RESIDUES OF FORESTRY HERBICIDES IN 
PLANTS OF INTEREST TO NATIVE AMERICANS: 
PHASE ONE - DEVELOPMENT OF METHODOLOGIES 
AND PILOT SAMPLING 


By


R. Segawa, A. Bradley, P. Lee, D. Tran, J. White, J. Hsu, and K. Goh


April 1997


Environmental Hazards Assessment Program

California Environmental Protection Agency
Department of Pesticide Regulation
Environmental Monitoring and Pest Management Branch
1020 N Street, Room 161
Sacramento, California 95814-5624


EH 97-01


Executive Summary 
of Report EH 97-01 -- Entitled

"Residues of Forestry Herbicides in Plants of Interest to Native Americans:
Phase One - Development of Methodologies and Pilot Sampling"

Environmental Monitoring and Pest Management Branch
Division of Enforcement, Environmental Monitoring,
and Data Management


PURPOSE 

This project is the first phase of a two-phase study to assess exposure of basketweavers to forestry herbicides. Its purpose is to develop sampling and analytical methodologies to measure residues of three herbicides in plant material, to collect plant material from inside and outside treatment areas in several national forests in California, and to provide an estimate of the occurrence of these herbicides in the collected plant material. This information will be used to design the second phase of the study: a comprehensive sampling and analytical study of selected plants.

BACKGROUND 

The California Indian Basketweavers Association and members of some Native American tribes have expressed concern to the Department of Pesticide Regulation (DPR) about potential exposure of basketweavers and gatherers to herbicides used in forest vegetation management. The basketweavers have also expressed their concerns to the U.S. Environmental Protection Agency and to the U.S. Forest Service. These unique exposure scenarios are not characterized in the risk assessment of these herbicides.

When trees from an area of national forest are removed (due to forest fires or logging), the U.S. Forest Service prepares the site and replants conifers. As part of its national forest site preparation and conifer release operations, three herbicides are used to control plants that compete with the conifers: glyphosate, hexazinone, and triclopyr. The U.S. Forest Service requested DPR's assistance in assessing exposure of basketweavers to these three forestry herbicides.

STUDY METHODS 

DPR scientists selected specific plant species and plant parts to sample based on discussions with Native Americans and site observations. The following plant parts and plants were selected for sampling in this study: stem or shoot samples (bitter cherry, buckbrush, deer brush, dogwood, willow), root samples (bracken fern, soaproot), leaf or stalk samples (deergrass, golden fleece, pearly everlasting), and fruit or nut food samples (elderberry, manzanita, oak). Plant samples were collected from 17 different treatment units, representing the five major types of treatments (pesticide/application method combination) in Lassen, Eldorado, Stanislaus, and Sierra National Forests. DPR staff collected samples inside and outside treatment units after the herbicide treatment had started killing target plants following applications which occurred in late November 1995 through spring 1996.

California Department of Food and Agriculture's Center for Analytical Chemistry developed analytical methods for these plant species and plant parts for hexazinone, glyphosate, and triclopyr. In order to analyze more plant species, the analytical methods used for this phase of the study only indicate if herbicides were detectable, not the exact concentration.

RESULTS 

Of the 14 plants selected for sampling, 13 plants had sufficient plant material to collect (oak acorns were not collected). For these 13 plants, staff collected 211 samples. Of the 92 samples collected from inside treatment areas, 49% contained detectable residues of herbicides. Inside treatment areas, residues continued to be detected for several months after application. Of the 119 samples collected outside treatment areas, 3% contained detectable residues of herbicides.

The lowest concentration that was reliably detected was 0.10 parts per million (ppm) for glyphosate, 0.05 to 0.10 ppm for hexazinone, and 0.01 to 0.07 ppm for triclopyr in various plants.

CONCLUSIONS AND RECOMMENDATIONS 

Since all three herbicides are detectable in plants located inside and outside the treated areas, DPR staff recommends that the second phase of this study include all three herbicides and that samples should be collected from inside and outside treated areas. Dissipation monitoring should be conducted for at least eight months or until residues are not detectable. The second phase of this study should maximize number of treatment unit samples (rather than number of samples within a treatment unit) when monitoring off-site movement.

DPR staff recommends that dissipation monitoring be conducted for one brush-type plant, one foliage-type plant, one root-type plant and one food-type plant. When available, information about plant availability from this study will be used to determine which plant to sample.

The second phase of the study will include additional development of selected analytical methods so that specific concentrations can be reported. Using that quantitative information, DPR scientists can estimate the dissipation of these herbicides on selected plants in the second phase. Off-site movement will be monitored in phase two; DPR staff recommends that the qualitative analytical methods (methods that only detect the presence or absence of a chemical, not the concentration) developed in this project be used to conduct subsequent off-site movement monitoring.



John S. Sanders, Ph.D., Chief                                                     April 1997
Environmental Monitoring and
Pest Management Branch
(916) 324-4100

ACKNOWLEDGMENTS

We would like to express our deepest thanks to the following tribal consultants living near or use the national forests who generously shared valuable information on plants of importance, their methods of gathering, processing, and uses. In the Eldorado National Forest the consultants were Jeff Brownell, Louise Griffin, Fred Short, Darlene Brown Toyeko, and Fred Velasquez. In the Sierra National Forest the consultants were Edith Beecher, Martha Beecher, Melba Beecher, Maria Cohoe, Esther Ortiz, Norma Turner, and Gladys McKinney. In the Stanislaus National Forest the consultants were Jennifer Bates, Reba Fuller, David Lingo, Phyllis Montgomery, and Lawrence Wilson. We also thank the survey respondents and supporters: Alec John A-Hiokatoo, Mary Baugh, Doris Case, Ramona Dutschke, Karen Dutschke, Alvin Walloupe, and Helen Wilson. We also appreciated the supports of tribal chairs: Thane V. Batty (Big Sandy), Susie L. Long (Yurok Tribe), and Cheryl A. Seidner (Wiyot Tribe).

We would like to thank the California Indian Basketweavers Association (CIBA) for bringing to our attention the concerns of exposure to forestry herbicides. We especially appreciate the guidance and consultation from Jennifer Bates (CIBA Director), Sara Greensfelder (CIBA Executive Director), Steve Nicola, Emily Schwalen, Kathy Wallace, Denise Davis, and Margaret Mathewson.

Special thanks to personnel of the U.S. Forest Service for their funding and assistance, especially the program managers, silviculturists, botanists, and tribal relations managers in each forest. Without them this project would have been impossible. We thank John Borrecco at the U.S. Forest Service San Francisco office for arranging the financial support, providing many hours of consultation and establishing contacts for us with each national forest. We also thank Sonia Tamez and Nancy Lollar for facilitating our collaboration with the tribal relations managers at each forest and processing grant applications, respectively.

The following people were most generous in helping us with various activities in coordinating consultations with local Native Americans, untiringly taking us on many field site visits, helping us identify plant species, gathering plant materials, and providing consultations:

Eldorado National Forest - Dave Bakke, Robin Barron, Bob Carroll, Krista Deal, Carmen Galindo, Denise McLemore, Judy Rood, and Mike Taylor

Lassen National Forest - Marvin Benner and Lauren Payne

Sierra National Forest - Dirk Charley, Joanna Clines, Lori Planas, and Mark Smith

Stanislaus National Forest - Phyllis Ashmead, Jim Behm, Mike Brown, Jennie Haas, Jayne Montoya, Mike Rutty, John Schmechel, Joe Sherlock, Jim Mullin and Lynn Webb.

Thanks also to all environmental monitoring and laboratory personnel who participated in this project, particularly Carissa Ganapathy, Heinz Biermann, Terri Barry, Nancy Miller, Madeline Brattesani, Lianna Easter and Cathy Cooper. We appreciate the continuing support and guidance of John Sanders, Tobi Jones and Paul Gosselin on this project.


DISCLAIMER

The mention of commercial products, their source, or use in connection with material reported herein is not to be construed as an actual or implied endorsement of such product.

TABLE OF CONTENTS
                                                                                                                                page 

EXECUTIVE SUMMARY.........................................................................................I

ACKNOWLEDGMENTS...........................................................................................iv

DISCLAIMER.............................................................................................................v

TABLE OF CONTENTS............................................................................................vi

INTRODUCTION.......................................................................................................1

U.S. FOREST SERVICE HERBICIDE PROGRAMS

Treatment Site Description.........................................................................................2

Herbicide Application Methodology..........................................................................3

MATERIALS AND METHODS

Plant Selection...........................................................................................................4

Sampling Site Selection.............................................................................................4

Sampling Methodology..............................................................................................5

Analytical Methodology............................................................................................6

SAMPLING RESULTS

Results by Herbicide and Application Method.........................................................9

Results by Plant Species..........................................................................................10

Results by Location..................................................................................................10

Results Over Time...................................................................................................11

CONCLUSIONS AND RECOMMENDATIONS....................................................11

REFERENCES........................................................................................................13

Appendix 1: Discussions with Native Americans

INTRODUCTION 

The use of native plant materials is a tradition among Native American tribes in California. Native Americans use many different plants for food, medicine, basketry, and other cultural activities. The California Indian Basketweaver Association has alerted the Department of Pesticide Regulation (DPR), the U.S. Environmental Protection Agency, and the U.S. Forest Service of potential exposures to forestry herbicides that may be occurring to Native Americans through the use of plant materials. These unique exposure scenarios are not characterized in the risk assessment of these herbicides. The U.S. Forest Service has established programs within their national forests to work with tribal representatives to identify and protect collecting areas from herbicide spraying. However, not all Native Americans participate in these programs, and may collect plant materials in unidentified locations. Additionally, Native Americans are concerned that the protective measures may not be sufficient.

Glyphosate, hexazinone, and triclopyr are herbicides that are used for site preparation and release operations in conifer tree plantations in California's national forests. Glyphosate is a nonselective, postemergent contact herbicide. Triclopyr is a systemic herbicide used extensively to control woody weeds and many broad-leaf weeds. Hexazinone is a contact and residual herbicide. All three herbicides are applied from spring through fall months. Applications of these herbicides are made by commercial applicators under contract with the U.S. Forest Service. While most applications are made by ground application using backpack sprayer, some large acreage projects may use aerial application of granular products.

The U.S. Forest Service has requested the assistance of DPR in assessing exposure of plant gatherers and basketmakers to forestry herbicides. In order to estimate potential exposure, the residue concentrations of forestry herbicides must be determined.

This study is being conducted in two phases. This report presents the results of the first phase and makes recommendations for the second phase. The objectives of the first phase are:

1. Identify sites within the Eldorado, Lassen, Sierra, and Stanislaus National Forests that are scheduled for herbicide treatment for fall 1995 and spring 1996.

2. Catalogue sites for plant collection and residue analysis in Eldorado, Lassen, Sierra, and Stanislaus National Forests, both within and outside treatment areas, where relevant plant species may be exposed to forestry herbicides.

3. Develop preliminary sampling and analytical methodologies to measure residues of glyphosate, hexazinone, and triclopyr in candidate plant materials.

4. Conduct pilot sampling to make a preliminary estimate of the occurrence of herbicide residues in candidate plant materials.

The second phase will consist of a comprehensive sampling and analytical study of selected plants.

U.S. FOREST SERVICE HERBICIDE PROGRAMS 

Treatment Site Descriptions 

The U.S. Forest Service currently conducts herbicide treatment programs within four of California's National Forests: Eldorado, Lassen, Sierra, and Stanislaus (Figure 1). Herbicide treatments occur in areas that are being reforested due to recent fires or logging. Herbicides are applied in reforested areas to reduce competing vegetation such as bear clover, manzanita, shrubs and grasses.

The characteristics of individual treatment areas varied greatly. The size of the individual treatment areas varied from one to hundreds of acres. The elevation varied from 3,000 to 7,000 feet. The slope varied from 0 to 100% or more. The types and abundance of plants of interest within the treatment areas also varied greatly. The treatment areas examined contained from none to five plant types in sufficient abundance for sample analysis. Many of the treatment areas contained protected areas, usually streams or drainages, which were left untreated. In addition, some plants such as oaks were intentionally avoided. Plant samples were collected from both treated and untreated areas. Complete descriptions of the treatment areas are given in the Environmental Impact Reports or Environmental Assessments available from the U.S. Forest Service.

Herbicide Application Methodology 

For the purposes of this study, herbicide applications have been classified by active ingredient and method of application. Currently, the U.S. Forest Service uses four different products, each containing one of three active ingredients. These herbicides can be applied from the ground or air. The following types of applications are currently used:

            hexazinone/air:                     Pronone®/air
            hexazinone/ground:               Pronone®/ground
                                                          Velpar®/ground
           glyphosate:                             Accord®/ground
           triclopyr:                                Garlon®/ground

In addition, glyphosate and triclopyr may be combined and applied as a tank mix.

Plant samples were collected from each of the five major types of treatments, as detailed below:

            Lassen - Pronone®/ground
            Eldorado - Garlon®/ground
            Stanislaus (Paper Project) - Pronone®/air, Accord®/ground, Garlon®/ground
            Stanislaus (Hamm-Hasloe) - Pronone®/ground, Accord®/ground,
            Garlon®/ground
            Stanislaus (Groveland Conifer Release II) - Accord®/ground
            Sierra - Velpar®/ground, Accord®/ground

The first treatments occurred in late November 1995 and continued through spring 1996. All treatments were conducted by commercial applicators, under contract to the U.S. Forest Service. Most applications were directed treatments; individual plants were treated rather than entire areas. Therefore, the application rates varied with the number of target plants within treatment areas. Typical application rates for these herbicides were 1.0 - 1.5 pounds glyphosate per acre, 3.0 - 3.5 pounds hexazinone per acre, and 1.0 - 1.5 pounds triclopyr per acre.

MATERIALS AND METHODS 

Plant Selection 

Specific plant species and plant parts were selected for study based on discussions with Native Americans (Appendix 1) and site observations. The following fourteen plants and parts were selected for study.

            Anaphalis margaritacea (pearly everlasting-foliage)
            Arctostaphylos spp. (manzanita-berries)
            Ceanothus cuneatus (buckbrush-shoots)
            Ceanothus integerrimus (deer brush-shoots)
            Chlorogalum pomeridianum (soaproot-bulb)
            Cornus spp. (dogwood-shoots)
            Ericameria arborescens (golden fleece-foliage)
            Muhlenbergia rigens (deergrass-stalks)
            Prunus emarginata (bitter cherry-shoots)
            Pteridium aquilinum var. pubescens (bracken fern-roots)
            Quercus spp. (oak-acorns)
            Rorippa nasturtium-aquaticum (watercress-foliage)
            Salix spp. (willow-shoots)
            Sambucus mexicana (elderberry-berries)

Some commonly used plants such as redbud and wild onion were not selected for study because they did not occur within treatment areas.

Sampling Site Selection 

Samples were collected from 17 different treatment units. The specific units were chosen primarily for the abundance and types of plants within the units as well as the treatment method. Other selection factors included accessibility (those with easier access were chosen) and location (different regions and elevations were chosen). Descriptions and locations of each treatment unit sampled are given in Table 1 and Figures 2 - 5.

Sampling Methodology 

The dates of sample collection were dictated by treatment schedules and plant growth. The first treatments occurred in late November 1995 and continued through spring 1996. Samples were collected after the herbicides had started killing target plants. Sampling continued until seasonal growth was sufficient to allow sampling for each plant type. Those plants treated showing apparently severe toxicity (all leaves brown in color) were assumed to contain herbicide residues and were not sampled.

Samples were collected from plants located outside as well as inside treatment areas. Samples from outside treatment areas were mainly collected in or near streams and drainages adjacent to treatment areas. The U.S. Forest Service requires a buffer of 10 - 300 feet on either side of the drainages, depending on the application method and type of drainage. Plants from outside treatment areas were collected from 5 - 50 feet from the application boundary.

Each sample consisted of plant material from 1 to 20 plants. The specific plants sampled were selected based on availability and accessibility. Many of the plant species occurred infrequently within or near treatment areas. Therefore, samples were collected from accessible areas that had a sufficient quantity of plants for analysis; random selection was not employed. Brush samples (bitter cherry, buckbrush, deer brush, dogwood, willow) consisted of the plant shoots or stems. The leaves were removed from all brush samples, except buckbrush. Root samples (bracken fern, soaproot) consisted of the plant root or bulb. Foliage samples (deergrass, golden fleece, pearly everlasting) consisted of the foliage or grass stalks. Food samples (elderberry, manzanita, oak) consisted of the fruit or nut. Those plants that had all brown leaves were assumed to contain herbicide residues and were not sampled. Each sample collected was subjectively rated for toxicity, with one indicating apparent toxicity (a few all-brown leaves), two indicating possible toxicity (a few partially-brown leaves) and three indicating no apparent toxicity (no brown leaves).

All samples were collected and stored in glass jars. All plant material was placed into the jars while wearing disposable gloves; gloves were changed with each sample. All sampling tools such as clippers, and trowels were cleaned with water and alcohol after each sample. All samples were cooled immediately after collection, then kept refrigerated until analysis. All samples were photographed to confirm the species and condition (phytotoxicity) of the sample.

No major problems were encountered in developing the sampling methodology. The worst problem was finding sufficient material for some of the plants, particularly bitter cherry, deergrass, dogwood, elderberry, and oak. These plants occurred very infrequently within or near treatment areas. No special tools or materials were required to collect the samples. The foliage samples (deergrass, golden fleece, pearly everlasting) took the least time sample and therefore had the lowest sampling cost. The root samples took the most time to sample and had the highest sampling cost. The only potential problem in sampling during phase two is the possibility of sample contamination. For example, not changing gloves between samples or placing collected plant material on the ground could cause inadvertent contamination of samples. The standard operating procedures developed for phase two will address this issue specifically.

Analytical Methodology 

The analytical methods have only been partially validated. Therefore, the results of individual samples are expressed as a range of concentration. The analytical methods will be fully validated as part of phase two.

The primary laboratory for this study was the California Department of Food and Agriculture's Center for Analytical Chemistry. Analytical methods were developed for a variety of plant species and parts for three chemicals: hexazinone, glyphosate and triclopyr. The reporting limits (concentration that can be reliably detected) were determined using EHAP standard operating procedure QAQC001.00 (Segawa, 1995).

Complete validation of the laboratory methods (as described in EHAP standard operating procedure QAQC001.00 [Segawa, 1995]) was not conducted for phase one of this study. The pilot sampling conducted in phase one only determined the presence or absence of herbicides in plant samples. Therefore, a quantitative determination of residue levels was not necessary. By postponing method validation, a greater number of samples were collected and analyzed for the pilot study. In addition, only plants selected for further study will need to be validated. Funds were not expended to validate methods that may not be used.

Glyphosate Analytical Method 

Glyphosate samples were analyzed using a method developed by Monsanto, but significantly revised as described here. Twenty grams from each plant sample were first chopped and homogenized with dry ice. After the dry ice sublimed, the plant material from each sample was extracted with 0.1 N hydrochloric acid. The aquesous extract was then concentrated on a column containing a chelating agent, Chelex 100. The eluent from this column was cleaned up by passing it through two other columns, one anion exchange and one cation exchange column. After cleanup, the eluent was evaporated almost to dryness, then redissolved using water. The final extracts were analyzed using a high pressure liquid chromatograph with a Pickering post-column reactor and fluorescence detector. The reporting limit (concentration that can be reliably detected) for each plant type was 0.1 ppm (fresh weight basis). Reporting limits for each plant and herbicide are given in Table 2.

This method was the most difficult to develop because of glyphosate's high water solubility, which caused numerous compounds to be co-extracted. These other compounds interfere with the analysis and foul the instrument. Therefore, most other glyphosate methods, including Monsanto's, have higher reporting limits than the method employed for this study and the instrument column must be replaced frequently at a cost of several thousand dollars. In addition, the dirty samples cause the instrument response to be unstable; frequent standards must be analyzed to verify calibration. The method for this study uses a greater amount of plant material and an extensive cleanup procedure. Therefore, the reporting limits are lower, the instrument column lasts ten times longer, and the instrument response is very stable. This method takes the longest to perform and requires some expensive materials; it has the highest cost of the three methods.

Hexazinone Analytical Method 

Twenty-five grams from each plant sample were first chopped and homogenized with dry ice. After the dry ice sublimed, the plant material from each sample was extracted with acetonitrile. The acetonitrile extract was evaporated almost to dryness. The extract was cleaned up by placing it on an aminopropyl bond elut® column and eluting with a methanol/methylene chloride mixture. The final extract was analyzed using a high pressure liquid chromatograph with an ultraviolet detector and reverse-phase column. The reporting limits ranged from 0.05 to 0.2 ppm (fresh weight basis). Reporting limits for each plant and herbicide are given in Table 2.

There were few problems encountered with this method. The greatest difficulty was analyzing some of the golden fleece samples. Golden fleece and soaproot can have compounds that are co-extracted and interfere with the analysis. Additional cleanup was not successful. However, this seems to be a problem only with late-season golden fleece. For the above reasons, hexazinone in golden fleece and soaproot were analyzed with a gas chromatograph and nitrogen/phosphorous detector. This method takes the least time to perform and requires no special materials; it has the lowest cost of the three methods.

Triclopyr Analytical Method 

Ten to forty grams from each plant sample were first chopped and homogenized with dry ice. After the dry ice sublimed, the plant material from each sample was extracted with a benzene/sulfuric acid mixture. An aliquot of the benzene extract was cleaned up by first extracting with a sodium bicarbonate solution and then ethyl ether. The aqueous solution was reacidified with sulfuric acid, then reextracted with methylene chloride. The extract was concentrated and derivatized with diazomethane. The final extracts were analyzed using a gas chromatograph with an electron capture detector. The reporting limits ranged from 0.01 to 0.07 ppm (fresh weight basis). Reporting limits for each plant and herbicide are given in Table 2.

The low reporting limits for triclopyr were possible because of the efficient cleanup and derivatization that is possible with this herbicide. The solubility of triclopyr changes dramatically with pH. Triclopyr is soluble in organic solvents under acidic conditions and soluble in water under basic conditions. By changing the pH, both water-soluble and water-insoluble interferences can be removed. In addition, derivatization with a methylating agent (diazomethane) gives an enhanced instrument response. No problems were encountered with this method. This method is less expensive than glyphosate and more expensive than hexazinone. Of the three herbicides, triclopyr is in the middle for both time to complete and cost of materials.

SAMPLING RESULTS 

Samples were collected for 13 of the 14 plants selected for study. Samples of acorns could not be obtained due to low acorn production. For the other 13 plants, 211 samples were collected. Of the 92 samples collected from inside treatment areas, 49% (45 samples) contained detectable residues of herbicides. Only plants that were fully or partially healthy were sampled. Plants with all brown leaves were assumed to contain herbicide residues and were not sampled. Had all plants inside treatment areas been sampled, the proportion of positive samples would likely have been higher. Additionally, the proportion of positive samples decreased over time, as discussed later. Of the 119 samples collected from outside treatment areas 3% (4 samples) contained detectable residues of herbicides. Samples from outside treated areas were collected 5 - 50 feet from the treatment area boundaries. Results of the sampling are summarized below.

Plant health showed some correlation with herbicide detection. Each sample collected was subjectively rated from one to three for toxicity. For samples collected inside treatment areas, of the samples rated one (apparent toxicity, a few brown leaves) 65% were positive. Of the samples rated two (possible toxicity, a few partially-brown leaves) 38% were positive. Of the samples rated three (no apparent toxicity, no brown leaves) 36% were positive.

None of the treatment units selected for sampling had acorns suitable for analysis. Several other treatment units were also surveyed, but they also did not have any suitable acorns. The few acorns that were found contained worms and/or mold which made them unsuitable for analysis.

Results by Herbicide and Application Method 

Table 3 shows the sampling results by herbicide and application method. Overall, the three herbicides show similar results. Inside treatment areas, 52%, 47%, and 50% of the samples were positive for glyphosate, hexazinone and triclopyr, respectively. Outside treatment areas, 5%, 2%, and 3% were positive for glyphosate, hexazinone and triclopyr, respectively.

Hexazinone, did show dramatic differences with application method. Inside treatment areas, 94% of the samples were positive in areas where hexazinone was applied by air and 15% of the samples were positive in areas where hexazinone was applied by ground. Some of the factors that may account for the difference include time of sampling and which plants were targeted. Some of the samples collected from ground treatment areas (Lassen) were collected several months after treatment, as opposed to several weeks after treatment for other areas. Lassen applications were conducted in the fall of 1995, while the other treatments were conducted in the spring of 1996. Most of the ground treatments were directed applications, that is specific plants were treated. Aerial treatments were broadcast applications, where entire areas were treated. In addition, only one aerial treatment unit was sampled, while eight ground treatment units were sampled. The results from the aerial treatment unit could be anomalous. The sampling in phase two should give more information on this difference.

Results by Plant Species 

Table 4 shows the sampling results by plant. The disparity in number of samples is due to the availability of plants. For example, few bitter cherry plants could be found in any of the treatment units. The same is true for the disparity between the number of samples inside versus outside treatment areas. Watercress and willow plants were found only in drainage and stream areas (outside treatment areas).

While there was variability in the proportion of positive samples, no pattern was apparent. Inside treatment areas, the proportion of positive samples ranged from 25 - 75% (for plant species that had more than two samples). Even similar plant types showed similar variation. For brush-type plants, bitter cherry had 25% positive samples and buckbrush had 71% positive samples. For foliage-type samples, pearly everlasting had 25% positive samples and golden fleece had 75% positive samples. Outside treatment areas, the four positive samples each came from a different plant species.

Results by Location 

Table 5 shows the results by treatment unit. Inside treatment areas, 78% (7 of 9) of the units had positive samples (for units that had more than two samples). Outside treatment areas, 20% (2 of 10) of the units had positive samples (for units that had more than two samples). Two locations were markedly different from the others: Matsen 38/42 and 027-019, both in the Stanislaus National Forest.

Inside treatment areas, Matsen 38 and 42 are stands within the same aerial treatment unit in the Stanislaus National Forest. As discussed earlier, this unit had 94% positive samples, which is much higher than proportion of positive samples collected inside other units. The combined total of positive detection inside other treatment units was 37%.

Outside treatment areas, stand 027-019 in the Stanislaus National Forest had three of the four positive samples. All positive samples were collected near a live stream inside the treatment unit. This stream had a buffer of 50 feet. There were no obvious reasons why most of the positive samples outside treatment areas came from this unit. This unit is one of the steeper units, but at least one other was just as steep (Murphy 38), and had no positive detections. Soil type and ground cover are also similar to other units. The positive samples could be due to drift or overspray, but this application was not observed by DPR staff and no samples were collected at the time of treatment. Therefore, no evidence supporting or opposing this possibility is available.

Results Over Time 

Table 6 shows the results over time. Inside treatment areas, the proportion of positive samples decreased over time. Within the first eight weeks of treatment, 74% of the samples were positive; no positives were detected 41 - 48 weeks after treatment. A comparison of the frequency of positive detections between the three herbicides was inconclusive due to insufficient number of samples.

CONCLUSIONS AND RECOMMENDATIONS 

Sampling results indicate that all three herbicides (glyphosate, hexazinone, triclopyr) are detectable in plants of interest to Native Americans and occur with comparable frequency.

            Recommendation: All three herbicides should be studied in phase two. 


Herbicide residues are detectable in plants that are located within treated areas as well as outside treated areas, but the frequency of detection is much higher within treated areas (49% vs. 3%).

            Recommendation: Samples should continue to be collected from inside and 
            outside treated areas (dissipation and off-site monitoring) for phase two. 

            Recommendation: Complete method validation is required to obtain 
            quantitative laboratory results. Quantitative results and method validation 
            are only necessary for those plants chosen for dissipation monitoring. 
            Off-site monitoring can be accomplished with qualitative analytical methods. 

Due to an insufficient number of samples for some plants, a definitive comparison of plant species cannot be conducted.

            Recommendation: Dissipation monitoring should be conducted for one 
            brush-type plant (buckbrush or deer brush), one foliage-type plant 
            (golden fleece or pearly everlasting), one root-type plant (bracken fern or soaproot), 
            and one food-type plant (elderberry or manzanita berry). Bitter cherry, deergrass, 
            dogwood, oak-acorn, watercress, and willow should not be monitored for 
            dissipation because they occur infrequently inside treated areas. Based 
            solely on frequency of occurrence, buckbrush and golden fleece should be 
            sampled instead of deer brush and pearly everlasting. Bracken fern and 
            soaproot had similar frequency of occurrence. There were insufficient 
            elderberry samples for comparison to manzanita. Other factors such as 
            cultural importance and cost will also influence the choice of plants. 

Outside treatment areas, residues are detected infrequently (3% of the samples and 20% of the treatment units). It is likely that detection of herbicides in plants outside of treatment areas is due to unusual circumstances of a particular site or application, rather than a systematic occurrence.

            Recommendation: Off-site monitoring should be maximized for number of 
            treatment units sampled rather than number of samples within a treatment unit. 

Inside treatment areas, residues continue to be detected for several months after application.

            Recommendation: Dissipation monitoring should be conducted for at least 
            eight months or until residues are not detectable. This may require monitoring 
            to be extended for two growing seasons. 

REFERENCES 

Segawa, R. 1995. Standard Operating Procedure QAQC001.00, Chemistry Laboratory Quality Control. Department of Pesticide Regulation.



Table 1. Sampling sites treated with herbicides in four National Forests in California 


forest 
district 
county  		location 
twn/rng-sec 
stand  		treatment date 
herbicide1  		application method 
Eldorado  Pacific  El Dorado  11N/13E-15  501-044  6/17/96  Garlon  ground 
Lassen  Almanor  Plumas  29N/05E-12  Domingo 17  11/28/96  Pronone  ground 
Almanor  Plumas  29N/05E-12  Domingo 18  11/28/96  Pronone  ground 
Almanor  Plumas  29N/04E-21  Mineral 01  11/29/96  Pronone  ground 
Almanor  Plumas  29N/04E-21  Mineral 04  11/29/96  Pronone  ground 
Almanor  Plumas  29N/04E-20  Mineral 30  11/29/96  Pronone  ground 
Sierra  Kings River  Fresno  10S/25E-15  403-010  3/14/96  Velpar  ground 
Kings River  Fresno  10S25E-21  403-011  5/17/96  Accord  ground 
Pineridge  Fresno  10S/23E-03  339-005  3/16/96  Velpar  ground 
Pineridge  Fresno  10S/23E-15  339-007  5/23/96  Accord  ground 
Stanislaus  Groveland  Mariposa  02S/17E-01  027-019  6/15-18/96  Accord/Garlon  ground 
Groveland  Mariposa  02S/17E-02  027-038  3/25/96  Pronone  ground 
Groveland  Mariposa  02S/19E-28  038-010  4/8-12/96  Accord  ground 
Groveland  Mariposa  03S/20E-04  040-138  4/10/96  Accord  ground 
Mi Wok  Tuolumne  01N/16E-14  Matsen 38/42  3/15/96  Pronone  air 
Mi Wok  Tuolumne  01N/16E-36  Murphy 23  3/15/96  Pronone  air 
Mi Wok  Tuolumne  01N/16E-02  Murphy 38  5/7-24/96  Accord/Garlon  ground 


1 Garlon = triclopyr; Pronone = granular hexazinone; Velpar = liquid hexazinone; Accord = glyphosate


Table 2. Analytical method reporting limits for glyphosate, hexazinone, and triclopyr. The reporting limit is the lowest concentration that can be reliably detected. 

reporting limit (ppm) 
plant  glyphosate  hexazinone  triclopyr 
bitter cherry shoot  0.10  0.10  0.07 
bracken fern root  0.10  0.07  0.03 
buckbrush shoot  0.10  0.10  0.10 
deer brush shoot  0.10  0.10  0.03 
deergrass stalk  0.10  0.05  0.05 
dogwood shoot  0.10  0.10  0.03 
elderberry fruit  0.10  0.05  0.01 
golden fleece foliage  0.10  0.10  0.07 
manzanita berry  0.10  0.05  0.03 
oak acorn  0.10  0.10  0.01 
pearly everlasting foliage  0.10  0.20  0.07 
soaproot bulb  0.10  0.05  0.03 
watercress foliage  0.10  0.05  not analyzed 
willow shoot  0.10  0.05  0.03 




Table 3. Summary of sampling results by herbicide and application method for all plant samples taken inside or outside treatment boundaries in the Eldorado, Lassen, Sierra and Stanislaus National Forests, California 

inside treatment areas  outside treatment areas 
herbicide/method
 #positive  #samples  %positive  #positive  #samples  %positive 
glyphosate/ground  16  31  52  2  44  5 
hexazinone1  20  43  47  1  46  2 
Pronone®/air  16  17  94  0  17  0 
Pronone®/ground  1  10  10  0  19  0 
Velpar®/ground  3  16  19  1  10  10 
triclopyr/ground  9  18  50  1  29  3 
TOTALS  45  92  49  4  119  3 


1 Pronone® = granular formulation, Velpar® = liquid formulation


Table 4. Summary of sampling results by plant for samples collected inside or outside areas treated with glyphosate, hexazinone or triclopyr in the Eldorado, Lassen, Sierra, or Stanislaus National Forests, California. 

inside treatment areas  outside treatment areas 
plant
 #positive  #samples  %positive  #positive  #samples  %positive 
bitter cherry shoot  1  4  25  0  2  0 
bracken fern root  6  17  35  0  18  0 
buckbrush shoot  5  7  71  0  6  0 
deer brush shoot  5  12  42  0  19  0 
deergrass stalk  1  2  50  0  6  0 
dogwood shoot  1  1  100  0  2  0 
elderberry fruit  0  2  0  0  1  0 
golden fleece foliage  15  20  75  1  14  7 
manzanita berry  6  12  50  1  8  13 
pearly everlasting foliage  2  8  25  0  6  0 
soaproot bulb  3  7  43  1  15  7 
watercress foliage  0  0  --  0  4  0 
willow shoot  0  0  --  1  18  6 
TOTALS  45  92  49  4  119  3 




Table 5. Summary of sampling results by location for samples collected inside or outside areas treated with glyphosate, hexazinone or triclopyr in the Eldorado, Lassen, Sierra, or Stanislaus National Forests, California. 

inside treatment areas  outside treatment areas 
forest
stand
 #positive  #samples  %positive   #positive  #samples  %positive 
Eldorado  501-044  3  8  38  0  6  0 
Lassen  Domingo 17  1  1  100  0  0  -- 
Domingo 18  0  3  0  0  4  0 
Mineral 01  0  3  0  0  2  0 
Mineral 04  0  2  0  0  0  -- 
Mineral 30  0  0  --  0  2  0 
Sierra  339-005  2  10  20  1  5  20 
339-007  2  2  100  0  4  0 
403-010  1  6  17  0  5  0 
403-011  5  7  71  0  0  -- 
Stanislaus  027-019  11  16  69  3  24  13 
027-038  0  1  0  0  11  0 
038-010  0  0  --  0  2  0 
040-138  2  12  17  0  15  0 
Matsen 38/42  16  17  94  0  15  0 
Murphy 23  0  0  --  0  2  0 
Murphy 38  2  4  50  0   22  0 
TOTALS  45  92  49  4  119  3 




Table 6a. Summary of sampling results over time. Samples were from plants collected inside areas treated with herbicides in Eldorado, Lassen, Sierra, and Stanislaus National Forests.

#positive/#samples 

weeks after herbicide application 
herbicide
 1-8 wk   9-16 wk  17-24 wk  25-32 wk  33-40 wk  41-48 wk  
glyphosate  7/10  7/14  0/5  2/2  not sampled  not sampled 
hexazinone  11/14  5/8  3/11  0/1  1/3  0/6 
triclopyr  5/7  3/10  1/1  not sampled  not sampled  not sampled 
TOTALS  23/31  15/32  4/17  2/3  1/3  0/6 




Table 6b. 
%positives 

weeks after herbicide application 
herbicide
 1-8 wk   9-16 wk  17-24 wk  25-32 wk  33-40 wk   41-48 wk 
glyphosate  70  50  0  100  not sampled  not sampled 
hexazinone  79  63  27  0  33  0 
triclopyr  71  30  100  not sampled  not sampled  not sampled 
TOTALS  74  47  24  67  33  0 



Figure 1. National forests in California with herbicide programs 



Figure 2. Sampling sites in the Eldorado National Forest 



Figure 3. Sampling sites in the Lassen National Forest 



Figure 4. Sampling sites in the Sierra National Forest



Figure 5. Sampling sites in the Stanislaus National Forest



APPENDIX 1

DISCUSSIONS WITH NATIVE AMERICANS
ABOUT PLANTS OF INTEREST

Eldorado National Forest Field Trip Report 

(Trip Date: 11/2/95)

Attendees:

USFS Denise McLemore (chair), Dave Bakke, Robin Barron, Krita Deal, Judy Rood,

Mike Taylor, Jenia Walter, Carmen Galindo

Native Americans Louise Griffin, Fred Velasquez, Jeff Brownell, Fred Short,

Darlene Brown Toyeko, Alec John A-Hiokatoo.

CIBA Steve Nicola

CDPR Kean S. Goh, Randy Segawa, Carissa Gana

We met at the Pacific Ranger Station Office at 9:00 AM on Novemeber 2, 1995. The objectives of the meeting and field trip were to 1) inform the local tribal people of DPR's upcoming herbicide residues monitoring project; 2) have the Native Americans survey the suitability of the set aside areas that are protected and non-herbicide treated, for plant gathering purposes (USFS's objective); 3) have the Native Americans (consultants) identify plant and plant parts of important that we might include in the analytical methods development and the residue monitoring.

Windmiller Site 

This is a set-aside area at 5,000 ft elevation with a small brook running through the site. This is adjacent to herbicides treated areas, and the flora are representative of this treatment zone. Key species and plant parts of interest as pointed out by Native American consultants (notably Fred Velasquez, Jeff Brownell, and Louise Griffin) and botanical names provided by USFS's botanists (notably Robin Barron and Mike Taylor) are as follows. The list are in the sequence of encounter. (* are dorminant species)

*Pearly Everlasting, Anaphalis margaritacea (L.) leaves and flowers, medicine
Sierra Gooseberry, Ribes roezlii (Regel) edible fruits
*Bitter Cherry, Prunus emarginata shoots for basketry
*Deer Bruch, Ceanothus integerrimus stems for basketry
Watercress, Rorippa nasturtium-aquaticum edible foliage
Soaproot, Chlorogalum pomeridianum bulb as soap and bulb fiberous-coat for making brushes
*Brodiaea spp. edible bulbs
*Greenleaf Manzanita, Arctostaphylos patula edible fruits, raw or cider
*Mewuk Manzanita, Arctostaphylos Mewukka edible fruits, raw or cider
Mountain Pennyroyal, Monardella odoratissima stems and flower heads medicinal herb
Pratt's Buckwheat, Eriogonum prattenanium, seeds as food
Wild rose, Rosa spithamea, edible rosehips (fruits)
Elderberry, Sambucus maxicana fruits for drink/jelly
Wild onions, Allium spp. edible bulbs
Tarweed, Hemizonia virgata, plant as bath decoction
Farewell-to-spring, Clarkia spp., seed flour-meal.


Sayles Canyon 

This is at 7,000ft elevation with steep terrain and lush vegetation. The site bordered a flowing creek, and is representative of flora of treatment area at this elevation. Species identified are:

Serviceberry, Amelanchier pallida, food berries and basketry shoots.
Thimbleberry, Rubus parviflorus, edible berries
Mugwort, Artemisia douglasiana, medicinal herbs


References

Native American and Pioneer Uses of Plants in the Wright Lake Area compiled by Jenia Walter, Karen Bennett, and Krista Deal. 1993.

Native Plants of the Lava Cap and Traditional Native American Uses. 4pp.

Native American Uses of Plants in the Cleveland Fire Area 6pp.

K.S. Goh 11/6/95


Stanislaus National Forest: Mi-Wok Ranger District Trip Report 

Trip date: November 13, 1995

PART I: Consultation with local Native Americans 

Attendees:

USFS: Phyllis Ashmead (botanist), Jayne Montoya (Tribal Government Relations Manager), Mike Rutty (Stanislaus NF), Joe Sherlock (Mi-Wok RD), John Schmechel (Groveland RD), and Jim Mullin (Summit RD).

Native American Consultants: Jennifer Bates, Reba Fuller, and David Lingo

CDPR: Adrian Bradley, Kean S. Goh, Carissa Gana, and Randy Segawa

We met at the Mi-Wok Ranger Station at 9 a.m. on November 13, 1995 for the discussion, and travel to visit field sites in the afternoon. The objectives of the meeting and field trip were to 1) brief the local tribal people and USFS staff of DPR's upcoming herbicide residues monitoring, 2) solicit input from local tribal consultants on important plants and plant parts that we should look for and include in DPR's monitoring, 3) select potential sites for residue monitoring for next spring and fall, 4) identify plant species to scientific names, 5) gather new plant materials for developing analytical methods for triclopyr, hexazinone, and glyphosate.

The following is a list of major local plants used by local Native Americans as provided by the local consultants. Scientific names are from references and as collaborated by the botanist.

Acer macrophyllum (Big-Leaf Maple, basketry sticks)
Achillea millefolium (Yarrow, medicinal herb)
Allium spp. (Wild onion, food leaves)
Angelica breweri (Angelica, medicinal roots)
Arctostaphylos spp (Manzanita, food fruits)
Artemisia douglasiana (sagebrush, mugwort, medicinal/ceremonial herb)
Asclepias spp (milkweed, cordage)
Brodiaea spp (food tubers)
Carex sp. (White root, basketry rhizomes)
Ceanothus spp (Deer brush, buck brush, basketry shoots)
Chlorogalum pomeridianum ( Soaproot, fibrous coverings for brushes, bulb for basket coating)
Cornus spp. (Dogwood, basketry shoots and sticks)
Corylus cornuta (Hazel, basketry shoots)
Fragaria spp. (wild strawberries, fruits)
Muhlenbergia rigens (Deergrass, basketry flowering stalks)
Prunus spp (Bittercherry, chokecherry, basketry shoots)
Pteridium aquilinum var. pubescens (Braken fern, basketry rhizomes) 
Quercus kelloggii (Black oak, food acorns)
Rhus trilobata (Sourberry, basketry shoots)
Ribes spp (Sierra currant, gooseberry, food berries)
Salix spp (Willows, basketry shoots)
Sambucus spp (Elderberry, food berries)
Mushrooms (morel, buckskin, puffball, etc.)



PART II: Field visit 

USFS: Phyllis Ashmead, Jayne Montoya, and Joe Sherlock

CDPR: Adrian Bradley, Kean S. Goh, Carissa Gana, and Randy Segawa

Native American consultants were not available for the field visit.

Matsen Compartment 30, stand 52 

This is a part of various planted, mixed conifer stands with total acreage of 533 acres above

3,000 ft elevation and slope of 10-50%. Some of these stands are previously treated once with glyphosate using backpack sprayers and are scheduled for aerial hexazinone granule treatment in spring 1996. This areas contained a riparian area. A 100 ft buffer will be maintained surrounding the creek. Major species of important plants to Native Americans found at this site in decreasing order of prevalence are:

Golden fleece, Ericameria arborescence,(collected leafy stems for residue methods development)
Buck brush, Ceanothus cuneatus, (collected young shoots)
Mariposa manzanita, Arctostaphylos viscida mariposa (no berries available)
Black oak, Quercus kelloggii (no acorns available)
Sierra Gooseberry, Ribes roezlii (no berries available)
Major species found in the buffer zone, creek area with a small flowing spring:
Buck brush, Ceanothus cuneatus 
Willow, Salix sp. 
Mugwort, Artemisia douglasiana (insufficient quantity for method development)
Watercress, Rorippa nasturtium-aquaticum (insufficient quantity for method development)

References: 

Local Plants Used by Native Americans, Stanislaus National Forest.

Native American Uses of Plants, four-page handout from Phyllis Ashmead.

Paper Reforestation Project, Final Environmental Impact Statement.

ksg 11/27/95 miwok



Sierra National Forest Trip Report 

Trip Date: November 20 and 21, 1995

Part I. Tribal Relations Program Briefing 

Attendees: over 20 people attended including members of the mono tribes, district foresters, CalTran staff, and other state agencies staff.

Native Americans: Thane Baty (Big Sandy Tribal Chairman), Norman Turner, Ester Ortiz, Melba Beecher, Gladys McKinney (CIBA), et al.

USFS: Jim Boynton (Forest Supervisor), Dirk Charley (Tribal Relations Program Manager), Jerry Bishop (American Indian Special Emphasis Program Manager), and Mark Smith (Silviculturist), et al.

DPR: Adrian Bradley, Carissa Gana, Kean S. Goh, and Randy Segawa

Objectives:

We met at the Big Sandy Rancheria, Auberry, on November 20, 1995, at 1:00 p.m. As part of the meeting, we presented DPR's forestry herbicide residues monitoring project to the tribal members and USFS staff and also solicit local inputs and address issues associated with the monitoring.


Part II. Field visit 

Attendees:

USFS: Mark Smith (silviculturist), Joanna Clines (botanist)

CalTran: Dana York (botanist)

Native American Consultants (coordinated by Dirk Charley): Melba Beecher, Edith Beecher, Martha Beecher, Marie Cohoe, Gladys McKinney, Ester Ortiz, Norma Turner, and Cecilia Wesly

DPR: Adrian Bradley, Kean S. Goh, and Randy Segawa

Objectives:

We met briefly at the Big Sandy Rancheria, Auberry, on November 21, 1995, 9:00 a.m. and proceeded to visit field sites targeted for herbicide treatment next year. The objectives of the meeting and field trip were to 1) consult with local Native Americans on important plant species and plant parts used for basketry, foods, medicines, etc.; 2) select sites with important plant species in consultations with USFS and Native Americans for herbicide residues monitoring; 3) identify plant species to scientific names; and 4) gather plant materials for analytical methods development.

Important Plant Species 

The following is a list of plants and uses important to the local tribes as provided by the Native American consultants, notably Norma Turner (Mono) with assistance from Joanna Clines with botanical names:

Black oak, valley oak (acorn food)
Bracken fern (basketry roots)
Cascara, coffeeberry (bark medicine)
Chaparral, buckbrush (basketry shoots)
Clover (food, salt)
Elderberry (berry fruit, blossom and leave medicine)
Golden rod (root medicine)
Laurel, bay (leaves and sticks)
Manzanita (leave medicine, berry fruit, wood pipestem)
Milkweed (stem cordage, sap medicine)
Mountain mahogany (digging sticks)
Mushrooms (food)
Redbud (basketry shoots)
Slippery elm, flannelbush (shoot cordage)
Soaproot (fibrous covering brushes, bulb shampoo)
Sourberry (fruits and basketry sticks)
Sugar pine (pitch gum)
White root, Carex spp (basketry root)
Wild grape (cordage, fruit)
Wild berries-gooseberry, serviceberry, blackberry, cherry, plum (fruits)
Willows (basketry shoots)
Wormwood (medicinal and ceremonial herb)
Yerba santa (medicinal herb)

Field Sites Visited 

We visited four major sites 339-5, 336-136, 9S07B, and 336-219. The following two are potential monitoring sites. Sites information and maps are provided by Mark Smith; plant species identifications are by Native American consultants and Joanne Clines.

339-5: This site was treated on July 94 with glyphosate and will be retreated with hexazinone by hand in spring 1996. This 12-acre site is at 3,000 ft elevation with 30% slope and planted with mixed conifers. Important plant species of interest to Native Americans are sparce they includes

(in decreasing order of abundance) soaproot, black oak, deerbrush, willow (Salix lasiolepis), and golden fleece.

9S07B (Deergrass patch): This is a gently slopping site about 10 acres at 3,600 ft elevation. It has an ephemeral drain (currently dry) at the bottom of the site. This site may be treated with glyphosate or hexazinone with appropriate buffer bordering the drain. Important plant species of interest to Native Americans are buckbrush, deergrass, whiteleaf manzanita, and wild rose.

References: 

Cultural Resources of the Native American on or near the Sierra National Forest compiled by Ron W. Goode, edited by Joanna Clines.