloppy implementation. Employees operate unaware of safeguards and shortcuts are emphasized over rules. Ad hoc variance from the planning documents are used whenever written procedures fail to anticipate real conditions, including unforeseen problems such as sudden changes in wind speed, public presence at a treatment site, lack of safety kits, vehicles in poor condition, etc.

During the use of chemical treatments, responsible personnel should always be available and preferably present at the treatment site. Managers need to anticipate the amount of staff time that will be necessary to implement projects. Staff need to be carefully chosen to maximize efficient use of personnel resources.

Case example: Okanogan NF spill risks go unstated The Okanogan NF EA for noxious weeds (1999) included an Appendix purported to be a “Spill / Release Control Plan”, however the actual plan was only included there by reference. In its place was a set of measures designed to lessen, but not necessarily undo, the impacts of a major spill in the case of an accident. Lacking from the EA was an analysis of the potential likelihood of such a spill. The likelihood for the 1990s, it turns out, was 100%, because on October 3, 1994, an herbicide truck contracted by the Tonasket Ranger District to spray weeds on the Okanogan NF crashed and turned over, spilling five gallons of herbicide into Nicholson Creek.

Turning to the national risk analysis for conducted for glyphosate (Syracuse Environmental Research Associates, 1996), one finds that no such spill scenario was ever considered there either. Instead, the report only analyzes risks from relatively minute spills, such as applicators splashing the material on their hands, while completely dismissing the possibility that a major spill could occur:

For this risk assessment, several very conservative scenarios are developed. As discussed below, most of these scenarios should be regarded as extreme, some to the point of limited plausibility.

Decision-makers and the public should have been made aware of these risks before serious harm occurred.

Case example: Okanogan NF estimates of risk are unwarranted The Okanogan NF 1997 Integrated Weed Management Environmental Assessment (EA) (1997) found that the use of herbicides in their plan would pose,

...minimal risks no greater than the risks predicted in the PNW Region FEIS for Competing and Unwanted Vegetation . . .

And that,

It is unlikely that any members of the general public would receive sufficient exposure to develop any adverse effects from the treatment . . .

Based on this sweeping conclusion, there were few safety precautions incorporated into the plan, except to claim that Forest Service workers would receive training, adhere to label directions, and follow the Forest Service Pesticide Use Manual (FSM 2150). The Forest prepared another EA (Okanogan NF, 1999), in which the same statements were cut and pasted, verbatim, into the new text.

But in proclaiming health effects minimal and unlikely, the Okanogan NF was ignoring its own analyses prepared by the Washington Office explicitly for reference by the agency in planning documents (Syracuse Environmental Research Associates, 1996). The preparers of the EA were apparently unaware of the fact that the Forest Service Washington Office had contracted a thorough literature review of glyphosate that went far beyond the scope of the 1988 FEIS.

The Syracuse Environmental Research report listed numerous reasons for stringent precautionary measures to be taken with glyphosate:

Glyphosate is a skin and eye irritant. This effect must be considered in the handling of commercial formulations. In addition, the toxicology of the combustion products of glyphosate has not been well characterized and this adds uncertainty to the risk assessment for brown-and-burn operations ...

Incidental occupational exposure may occur from improper handling or use of the herbicide or from accidental contamination of the skin or clothing by a spill. All of these scenarios can be modeled using Fick’s first law ... For this scenario, the estimated absorbed dose, using Fick’s first law, is approximately 0.00012 mg/kg ...If, however, the scenario involves contaminated clothing (e.g., the chemical spilled inside of gloves), which might be worn for a long time, absorbed doses could be much higher. For example, contaminated gloves worn for 1 hour would lead to an exposure 60 times greater than that described for the immersion scenario [i.e., 0.0069 mg/kg] ...

The toxicity of glyphosate is relatively well characterized in humans and experimental mammals, although the mechanism of action is not clear. The acute toxicity of glyphosate is relatively low, with oral LD50 values ranging from approximately 1,000 to 4,000 mg/kg. Most of the data regarding human exposure to glyphosate involves the consumption of large quantities of glyphosate during attempted suicides. The signs of toxicity are generally consistent with massive mucosal irritation and tissue degeneration. In addition, glyphosate may interfere with normal metabolic biochemical functions ...

Glyphosate contains small amounts of a nitrosamine, N-nitrosoglyphosate (NNG), and is metabolized, to a small extent, to aminomethylphosphonate (AMPA). The potential effects of these compounds are encompassed by the available toxicity data on glyphosate and glyphosate formulations.

During appeal of the 1999 EA for noxious weeds on the Okanogan and Colville National Forests, Gary Smith, Noxious Weed Coordinator of the Regional Office, went on record to deny requests to incorporate additional references to the toxicity of glyphosate made since the Syracuse Environmental Research Associates report, stating,

I know that the peer review process used by SERA for Forest Service pesticide risk assessments includes qualified scientific experts outside of the Forest Service. Their comments would already be incorporated into the final document you have retrieved [the EA].

Besides being false, the statement indicates a disregard for concerns of public safety.

Solutions

1) Material Safety Data Sheets 2) An approved plan of Forest safety precautions 3) Exposure incident reporting forms 4) Herbicide label directions 5) Pesticide background sheets

Section D. Treatments should receive adequate public notification.

NEPA (§1506.6 (ix)) specifies that notice of pending actions may include posting of notices on and off site in the area where the action is to be located. In light of the known and potential effects, public notification should accompany all herbicide treatments on public lands. Treatment locations should be available on request by visitors and treated areas should be posted with large, visible signs before, during and following the treatment for the remainder of the season. Workers in treated areas should be notified of chemical treatments and given opportunities for alternative assignments.

Signed areas should have large, readable signs to insure maximum protection of the public and workers in the area, particularly for sensitive individuals and children. Signs should include information indicating who to contact in case of injury, should be dated and should be checked for an entire season following treatment. To date, the Forest Service does not have an adequate procedure for signing areas treated with herbicides, using tiny, stapled pieces of paper located out of sight from most traffic.

The use of signs to protect workers and the public is a standard practice, which the Forest Service would do well to heed. The protection of worker and public safety should be part of everyone’s responsibility. The Department of Pesticide Regulation of California (1999) cites a reasonable set of regulations that can protect worker safety:

Under the reporting regulations, after every pesticide application pest control operators must give farmers a written notice that includes the date and time the application was completed and the restricted-entry and preharvest intervals. The restricted-entry interval is the period required between a pesticide application and when workers may re-enter the field. The preharvest interval is the time between an application and the earliest date the crop may be harvested. Farmers are required to post signs at fields treated with certain pesticides. The signs must include information on pesticide use including when it is safe for workers to re-enter the treated area. Farmers must also make records of pesticide use available to workers. Use reporting makes this information readily available.

Prior to treating areas with herbicides, public notification should also occur in local newspapers, on local public radio, Forest Service office bulletin boards, Forest Service web sites and any other readily available locations.

Despite the best efforts to protect workers and the public, injuries can occur. Even with the use of adequate warning signs, in some cases projects will result in unintended harm to sensitive members of the public. There should be an approved Forest Chemical Safety Plan to refer to in case of an accident or claim made by a member of the public. A recommended Chemical Safety Plan is included in Appendix B.

Solutions