ENVIRONMENTAL PROTECTION AGENCY OFFICE OF ENFORCEMENT EPA -330/2-77- 014 HURBICIDE USU SUPERIOR AND CHIPPEWA MINNESOTA (JULY NATIONAL ENFORCEMENT INVESTIGATIONS CENTER DENVER, COLORADO and REGION V - CHICAGO rtO S7*. OCTOBER 1977 ------- ENVIRONMENTAL PROTECTION AGENCY OFFICE OF ENFORCEMENT EPA-330/2-77-014 HERBICIDE USE OBSERVATIONS SUPERIOR AND CHIPPEWA NATIONAL FORESTS MINNESOTA (July 1977) October 1977 National Enforcement Investigation Center - Denver and Region V - Chicago ------- CONTENTS I INTRODUCTION 1 II SUMMARY AND CONCLUSIONS 3 General 3 Specific 3 III DESCRIPTION OF STUDY AREA 5 IV USE OBSERVATIONS 6 Pre-application 6 Application 8 Post-application 9 V SAMPLING DEVICES AND METHODS 13 TABLES 1 Results of Spray Drift and Environmental Sampling. Superior National Forest, Minnesota 2 VMD> Deposition Density and Drift Distances. SuDerior National Forest, Minnesota 3 Results of Spray Drift and Environmental Sampling. Chippewa National Forest, Minnesota 4 VMD> Deposition Density and Drift Distances, Chippewa National Forest FIGURES 1 Sampling Stations, Superior National Forest 2 Sampling Stations, Chippewa National Forest ------- HERBICIDE USE OBSERVATIONS SUPERIOR AND CHIPPEWA NATIONAL FORESTS, MINNESOTA JULY 1977 I. INTRODUCTION Silviculture includes the practice of controlling forest establishment, composition and growth. It is a continuous management process that begins when mature timber is harvested and the site is prepared for a new crop of trees. Depending on many factors, a forester often uses pesticides to prepare an area for reforestation. Herbicides are widely used to control undesirable plant species and weeds in reforestation areas. To delineate the actual and potential environmental damage caused by the aerial application of 2,4-D on forest areas, the EPA National Enforcement Investigations Center (NEIC) and Region V conducted an herbicide use observation study. This study was conducted in the Superior and Chippewa National Forests in Northern Minnesota, from July 12 through 15, 1977. The objectives of the study were as follows: 1. Determine that applications of 2,4-D were performed consistent with Federal Insecticide, Fungicide and Rodenticide Act (FIFRA) regulations and appropriate state codes. 2. Determine that applications of 2,4-D were performed consistent with U. S. Forest Service Environmental Analysis Reports (EAR) and Environmental Impact Statement (EIS) recommendations. 3. Determine the extent and environmental effects of herbicide movement from target areas onto non-target areas. The study determined residue levels of 2,4-D in ambient air, vegetation, soil and water. Various types of spray droplet cards and high-volume air ------- 2 sampling devices were used to characterize spray drift. While storage and mixing of the material were not observed, on-site handling and loading operations were. ------- II. SUMMARY AND CONCLUSIONS 3 General In July 1977, the NEIC and Region V conducted a four-day study in two Northern Minnesota national forests to monitor the aerial application of 2,4-D used to control undesirable plant species. On-site observations, combined with the use of various monitoring techniques, enabled the EPA to determine user compliance with label instructions and to evaluate both beneficial and harmful effects associated with the treatment program. Technologically it is impossible to aerially spray with "no drift - no contamination", therefore, the forest sprayings observed by EPA were evaluated to determine lasting or other major adverse impacts on the envi ronment. On-site observations revealed that the applicator used a registered chemical herbicide consistent with Federal regulations. Environmental sampling showed the use of 2,4-D resulted in contamination of non-target forest land (340 yg/gm) and waterways (1.4 yg/1). The most valuable techniques used to document the practices and environmental effects of the aerial applications were on-site evaluations by trained observers and analyses of spray-drift cards, air, water and soil to determine drift and 2,4-D residues at the study sites. Specific 1. Pre-application sampling revealed no 2,4-D contamination of either of the two national forest study sites. 2. Possible human health hazards were observed during on-site loading operations. Workers handling the herbicides wore no safety apparel. ------- 4 At a minimum, those individuals should have been wearing long-sleeved shirts and rubber gloves. 3. In the Superior and Chippewa National Forests no visible drift was ob- served beyond the spray sites. However, factors observed that con- tributed to overspray and contamination of the surrounding area were: (a) difficulty in delineating the physical boundaries of the area be- cause insufficient numbers of balloons were used to mark the site, and (b) the aircraft pilot flew along the short axis of the area; if he had flown along the long axis of the site he would have used existing bal- loons more effectively. Drift contamination of the surrounding areas was kept to a minimum: maximum drift distance in the Superior Forest was about 6 meters, and in the Chippewa Forest about 100 meters. 4. Observations and analyses by NEIC and Region V personnel after che forests were treated revealed that an application rate of eight gallons per acre of 2,4-D provided extensive coverage of plant surfaces, thus insuring good plant kill. Chemical data from the Superior Forest showed that less than 1% of the herbicide penetrated the forest canopy and reached the ground. 5. Laboratory analyses revealed a spray-droplet Volume Median Diameter (VI1D) exceeding 1,000 microns for both forests. Such large-sized droplets considerably reduce the propensity for drift, as well as reducing drift to an average distance of less than 3.5 meters. 6. On-site observations and residue analyses for 2,4-D in air, water and soil and on high volume air sampler filters and Kromecote cards, coupled with spray-drift evaluations were valuable techniques for monitoring the use of 2,4-D. ------- III. DESCRIPTION OF STUDY AREA 5 Superior and Chippewa National Forests, located in the Arrowhead region of Northern Minnesota, are two primary multiple-use forests in the Midwest. Timber production and recreation are among the major uses of these forests. Two sites were selected for the observation study: (1) Superior National Forest, Gunflint District, about 25 miles north (T63N, R1E, Sections 7 and 18 and T63N, R1W, Sections 12 and 13) of Grand Marais, Minnesota. The 81-acre spray area is characterized by heavy timber and brush growth, with the spray site being dominated by juvenile aspen and brush. Private land borders the north edge of the spray area. (2) Chippewa National Forest, Marcell District, about 40 miles north (T149N, R25W, Sections 7, 8, 17, 18) of Grand Rapids, Minnesota. The 36-acre spray site is a well defined stand of juvenile aspen surrounded by a mixture of mature trees. Heliports close to the spray sites were used to facilitate the loading of the spray material (the site in the Chippewa National Forest was part of a spray area). Sixteen sampling stations were established in the Superior National Forest (Figure 1) and 29 sampling stations were located in the Chippewa National Forest (Figure 2). These sampling sites were situated around the spray area in such a manner that any spray drift or other environmental effect could be evaluated. Stations were located both on and off the spray sites and near adjacent v/aterways. Air, vegetation, soil and water were collected at selected times and locations within this sampling network. Samples were analyzed to determine the fate of 2,4-D sprayed on the areas to control undesirable plant species. *Esteron 99 Concentrate. Dow Chemical Co., EPA registration No. 464-201-AA. ------- PRIVATE LAND POND NOT TO SCALE Figure 1. Sampling Stations, Superior National Forest, Minnesota Spray Site G-5, July 1977 ------- Figure 2. Sampling Stations, Spray Site \ \ NOT TO SCALE Chippewa National Forest, Minnesota M-8, July 1977 ------- IV. USE OBSERVATIONS 6 PRE-APPLICATION On July 13, 1977, prior to the aerial application of the herbicide 2,4-D, environmental samples were collected from the Superior National Forest study site to establish background levels. Soil and water samples from the spray site contained no detectable 2,4-D residues (Table 1). The same type of samples were collected from the Chippewa National Forest on July 15, 1977,- a few hours prior to the aerial treatment of that site. No detectable herbicide residues were found (Table 3). Prior to the aerial application in both forests, sampling devices to monitor spray drift were placed around the study sites. These devices included: high-volume air samplers and spray-droplet cards (Kromecote cards, Thermofax paper and Linograph paper). Each device is described in the Methods section of this report. Meteorological data including wind direction, velocity, air temperature and humidity were collected. These factors were used to evaluate the influence of atmospheric conditions on the movement of 2,4-D beyond the target sites. At neither forest site was the EPA able to observe the mixing operations. This operation was conducted at points removed from the spray areas, with the mix being transported to the spray sites by tank trucks. Loading operations were performed on site. The vehicle containing the mix (a 1035 liter tank mounted on a flat-bed truck) was moved onto the heliport where closed seal couplings were used to attach a heavy gauge ------- 7 flexible hose between the mix tank and the helicopter spray tank. Approxi- mately 360 liters of the mix were pumped through the flexible hose into the aircraft spray tank. A tight seal was maintained throughout the loading operation and no visible leakage occurred. When the hose was disconnected a plastic bucket was used to capture the small (approximately 50 ml) volume of mix that remained in the line. This mix was manually added to the spray tank. In cooperation with the Forest Service, a fluorescent dye (Rhodamine WT) was added to each helicopter load of mix at a concentration of about 2,000 yg/l. The dye was used: (1) to enhance visible observations of drift, and (2) to enhance the spots on spray-drop cards. Prior to loading the helicopter in the Chippewa National Forest, an EPA Pesticides Inspector collected a sample of the mixed material. Later chemical analysis showed that both the mix and the formulated material conformed to label specifications (5.24% 2,4-D acid PGBEE1). A similar sample collected in the Superior National Forest showed extremely low (0.8% acid PGBEE) 2,4-D levels in the mixed material. Subsequent investi- gation showed that this sample was collected from the tank truck after it had sat overnight and the 2,4-D and water had separated. Chemical analysis of the sprayed material confirmed that the material was properly mixed when sprayed. During the loading operation personnel of both the applicator and the Forest Service wore no safety clothing. Although no Federal requirements exist for safety clothing to be worn when working with 2,4-D, data are available to indicate exposure to 2,4-D may be hazardous.2 1 Propylene glycolbutyl ether ester 2 Morgan, D.P. 1976. Recognition and Management of Pesticide Poisonings. U.S. EPA Report Mo. EPA-540/9-011. Washington, D. C. 56 pp. ------- 8 The EPA observation team wore safety clothing consisting of an approved canister-type respirator, face shields, hats, long-sleeved coveralls and rubber gloves. APPLICATION On July 13, 1977, between 8:30 a.m. and 10:30 a.m., 2,4-D was applied to the Superior National Forest study site. Warm and sunny weather prevailed during the application. Air temperature was about 65°F, humidity was 65% and the wind from the northwest at 1.6 Kms/hr, increasing to 8 Kms/hr. The spray aircraft, a Bell Jet Ranger 206-B helicopter, was equipped with a Micro-Foil boom using .03 cm nozzles set in rows of three on the boom segments, thus providing 180 nozzles per six-inch segment. Rate of appli- cation of the 2,4-D mix was nominally 7.5 liters/hectare with an active ingredient of 0.56 Kg/hectare. While the EPA team did not observe visible drift beyond the spray site, several conditions were observed that contributed to contamination of the surrounding area: (1) the spray site was not a distinct entity. Some difficulty was encountered in delineating the physical boundaries because balloon markers used to mark the site were insufficient in numbers to allow the helicopter pilot to delineate the site; thus areas outside the zone, including a portion of privately owned land, were 2,4-D treated and, (2) it appeared that, with the configuration of the spray site, the pilot should have changed his spray swaths from a north-south direction to an east-west direction. This change would have put the spray swaths on a more direct line with the marking balloons, along the long axis of the spray zone. ------- 9 On July 15, 1977, the 2,4-D treatment of the Chippewa National Forest site was conducted between 7:30 p.m. to 9:00 p.m. Conditions were similar to the previous application, with warm, sunny weather, and light breezes (<8 Kms/hr). The type helicopter and spray rig used for this 2,4-D application was identical to that used in the Superior National Forest. The EPA observation team reported no unusual conditions or events during the course of the application, but noted that the pilot was conserv- ative in that he stopped spraying before reaching the boundary of the target forest area. This site also was treated at 7.5 liters/hectare with an active ingredient of 0.56 Kg/hectare. POST-APPLICATION As noted earlier, the EPA did not observe the mixing operation. Sub- sequent conversations with U. S. Forest Service personnel revealed that all containers v/ere triple-rinsed, with the rinse water being added to the spray tanks. Empty drums were transported to a Class I dump (a site especially designed for the disposal of highly toxic or environmentally damaging compounds or containers). The used containers were crushed and buried in compliance with all pertinent FIFRA regulations. Superior National Forest Following the completion of the 2,4-D application, the EPA observation team entered the study site (treated forest and surrounding area) to retrieve spray-droplet cards and the high-volume air samplers. Water, soil and vegetation samples were also collected. Results of drift deposit analyses for residues on spray-cards (Kromecote, Thermofax and Linograph), high-volume air sampler filters, water soil and vegetation are presented in Table 1. ------- TABLE 1. RESULTS OF SPRAY DRIFT AND ENVIRONMENTAL SAMPLING SUPERIOR NATIONAL FOREST, MINNESOTA July 1977 Station Type 13 Water 13 Water 13 Soil 13 Soil 11 Soil Vegetation 5 Hi-vol filter 15 Hi-vol filter 1* Kromecote card 2* Kromecote card 3* Kromecote card 4* Kromecote card 5* Kromecote card 6 Kromecote card 7 Kromecote card 8 Kromecote card 9 Kromecote card 10* Kromecote card 11 Kromecote card 12 Kromecote card 15* Kromecote card 16* Kromecote card 17* Kromecote card Time Concentration Pre** N.D.*** * Post*** lU'yg/l Pre N.D. Post 0.02 yg/gm Post 1.7 yg/g Post 340 ^g/g Post 1,300 ug Post 21 yg Post 1.4 yg Post 160 ug Post N.D. Post 22 yg Post 2.0 vjg Post 110 yg Post 380 yg Post N.D. Post 66 yg Post 140 yg Post 150 yg Post 390 yg Post 0.2 yg Post 26 yg Post 67 yg * Off-site locations ** Pre-application *** Post-application **** None detected ------- TABLE 2. VMD, DEPOSITION DENSITY AMD DRIFT DISTANCE SUPERIOR NATIONAL FOREST, MINNESOTA JULY 1977 Calculated Deposition Density Drift Distance Station VMD (y) (Drops/cm2) (ft)*** 1* 1,415 11.3 2 2* 1,038 8.5 11 3* N.A.** N.A. N.A. 4* 648 4.8 19.5 5* N.A. N.A. N.A. 6 748 4.0 16 7 1,288 14.7 9 8 N.A. N.A. N.A. 9 1,123 8.4 10 10* 983 6.0 11 11 998 7.6 11 12 1,349 9.4 3.5 13 N.A. N.A. N.A. 14* N.A. N.A. N.A. 15* N.A. N.A. N.A. 16* 692 0.79 18 * Off-site locations ** Mot applicable *** Based on 5 mph wind and 20 ft fall ------- Surface water collected from a small temporary pond at Station 13 prior to application contained no detectable amounts of 2,4-D, while a post-application level of 1.4 yg/1 was found. At this same station, soil residue levels ranged from non-detectable (pre-application) to 0.02 yg/gm (post-application). Elsewhere on the spray site soil levels of 1.7 pg/g were recorded (Station 11), with composite samples of vegetation showing 340 pg/g of 2,4-D. There was a difference in residue levels between the soil and vegetation samples. The vegetative cover was dense enough to capture virtually all the 2,4-D, thus insuring good control of undesirable plant species and helping to avoid ground and possibly groundwater contamination. Kromecote card data show varying amounts of residues at the site (Table 1) with, as would be expected, the center portions of the spray area having the higher amounts. Data from the high-volume air samplers and Kromecote cards show some areas of concern. At Stations 2, 5, 15 and 16 on the north side of the area and Station 10 on the south side, all off the target zone, 2,4-D was found (Table 1). These residue levels (140 to 160 ug) indicate direct overspray rather than drift, and are sufficient to cause plant damage. It is apparent that insufficient numbers of balloon markers were used to outline the target area, thus allowing the overspray to occur. The VMD* of the 2,4-D spray ranged from 648 microns to 1,415 microns, with an average of 1,028 microns (Table 2). The maximum "calculated drift *VHD - Volume median diameter: that value which divides the droplet diameter into two equal parts, one-half above and one-half below the median or 50% calculative point. ------- 11 distance" was about 6 meters, therefore, drift contamination of surrounding areas was kept to a minimum. Chippewa National Forest As with the Superior National Forest, members of the EPA team retrieved sampling devices and collected post-application environmental samples following completion of spraying activities. Table 3 summarizes the results of environmental and spray-card sampling. No detectable amounts of 2,4-D were found in water samples from Station 25, a pond, located about 60 meters southwest of the spray area. No detectable 2,4-D residues were found in the soil at Station 1, but at Stations 5 and 10 post-application levels of 0.49 yg/g and 3.6 yg/gm, respectively, were recorded. Filters from the high-volume air samplers had from 18 yg to 2,100 ug (with the latter value coming from a unit that was sprayed directly). The Kromecote cards showed the variation normally expected, ranging from detection limits (0.2 yg) to 110 yg. Minor amounts of spray drift were recorded at stations bordering the spray site. For unexplained reasons, the areas around Stations 13A and 15 (both on the target site) received no 2,4-D treatment. It will be noted that VMD falls into two basic categories, a VMD of <500 microns and >1,250 microns (Table 4). The smaller droplets occurred around the periphery of the treated area, with the larger droplets being found on the target site. The data indicate the spray zone borders were affected by both overspray and minor drift, some of which possessed the potential to contaminate the immediately surrounding area (i.e., 57 micron drops at Station 23 could drift 102 meters). Existing data do not allow a further interpretation of this point to be made. ------- TABLE 3. RESULTS OF SPRAY DRIFT AND ENVIRONMENTAL SAMPLING CHIPPEWA NATIONAL FOREST, MINNESOTA JULY 1977 Station Type Time Concentration 25* Water Pre* * N.D.**** 25* Water Post*** N.D. 1* Soil Pre N.D. 1* Soil Post N.D. 5 Soil Pre N.D. 5 Soi 1 Post 0.49 yg/g 10 Soil Pre N.D. 10 Soil Post 3.6 yg/g 1* Hi-vol filter Post 18 yg 9* Hi-vol filter Post 2,100 yg 21* Hi-vol filter Post 56 yg 16 Hi-vol filter Post 150 yg 1* Kromecote Post N.D. 2* Kromecote Post N.D. 3 Kromecote Post 25 yg 4* Kromecote Post 0.2 yg 5 Kromecote Post 62 yg 6* Kromecote Post 0.2 yg 7 Kromecote Post 19 yg 8* Kromecote Post N.D. 9* Kromecote Post 49 yg 10 Kromecote Post 38 yg 12* Kromecote Post N.D. 13* Kromecote Post 0.3 yg 13A Kromecote Post N.D. 14 Kromecote Post 110 yg 14A* Kromecote Post 0.6 yg 15 Kromecote Post N.D. 16 Kromecote Post 0.7 yg 17 Kromecote Post 39 yg 17A Kromecote Post 61 yg 18* Kromecote Post 1.4 yg 19* Kromecote Post N.D. 20* Kromecote Post N.D. 21 * Kromecote Post 0.5 yg 23* Kromecote Post 0.2 yg 24 Kromecote Post 2.8 yg 25* Kromecote Post N.D. 26 Kromecote Post 12 yg 27 * Kromecote Post 0.2 yg 28 Kromecote Post 33 yg 29 Kromecote Post 70 yg * Off-site locations ** Pre-application *** Post-application **** None detected ------- TABLE 4. VMD, DEPOSITION DENSITY AND DRIFT DISTANCES CHIPPEWA NATIONAL FOREST, MINNESOTA JULY 1977 Calculated Deposition Density Drift Distance Station VM.D (u) (Drops/cm2) (ft)*** 1* N.A.** N.A. N.A. 2 N.A. N.A. N.A. 3 1,379 5.2 7.5 4 286 0.38 98 5 1,443 6.5 8 6* 243 0.28 100 7 1,264 4.3 7 8* N.A. N.A. N.A. 9* 1,386 4.6 6.5 10 1,996 3.3 4.5 12* N.A. N.A. N.A. 13* N.A. N.A. N.A. 13A N.A. N.A. N.A. 14 2,001 9.5 4.5 14A* 205 0.38 69 15 N.A. N.A. N.A. 16 2,294 0.38 4 17 1,420 5.9 7.8 17A 2,035 4.7 4.3 18* 2,592 0.45 2.3 19* N.A. N.A. N.A. 20* N.A. N.A. N.A. 21* . 551 0.07 2.5 23* 57 0.02 335 24 N.A. N.A. N.A. 25* N.A. N.A. N.A. 26 1,417 2.3 7.8 27* N.A. N.A. N.A. 28 1,445 5.4 8 29 1,952 5.7 4.2 * Off-site locations ** Not applicable *** Based on 5 mph wind and 20 ft fall ------- 12 To summarize, on-site observations revealed that both National Forests were aerially treated with 2,4-D in a manner consistent with Federal regulations. Sampling showed that non-target forest land and waterways were contaminated by overspray or minor drift. The most valuable techniques used to document the practices and environmental effects of the aerial application were on-site evaluations by trained observers, VMD analyses of spray drift cards, and chemical analyses of soil and vegetation to determine drift and 2,4-D residues at the study site. ------- 13 V. SAMPLING DEVICES AND METHODS High-Volume Air Samplers This system uses a fiberglass filter 20 x 25.4 cm, on which airborne particles and spray are trapped. Approximately 1 m^/min (35 ft^/min) of air was drawn through the filter. The system was operated for three hours at each station. A total of 105 m3 of air passed through each filter. The filters were removed, stored in plastic bags and returned to the laboratory. The filters were extracted by submersion in a Pyrex dish containing 250 ml of water adjusted with sulfuric acid to a pH of 1.5. The water extract was further treated by extracting the 2,4-D, in acid form, with an organic solvent. The 2,4-D acid was esterfied with borontrifluoride in methanol and the resulting methyl ester analyzed by electron capture gas chromatography. Spray Droplet Cards Spray card clusters were constructed by stapling Linograph 480 paper, Thermofax 209 copy, type 640 paper and Kromecote cards onto 12-inch square poster board bases. The clusters were placed around and on the spray sites atop one-meter upright wooden platforms and attached by elastic bands. All cards were highly sensitive to the 2,4-D material. Visual images were enhanced by the addition of the fluorescent dye (Rhodamine WT). Linograph 480 and Thermofax 209 copy type 640 cards were sent to a contracting agency for analysis of the spray drop impingement. The analytical technique used a Quantimet Image Analyzing Computer, provided data on VMD ------- and deposition density. However, this optical scanner cannot distinguish between spray drops and imperfections in the spray card paper, dust, dirt, etc., thus some cards had unreasonably high values reported. Realizing these limitations and comparing the values with chemically analyzed cards, some values were set aside and labeled "Not Applicable" in Tables 2 and 4. Kromecote cards were chemically analyzed for 2,4-D residues. Treatment of these cards was the same as for the high-volume air sampler filters. ------- |