Agricultural and Food Science, Vol. 19 (2010): 13-23 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 13 © Agricultural and Food Science Manuscript received: February 2008 Effects of birch tar oils on soil organisms and plants Marleena Hagner1, Tiina Pasanen1, Bengt Lindqvist2, Isa Lindqvist2, Kari Tiilikkala2, Olli-Pekka Penttinen1 and Heikki Setälä1* 1Department of Ecological and Environmental Sciences, University of Helsinki, Niemenkatu 73, FI-15140 Lahti, Finland, *e-mail: heikki setala@helsinki.fi 2MTT Agrifood Research Finland, Plant Production Research, Rillitie 1, FI-31600 Jokioinen, Finland The use of birch tar oil (BTO) is a new innovation in plant and animal protection working against various weeds, harmful insects and rodents. Due to its novelty as a biocide/repellent/plant protection product, no comprehensive information on the effects of BTO on non-target soil organisms is available. In this study we examined the impact of BTO on non-target soil organisms (enchytraeids, nematodes and soil microbes) and plants using laboratory toxicity tests and field experiments. In addition, we determined the LC50 value of BTO to the earthworm Aporrectodea caliginosa and the EC50 value of BTO to the offspring production of the collembolan Folsomia candida. The effects of BTO on soil fauna were mostly insignificant. BTO seemed to be detrimental to the growth of plants directly after application, but this effect was short-term; after a period of 2.5 months, the growth of most of the plant species recovered completely from the ap- plication. The LC50 for A. caliginosa was 6560 mg BTO kg -1 dry soil and EC50 for juvenile production of F. candida was 5100 mg BTO kg-1 dry soil. The results indicate that the risk caused by BTOs (concentration 500−1360 l ha-1) to the soil environment is insignificant and short-term as compared to the many chemical products applied for similar purposes. Key-words: birch tar oil, plant protection, bio control, enchytraeids, nematodes, soil microbes, ecotoxicology Introduction The impact of pesticides on soils, their biota, and functions have become a matter of considerable concern especially in agriculture. Various pesticides are known to increase mortality of non-target soil organisms, thereby hampering the decomposition rate of organic matter (Bűnemann et al. 2006) and thereby altering soil physico-chemical quality. Because of the various negative effects due to pesticides, there is an increasing need to develop biological methods for pest control. A G R I C U L T U R A L A N D F O O D S C I E N C E Hagner, M. et al. Birch oil toxity 14 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 15 Birch tar oil (BTO) is a byproduct of process- ing birch wood through a pyrolysis process to pro- duce charcoal. There is ample anecdotal evidence suggesting the suitability of BTO for biocide and herbicide use, and/or repellents against insects, ro- dents, slugs and snails (Hagner et al. unpublished; personal observations by the authors, Hagner 2005, Hagner et al. 2010, Lindqvist et al. 2010, Salonen et al. 2008, Tiilikkala and Salonen 2008). Due to its novelty as a biological plant protection product, biocide, and repellent, and because of the high number of compounds evaporating in the py- rolysis process, no comprehensive information on the effective compounds of birch oil is available. Preliminary analyses indicate phenols to be one of the most promising groub of compounds in BTO acting as biocontrol agents, but various volatile compounds may also play a role. According to international regulations (EU 1996), the ecotoxicological effects of chemicals on the environment are to be assessed before using chemicals in the field. To the best of our knowl- edge, no studies have been documented with regard to the ecotoxicological effects of BTO. In Finland large scale field experiments using biological plant protection products can not be established if the ecological effects of these products are not known. We therefore examined the general eco- logical effects of BTOs on target organisms before conducting the actual experiments. In the current study the effects of BTO on non-target soil organ- isms were tested using two kinds of birch tar oils, both derived from the same pyrolysis process: 1) BTO1, the liquid material originating during the early phase of the distillation process when the temperature remains below 380 oC, and 2) BTO2, the more viscous component generating at the end of the process when temperature rises to 400 oC. BTO1 is effective against many weeds and insects and can be used as a herbicide and insecticide for example in potato fields (personal observation by the authors), while BTO2 has been shown to be an effective molluscicide (Hagner 2005, Linqvist et al. 2010). This study is part of a larger research pro- gramme in which the effect of BTOs as a mollusc repellent (Lindqvist et al. 2010) and their ecotoxi- cological effects on both terrestrial (the current study) and aquatic (Hagner et al. 2010) organisms were investigated. Here we aimed at 1) determining the ecotoxicological effects of BTO’s on non-target soil organisms and plants, and 2) to determine the lethal concentration (LC50) value and the no ob- served effect concentration (NOEC) value for the earthworm Aporrectodea caliginosa, and the ef- fective concentration (EC50) value and the NOEC value for the reproduction of the springtail (col- lembolan), Folsomia candida. Material and methods Ecological effects of BTOs on non-target soil organisms and plants – field investigations The ecotoxicological effects of BTOs on soil or- ganisms were monitored via changes in soil fauna and plant populations. Three different groups of organisms covering various trophic levels were chosen: enchytraeid worms (mostly omnivorous), nematodes (covering several trophic positions), and soil microbes (primary decomposers). Relatively high doses of BTO1 were used in the experiments because we aimed at getting information about the highest possible risks of the substance. Further, an effective control of perennial weeds is likely to require high doses (1.36 dl m-2) of this substance (Lindqvist et al., unpublished data). When used for weed control of annual crops the required dose of BTO is about one-third, and for controlling pest insects about a tenth of the dose applied for peren- nial grass control. In the toxicity tests we mostly applied BTO1 due to its higher viscosity which renders it use more convenient than that of BTO2. In the first experiment, referred to as a “garden experiment”, six 2 m2 plots, enclosed with wooden fences (height 30 cm), were constructed in June 2003 in the city of Lahti. The plots were established in five home gardens, each containing two replicate plots of each of the three treatments (see below). A G R I C U L T U R A L A N D F O O D S C I E N C E Hagner, M. et al. Birch oil toxity 14 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 15 No attention was paid to possible differences in soil characteristics between the study sites. Dominant plant species in the yards were Poa sp. and Festuca sp. (Poaceae), Taraxacum officinale (Asteraceae) and Trifolium repens (Fabaceae). In one yard Ae- gopodium podagraria (Apiaceae) was abundant. Two leaf-litterbags (mesh size of 1 mm) contain- ing 2 g (dry mass) of Calamagrostis arundinacea (Poaceae) –leaf litter was placed in the soil to a depth of ca. 1 cm in each plot to examine the effects of BTOs on decomposition rate. Three treatments were established in each area: plots were sprayed with 1) BTO1 (corresponding to the amount of 0.5 dl m-2; n = 10), 2) BTO2 (0.5 dl m-2; n = 10) and 3) tap water ( = control; n = 10 in each treatment). BTOs (and water control) were sprayed on the 2 m2 area using a compressed air pump once at the start of the study. Because the weather during the entire study period was dry and warm, water (2 l) was sprayed weekly on to each plot. Two soil samples (Ø 5 cm, 5 cm deep) were taken from each plot at 9, 17, 38 and 70 days after establishing the experi- ments. The samples were stored (1−20 days; so that the mean storage time for the soils in each of the three treatment was the same) at 5 °C and the num- bers of nematodes and enchytraeids were counted (for methods, see below). The effect of BTOs on plants (total plant coverage %) was estimated in concurrently with taking the soil samples. At the final sampling, plants were harvested at randomly chosen 50 × 50 cm2 areas in each plot, identified, dried (24 h, 105 oC) and weighed. The leaf-litter bags were removed and the contents dried (24 h, 105 oC) and weighed. The second experiment, referred to as the “field experiment”, was carried out in an experimental field in central Finland (MTT Agrifood Research Finland, Toholampi), from mid June to late July 2005. An arable field containing numerous weed species and situated between two turnip rape fields, was divided into ten contiguous plots (1 m × 2 m). Five randomly chosen plots were sprayed with BTO1 (1.36 dl m-2) once at the start of the study using a compressed air pump. The control plots (n = 5) were treated with water only. The experiment was conducted over 42 days, within which time soil samples were taken at 4, 10, 20, 29 and 42 days after the initiation of the study. At each sampling time, three soil samples (Ø 4 cm, 5 cm deep) were taken from each plot. The samples were stored in cooling boxes and transported to the laboratory for analysis of nematodes numbers, the biomass of enchytraeids, and the activity and biomass of soil microbes (for methods, see below). Toxic effects of BTO1 on soil organisms and plants – mesocosm studies Experiment 1 was established as a mesocosm study at the garden area of the Department of Ecological and Environmental Sciences, University of Hel- sinki, Lahti, in summer 2004. The experimental design consisted of 75 mesocosms established in 1500 ml glass jars (Ø 10 cm, height 19 cm) filled with 400 g of fresh homogenised garden soil taken from a lawn adjacent to the Departmental building. Earthworms, roots and other larger plant remains were removed from the soil. Grass (Festuca rubra, Festuca ovina and Poa pratensis) seeds were sown in the mesocosms and kept under a plastic cover in natural light and temperature conditions in the garden. The weather during the study was variable with heavy rains (cumulative rainfall during the study was 147 mm) to periods of dry and warm weather (average temperature during the study was 14.3 ºC). After a stabilisation period of one month, three treatments were established, each with five replicates: the mesocosms were treated once with: 1) 100% BTO1 (0.5 dl m-2), 2) 5% BTO1 (0.5 dl m-2) and 3) water ( = control). We aimed at keeping the moisture content of the soil constant during the 7-week study by adding 40−80 ml of water weekly to each mesocosm. Five jars per treatment were randomly selected at day 1, 7, 20, 29 and 48 for destructive sampling in which the effects of the treatments on the numbers of nematodes, biomass of enchytraeids and microbial activity were examined. After the last sampling, the plants were uprooted, dried (24 h, 105 oC), and weighed. Experiment 2 was identical to the experiment 1 described above except that the former was con- ducted in 200 ml plastic bowls containing 100 g A G R I C U L T U R A L A N D F O O D S C I E N C E Hagner, M. et al. Birch oil toxity 16 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 17 moist, homogenised garden soil, and no plants were sown into the mesocosms. The mesocosms were weighted weekly and the water evaporated from the mesocosms during the 7-week study was replaced by adding tap water into the mesocosms. The bowls were kept at room temperature (+ 22 oC) in constant darkness. Sampling of soil fauna Nematodes were extracted from 5 g (fresh) soil samples using the wet funnel method by Sohlenius (1979), and enchytraeids were extracted from 30- 80 g soil samples using the wet funnel technique described by O’Connor (1955). The numbers of nematodes and enchytraeids were counted under a binocular microscope. In the mesocosm and field experiments the length of the enchytraeids was meas- ured and the biomass of the different size classes was estimated using the formula described by Ab- rahamsen (1973). The total biomass of enchytraeids in the garden experiment was calculated using an average dry mass of 25 µg per individual (Persson et al. 1980). The corresponding average dry mass of the nematodes was 0.03 µg (Persson et al. 1980). Water content of the soil samples was determined by placing the soil into an oven (70 ˚C) for 24 hours. Measuring of microbial activity and biomass Microbial activity was measured using basal res- piration as an estimate. Microbial biomass was determined using the substrate induced respiration (SIR) method described by Anderson and Domsch (1978). Basal respiration in the autotrophic (with plants) mesocosms was measured from 5 g soil samples during a period of 50 h using a respirom- eter (Nordgren 1988). Basal respiration in the field soil (100 g fresh soil) and in the mesocosm soil without plants were determined using EasyQuant infrared-carbon analyzer. Soils from the field site were stabilized for a period of 24 h before con- ducting the first measurement. Basal respiration in the mesocosms without plants was measured di- rectly from the test jars so that an air sample was taken through a hole in the lid using the syringe from the head-space of the mesocosm. After this the hole was closed and a second air sample was taken two hours after closing the lid. Basal res- piration was calculated from the difference in the amounts of CO2 between the two measurements. Microbial biomass (SIR), determined only from the field soils, was quantified using the EasyQuant infrared-carbonanalyzer two hours after an appli- cation of glucose (the application was performed immediately after the second measurement). Toxicity tests against Aporrectodea caliginosa and Folsomia candida The LC50 and NOEC values of BTO1 for the earthworm A. caliginosa were defined using adult individuals collected from an arable field close to the city of Lahti. In practice, Eisenia fetida is a standard earthworm species in terrestrial ecotoxicology tests in the EU, but in Northern Europe its occurrence is limited to sites rich in organic matter (Kula and Larink 1998). Therefore Aporrectodea caliginosa, a dominant endogeic earthworm species in the agro-ecosystems and grasslands (Kula and Larink 1998), was applied in the current study. The worms were maintained in the laboratory in the original field soil for a period of two weeks before the start of the experiment. The experiment was carried out according to the OECD guideline 207 (1984). Prior to the start of the experiment, BTO1, dissolved in distillate water, was mixed with artificial soil com- prised of 70% coarse sand, 20% kaolinite clay and 10% finely ground Sphagnum peat, and adjusted to pH 6.5 with CaCO3. Moisture content was adjusted to 35 % of the dry weight. The BTO1 concentrations used were 2695, 5380, 6725, 8070, 9420, 10760, 12200 and 13500 mg BTO1 kg-1 dry soil (n = 4). These concentrations were selected on the basis of results obtained in preliminary tests. Control soils (n = 4) received no BTO1. For each test, 750 g of A G R I C U L T U R A L A N D F O O D S C I E N C E Hagner, M. et al. Birch oil toxity 16 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 17 the test medium was placed into each jar (vol. 1000 ml) and ten earthworms (ranging from 300 to 600 mg wet mass), which had been conditioned for 24 hours in artificial soil and then washed quickly before use, were placed on the test medium surface. The earthworms were not fed during the 2-week study. After 7 and 14 days of exposure, alive worms were removed and counted. During the exposure, the microcosms were kept in a climate cabinet (20±1 °C) in continuous light to keep the photophobic earthworms in the test jars during the test. The effect of BTO1 (EC50 and NOEC values) on the reproduction of springtails, F. candida, was determined using the protocol published by the International Standards Organization (ISO 1999). The springtails were derived from a pure culture maintained at the Department of Ecological and Environmental Sciences, University of Helsinki. The test was performed in 125 ml glass jars (Ø 11 cm, height 19 cm), each filled with 30 g (wet mass) of artificial soil (70% coarse sand, 20% kaolinite clay and 10% finely ground Sphagnum peat) and adjusted to pH 5.8 with CaCO3. The test concen- trations of BTO1 were 50, 85, 140, 235, 390, 655, 1090, 1820, 3030, 5055, 8425, 14040 and 23400 mg BTO1 kg-1 dry soil (n = 5). These concentra- tions were selected on the basis of results obtained in preliminary tests. Control soils (n = 5) received no BTO1. BTO1 was added to the artificial soil with deionised water and the moisture content was adjusted to 25 % of the dry weight. Soils were allowed to stabilise for one day prior to the ex- periment. Ten springtails (10−12 days old) were introduced to each replicate microcosm, and 2 mg of dry yeast was placed on the soil surface as a diet. The lids of the microcosms were otherwise closed during the experiment but removed twice a week for aeration. Fresh yeast was added once a week. The experiment was carried out in a climate cabinet (20 °C) under a light:dark cycle (16:8 h), and conducted for 28 days thus allowing sufficient time for the springtails to lay eggs. At the end of the experiment, the soil from each microcosm was mixed with 200 ml of tap water and gently stirred. The number of springtails floating on the surface of the water was recorded. Statistical analysis To test for treatment effects in the field, a repeated measures ANOVA was conducted. When interaction between the time and treatment was observed the level of one treatment was fixed and the effects of the other treatment were studied within this level using one-way ANOVA (simple-effects model) (Zar 1999). The data was not always normally distributed and the variances were sometimes heterogeneous, even after data transformation. We were interested in the BTO -induced temporal patterns in the field studies and therefore repeated-measures (not avail- able in non-parametric test packages) model was considered to be the most appropriate analysis, despite the fact that data transformation did not improve the data and as such warrants caution when interpreting the data. Enchytraeid and nematode data from the gar- den soils were analyzed using a repeated measures ANOVA with the custom-model. Custom model -option was applied when the degrees of freedom did not suffice due to the incomplete experimental design. Using this model interactions between the factors are not calculated. To find out whether ap- plication of BTO1 results in changes in the plant community composition, the Principal Component Analysis (PCA) was applied. To detect possible differences in the numbers of soil organisms between differentially treated mesocosms, a 2-way-ANOVA was used with time and BTO-treatment as factors. Data was not al- ways normally distributed and the variances were sometimes heterogeneous. Transformations had no effect on data. Effects on plants in the mesocosms were analyzed by one-way-ANOVA, and differ- ences between treatments were compared using Tukey’s -test. Probit analysis was used to determine LC50 val- ues for earthworms with 95% confidence intervals. The effective concentration (EC50) for juvenile pro- duction of Folsomia candida was calculated using DEBtox-package (2004). The DEBtox Growth Model was selected because it provided a better fit of reproduction data for BTO1 in F. candida. According to the Growth Model, the toxicant in- A G R I C U L T U R A L A N D F O O D S C I E N C E Hagner, M. et al. Birch oil toxity 18 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 19 fluences mother´s energy allocation on offspring production. All statistical analyses were performed using the SPSS statistical package (SPSS 1999). Results Effects of BTOs on soil fauna Birch tar oils had no consistent effects either on the numbers of enchytraeid worms in the garden soil in the city of Lahti (repeated measures ANOVA, p = 0.642), or on their biomass in the Toholampi field study (repeated measures ANOVA, p = 0.85). The biomass of enchytraeids in garden soil varied between 1.25 and 15 µg g-1 dry soil, and their biomass in the Toholampi field study ranged from 0 to 2.5 µg g-1 dry soil. Similarly, the effect of BTO1 on the bio- mass of enchytraeids in the mesocosm experiments without plants (2-way-ANOVA, p = 0.33) and with plants was insignificant (2-way-ANOVA, p = 0.52). The biomass of enchytraeid worms in these studies ranged from 0 to 22 µg g-1 dry soil. The numbers of nematodes did not differ signifi- cantly between the treatments either in the garden experiment (repeated measures ANOVA, p = 0.117), or in the mesocosm experiments without plants (2-way-ANOVA, p = 0.52), or with plants (2-way- ANOVA, p = 0.52). In garden soil the number of nematodes fluctuated quite much being between 10.5 and 49.5 individual g-1 dry soil. In the meso- cosm experiments with plants their numbers varied from 5.8 to 6.9 individual g-1 dry soil and in meso- cosm without plants from 14.7 to 31.7 individual g-1 dry soil. In the Toholampi field study, the number of nematodes in BTO1 treated plots decreased at the last sampling time as compared to the control soils (ANOVA, simple effects, p > 0.05) (Fig. 1). BTO1 had no effect on microbial activity in the mesocosms with plants (2-way ANOVA, p = 0.52). However, when plants were absent, microbial res- piration in the mesocosms treated with 100% BTO1 increased 1 day after BTO application, being sig- nificantly higher than in the control systems and in the mesocosms treated with 5% BTO1 (Tukey, p < 0.001) (Fig. 2). Similar to the mesocosm experi- ment without plants, the field experiment showed that microbial activity was positively affected soon after addition of BTO1 (ANOVA, simple effets, p > 0.05) (Fig. 3). However, from day 9 onwards, 0 10 20 30 40 50 0 10 20 30 40 Time (d) Number of nematodes g-1 dry soil BTO1BTO1BTO1 Fig 1. The number (mean values + SD, n = 5) of nema- todes in the BTO1-treated (black line) and control plots (grey line) in the Toholampi field experiment 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0 10 20 30 40 50 Time (d) μg CO 2 -C g-1 soil (dw) h-1 Fig 2. The effects of BTO1 treatments on the microbial activity during the mesocosms experiment without plants (mean values + SD, n = 5). Blue line = Control, black dashed line with squares = 5% BTO1, black line with triangles = 100% BTO1. 0 1 2 3 4 5 6 7 8 0 10 20 30 40 Time (d) μg CO2 g -1 soil (dw) h -1 Fig 3. The effect of BTO1 on microbial activity at the Toholampi field experiment (mean values + SD, n = 5). Blue line = Control, black line = BTO treated plots. A G R I C U L T U R A L A N D F O O D S C I E N C E Hagner, M. et al. Birch oil toxity 18 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 19 the influence of BTO1 on microbial activity was negative (ANOVA, simple effects, p > 0.05) (Fig. 3) and BTO1 had no effect on microbial biomass at any sampling time (repeated measures ANOVA, p = 0.38). There was a rapid loss in leaf-litter mass during the garden experiment with no significant difference between the treatments (ANOVA, p = 0.384). After 2.5 months, approximately 60% of the initial litter mass remained in the leaf-litter bags. Effects of BTOs on plants Plants showed signs of stress and began to wither immediately after BTO applications. In the garden plots, 40% and 60% of the total coverage of the plants withered within the first day of BTO1 and BTO2 application, respectively. However, plants started to recover one month after BTO treatments, and after 2.5 months no difference in plant biomass between the variously treated plots was observed (ANOVA, p = 0.532). Aegopodium podagraria made an exception; only a fraction (5 % of total cover- age) recovered in BTO2 treated plots, and 40% in BTO1 treated plots. Birch tar oils affected neither plant community composition nor plant species diversity as depicted from the results of the PCA analysis (results of the PCA-analysis not shown). The number of plant taxa varied between 2 and 14 at the experimental plots. In mesocosms with plants, 100% BTO1 had a negative effect (Tukey test, p = 0.01) on plant biomass: 50 days after application, the biomass of 100% BTO1-treated plants was 84% less than in the control mesocosms. Application of 5% BTO1 showed no consistent effects on plant biomass pro- duction (Tukey test, p = 0.76). EC 50 -values for collembolas and earth- worms In F. candida EC50-value for juvenile production was 5100 mg BTO1 kg-1 dry weight soil (S.D. 1280 mg kg-1) (Fig. 4). No mortality (NOEC = no observed effect concentration) occurred at 3033 mg BTO kg-1 dry weight soil. Soil pH was variable during the treatment: at the beginning of the test it varied between 5.8 (controls) and 4.5 (23400 mg BTO kg-1 dry weight) (soil pH was not adjusted between treatments). The mortality in the control jars was low and the animals reproduced during the test. At highest test concentrations (8425−23400 mg BTO1 kg-1 dry soil), all springtails died. Variation in reproduction in controls did not exceed 30%. In the earthworm test 14-day LC50 for Apor- rectodea caliginosa was 6560 mg BTO kg-1 dry weight (95% confidence limits 6323-6834 mg BTO kg-1 dry weight) (Fig. 5). NOEC value was 2694 mg BTO kg-1 dry weight. There was no mortality in the control jars. 0 100 200 300 400 500 600 700 800 900 0 85 235 655 1820 5055 14040 Concentration (mg kg-1) Number of offspring 0 1 2 3 4 5 6 7 8 9 10 Number of adults Number of offspring Number of adults Fig 4. Chronic toxicity of BTO1 to F. candida adults ( = lines with squares) and offspring ( = lines with circles). 0 20 40 60 80 100 0 2695 5380 6725 8070 9415 10760 12200 13500 Consentration (mg kg-1 ) Survival (%) 7 d 14 d Fig 5. Survival rate of the earthworm A. caliginosa as a function of concentrations of BTO1 after 7 ( = lines with circles) and 14 ( = lines with squares) day exposure. A G R I C U L T U R A L A N D F O O D S C I E N C E Hagner, M. et al. Birch oil toxity 20 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 21 Discussion Field tests Enchytraeid worms, nematodes, soil microbes, earthworms and springtails were selected as test biota because they fullfill several criteria required for toxicity tests. They are present in a wide range of ecosystems, occur abundantly, play a key role in the functioning of the soil ecosystem, are easy to use, collect and culture, come into contact with a variety of stress factors (the soil solution, the solid phase, and the gaseous phase in soil), and are sensitive to environmental stresses (Didden and Römbke 2001, Römbke and Moser 2002, Schloter et al. 2003). Importantly, changes in the soil faunal community composition has been shown to mark- edly affect e.g. microbial and plant biomass and community composition, thereby controlling the functioning of the entire ecosystem (Bradford et al. 2002, Wardle et al. 2004). BTOs proved to have a clear negative influence on plant biomass and species diversity in the home gardens, but this effect was short-term. Aegopo- dium podagraria, however, proved to be a sensitive species to BTOs with a poor recovery rate during the growing season. This species is a common, fast- growing weed in gardens that has hitherto been tedious to control. The reason for the sensitivity of A. podagraria to BTOs remains open. In the mesocosm experiment established under a plastic shelter, 100% BTO1 imposed a clear nega- tive effect on the biomass of plants which continued up to 50 days post application. It is highly likely that the efficacy of BTO1 was reduced under field conditions due to rain and solar radiation, which can explain the differences between the field (open sky) and mesocosm (plastic roof) experiments. BTO1 had no consistent effects on soil micro- bial biomass. Interestingly, both laboratory and field experiments suggest 100 % BTO1 having a positive effect on microbial activity immediately after application. This is often a typical reaction when added resources are rapidly consumed by mi- crobes (Meli et al. 2003). BTOs, being composed of numerous chemical compounds, supposedly in- clude substances that are readily available for soil microbes. On the other hand, the enhanced respira- tion activity can, at least partly, be explainable by the so-called “gasping-effect”, whereby toxic com- pounds enhance soil respiration but not microbial growth (Bååth 1989, Connell et al. 1999). In the Toholampi field experiment, BTO1 had a negative influence on soil microbial activity in the last samplings. As depicted from the SIR results, BTO1 did not reduce microbial biomass. This can result from the withering of the plants in the BTO- treated plots. There is an increasing amount of evi- dence suggesting that the structure and activity of soil food webs can, by large part, be controlled by the living plant community (Wardle 2002, Bardgett 2005) regulating, for example, an amount of certain organic compounds in the soil by root excretion (Smolander and Priha 2003, Wardle et al. 2004). We hypothesize that in the BTO-treated plots in To- holampi, a decreased amount of root excudates by the dead/wilting plants lead to drastically reduced nutritional resources for the soil microflora causing the microbes to enter into a dormant state (Mar- tikainen 2003). This negative effect is, however, likely to be short-term due to the resource input by the dead plant biomass later on in the growing season. BTOs appeared to have little effect on en- chytraeid worms and nematodes. This result was unexpected as both enchytraeids and nematodes are indirectly influenced by changes taking place above the ground (Wardle et al. 2004). The drastic changes by BTOs in the primary producers should therefore have cascaded down to the soil food web. That this proved not to be the case suggests that nematodes (as a group) and enchytraeid worms are less dependent on dead organic matter than on root exudates produced by living plants. Because BTOs had no effect on the survival of enchytraeids which are commonly considered as sensitive (Salminen and Haimi 1999, Salminen and Haimi 2001) key- stone species (Huhta et al. 1998) living in the soil surface layer (Nurminen 1967), the direct and in- direct effects of BTOs to soil food-webs appear insignificant. A G R I C U L T U R A L A N D F O O D S C I E N C E Hagner, M. et al. Birch oil toxity 20 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 21 Toxicity tests Earthworms are widely used in laboratory toxicity tests because of their important role in ecosystems and sensitivity to numerous chemical stressors (Didden and Römbke 2001). The LC50 (14 d) of BTO1 for Aporrectodea caliginosa was 6560 mg BTO1 kg-1 dry soil. Most OECD countries follow the classification system where LC50-values being > 1000 mg kg-1 (kg of dry soil weight) indicate pesticides to be practically nontoxic for earthworms (OECD 2003). The LC50 of BTO1 to Aporrectodea caliginosa was over six times higher. As far as we know, literature values for the toxicity of BTO1 in other species are not available for comparison. In general, the LC50 values are not comparable in experiments conducted on different experimental conditions and on different time scales. In the present study, the acute toxicity was within range of LC50 values reported in the literature indicating low toxicity of a single chemical (Russom et al. 1997). Springtails are present in most soil habitats at densities of up to 104 – 105 m-2 (Petersen and Lux- ton 1982) and play a vital role in the decomposition of organic material (Cragg and Bardgett 2001). The species F. candida is commonly applied as a test organism for estimating the effects of chemicals on non-target organisms (Fountain and Hopkin 2005) and it is among the most sensitive springtails to an array of chemicals (Chernova et al. 1995). The EC50 for the inhibition of juvenile production of F. candida was 5100 mg BTO1 kg-1 dry soil. Although the clear negative impact of this mix- ture on both the lethality of earthworms and the re- production of collembolas was seen, the observed response is attributed to the combination of chemi- cals present in BTO1 and evidently the responses were not correlated with concentration of the main component producing toxicity. In theory, if a single chemical from the BTO1 mixture has a prominent lipophilicity and specific target site, it can alone give rise to the observed responses. In the field, BTO1 acts as a non-selective herbicide by destroying virtually all growth of the aboveground parts of plants. Therefore, BTO treatments should be done before the emergence of cultivated seedlings. We have ample anecdotal evidence suggesting the suitability of BTO1 as a herbicide, for example in potato and carrot fields, and in the row width of berry shrubs and fruit trees. To conclude, birch tar oils appeared to have no consistent, direct negative effects on enchytraeid worms, nematodes or soil microbial biomass when used at concentrations between 500-1360 l ha-1. In- stead, 100 % BTO1 (500−1360 l ha-1) showed a positive effect on microbial activity immediately after application, suggesting that easily utilized substances in BTO1 were rapidly consumed by microbes. Futhermore, the high LC50 value for the earthworm A. caliginosa indicates BTO1 to be practically nontoxic to earthworms. Altogether the risks caused by BTO1 in the soil environment are insignificant and short-term as compared to the many of chemical products applied for similar purposes. In spite of the low environmental risks shown in this study a thorough investigation of the chemi- cal composition of BTO is required. Although BTO is included in the worldwide substance database (CAS number 8001−88−5, American Chemical So- ciety 2007), knowledge of the chemical composi- tion of BTO is essential also for the EU registration of BTO as a biological plant protection product. Acknowledgements. We are grateful to Asko Riihelä from the Lahti Region Environmental Service for finding the experimental sites for our studies. We warmly thank all the people who gave us permission to conduct experiments in their home yards. A G R I C U L T U R A L A N D F O O D S C I E N C E Hagner, M. et al. Birch oil toxity 22 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 23 References Abrahamsen, G. 1973. Studies of body-volume, body-sur- face area, density and live weight of Enchytraeidae (Oli- gocaheta). Pedobiologia 13: 6−15. American Chemical Society 2007: Homepage. Update (29 Apri4.209). Cite (13 June 209). Available on the internet: http://www.cas.org/expertise/cascontent/registry/reg- syst.html?WT.mc_id = casrn708&WT.srch = 1. Anderson, J.P.E. & Domsch, K.H. 1978. A physiological method for the quantitative measurement of microbi- al biomass in soils. Soil Biology and Biochemistry 10: 215−221. Bardgett, R.D. 2005. The biology of soil: a community and ecosystem approach. Oxford University Press. 242 p. Bradford, M.A., Jones, T.H., Bardgett, R.D., Black, H.I.J., Boag, B., Bonkowski, M., Cook, R., Eggers, T., Gange, A.C., Grayston, S.J., Kandeler, E., McCaig, A.E., New- ington, J.E., Prosser, J.I., Setälä, H., Staddon, P.L., Tord- off, G.M., Tscherko, D. & Lawton, J.H. 2002. Impacts of soil faunal community composition on model grassland ecosystems. Science 298: 615−618. Bűnemann, E.K., Schwenke, G.D. & Van Zwieten, L. 2006. Impact of agricultural inputs on soil organisms – a re- view. Australian Journal of Soil Research 44: 379−406. Bååth, E. 1989. Effects of heavy metals in soil on microbi- al processes and populations: A review. Water, Air, and Soil Pollution 47: 335−379. Chernova, N.M., Balabina, I.P. & Ponomareva, O.N. 1995. Changes in population growth rate of sprintails (Collem- bola) under the influence of herbicides. Polskie Pismo Entomoogiczne 64: 91−98. Connell, D., Lam, D., Richardson, B. & Wu, R. 1999. In- troduction to ecotoxicology. Blackwell Publishing. Unit- ed Kingdom. 170 p. Cragg, R.G. & Bardgett, R.D. 2001. How changes in soil faunal diversity and composition within a tropic group influence decomposition processes. Soil Biology and Biochemistry 33: 2073−2081. DEBtox, 2004. Publication v2.0. Department of Theoreti- cal Biology, Vrije Universitet, Amsterdam, Netherlands. Didden, W. & Römbke, J. 2001. Enchytraeids as indicator organisms for chemical stress in terrestrial ecosystems. Ecotoxicology and Environmental Safety 50: 25−43. EU (European Union) 1996. Technical guidance docu- ments in support of the Commission Directive 93/67/ EEC on risk assessment for new notified substances and the Commission Regulation (EC) 1488/94 on risk assessment for existing substances, Brussels, Belgium. Fountain, M.T. & Hopkin, S.P. 2005. Folsomia candida (Col- lembola): A “standard” soil arthropod. Annual Review of Entomology 50: 201−222. Hagner, M. 2005. Koivutisle torjunta-aineena: vaikutuk- set lehtokotiloon (Arianta arbustorum) ja maaperään (in Finnish). Masters thesis. University of Helsinki. Depart- ment of Ecological and Environmental Sciences. 39 p. Hagner, M., Penttinen, O-P., Pasanen, T., Tiilikkala, K. & Setälä, H. 2010. Acute toxicity of birch tar oil on aquatic organisms. Agricultural and Food Science 19: 24–33. Huhta, V., Persson, T. & Setälä, H. 1998. Functional impli- cations of soil fauna diversity in boreal forests. Applied Soil Ecology 10: 277−288. ISO 1999. Soil quality-inhibition of reproduction of Colle- mbola (Folsomia candida) by soil pollutants. Rep. No. ISO 11267:1999(E). Geneva. 16 p. Kula, H. & Larink, O. 1998. Tests on the earthworms Eisenia fetida and Aporrectodea caliginosa. In: Løkke, H. & van Gestel, C. A. M. (ed.) Handbook of soil invertebrate tox- icity tests. John Wiley & Sons Ltd, Chichester. p. 95−112. Lindqvist, I., Lindqvist, B., Tiilikkala, K., Hagner, M., Penttinen,O-P., Pasanen. T. & Setälä, H. 2010. Birch tar oil is an effective mollusc repellent: field and laborato- ry experiments using Arianta arbustorum (Gastropoda: Helicidae) and Arion lusitanicus (Gastropoda: Arionidae). Agricultural and Food Science 19: 1–12. M ar tikainen, P. 20 03. M etsämaan mikrobisto – In: Mälkönen, E. (ed.), Metsämaa ja sen hoito (in Finnish). Karisto, Hämeenlinna. p. 101−114. Meli, S.M., Badalucco, L., English, L.C. & Hopkins, D.W. 2003. Respiratory responses of soil micro-organisms to simple and complex organic substrates. Biology and Fertility of Soils 37: 96−101. Nordgren, A. 1988. Apparatus for the continuous long- term monitory of soil respiration rate in a large number of samples. Soil Biology and Biochemistry 20: 955−957. Nurminen, M. 1967. Ecology of enchytraeids (Oligochae- ta) in Finnish coniferous forest soil. Annales Zoologici Fennici 4: 147−157. O’Connor, F. B. 1955. Extraction of enchytraeid worms from a coniferous forest soil. Nature 175: 815−816. OECD 1984. OECD guideline for testing of chemicals. No. 207. “Earthworm, Acute Toxicity Tests”. Paris. Organi- zation for Economic Development. 9 p. OECD 2003. OECD environmemt, health and safety pub- lications. Series on pesticides, No. 15. Persistent, bio- accumulative, and toxic pesticides in OECD member countries. 67 p. Persson, T., Clarholm M., Lundkvist, H., Söderström, B. & Sohlenius B. 1980. Trophic structure, biomass dynam- ics and carbon metabolism in a Scots pine forest. − In: Persson T (ed.), Structure and function of northern co- niferous forest − an ecosystem study. Ecol Bull (Stock- holm) 23: 419−459. Petersen, H. & Luxton, M. 1982. A comparative analysis of soil fauna populations and their role in decomposi- tion processes. Oikos 39: 288−388. Russom, C.L., Bradbury, S.P., Broderius, S.J., Ham- mermeister, D.E. & Drummond, R.A. 1997. Predicting modes of toxic action from chemical structure: acute toxicity in the fathead minnow (Pimephales promelas). Environmental Toxicology and Chemistry 16: 948−967. Römbke, J. & Moser, T. 2002. Validating the enchytraeid reproduction test: organisation and results of an interna- tional ringtest. Chemosphere 46: 1117−1140. Salminen, J. & Haimi, J. 1999. Horizontal distribution of copper, nickel and enchytraeid worms in polluted soil. Environmental pollution 104: 351−358. Salminen, J. & Haimi, J. 2001. Life history and spatial dis- tribution of the enchytraeid worm Cognettia sphagneto- rum (Oligochaeta) in metal-polluted soil: below-ground sink-source population dynamics? Environmental Toxi- cology and Chemistry 20: 1993−1999. Salonen, J., Tiilikkala, K., Ruuttunen, P., Lindqvist, I. & Lindqvist, B. 2008. Birch Tar Oil: A Potential Herbicide A G R I C U L T U R A L A N D F O O D S C I E N C E Hagner, M. et al. Birch oil toxity 22 A G R I C U L T U R A L A N D F O O D S C I E N C E Vol. 19(2010): 13–23. 23 from the Forests of Finland. In: Abstracts of the 5th In- ternational Weed Science Congress. Weeds local prob- lems/global challenge. Vancouver, British Columbia, Canada. p. 286−287 Schloter, M., Dilly, O. & Munch, J.C. 2003. Indicators for evaluating soil quality. Agriculture, Ecosystems and En- vironment 98: 255−262. Smolander, A. & Priha, O. 2003. Puulajin merkitys metsä- maan viljavuuden hoidossa. In: Mälkönen, E. (Ed.), Met- sämaa ja sen hoito (in Finnish). Karisto, Hämeenlinna. p. 155−158 Sohlenius, B. 1979. A carbon budget for nematodes, roti- fers and tardigrades in a Swedish coniferous forest soil. Holarctic Ecology 2: 30−40. SPSS, 1999. SPSS for Windows 10.0. publication 9.0.1. SPSS, Chicago. Tiilikkala, K. & Salonen, J. 2008. Birch tar oil (BTO) a potential biological pesticide from forests. In: Biorefinery Joint Call Info Day. Information Booklet. Bryssel. p. 85 Wardle, D.A., Bardgett, R.D., Klironomos, J.N,. Setälä, H., van der Putten, W.H. & Wall, D.H. 2004. Ecological linkages between aboveground and belowground biota. Science 304: 1629−1633. Wardle, D.A. 2002. Communities and ecosystems − linking the aboveground and belowground components, Princ- eton University Press, Princeton. 392 p. Zar, J.H. 1999. Biostatistical analysis. 4 th edition. Prentice- Hall, Inc. Englewood Cliffs, New Jersey. 736 p. SELOSTUS Koivuöljytisleen vaikutukset maaperän eliöihin ja kasveihin Marleena Hagner, Tiina Pasanen, Bengt Lindqvist, Isa Lindqvist, Kari Tiilikkala, Olli-Pekka Penttinen ja Heikki Setälä Helsingin yliopisto ja MTT Effects of birch tar oils on soil organisms and plants Introduction Material and methods Ecological effects of BTOs on non-target soil organisms and plants – field investigations Toxic effects of BTO1 on soil organismsand plants – mesocosm studies Sampling of soil fauna Measuring of microbial activity and biomass Toxicity tests against Aporrectodea caliginosa and Folsomia candida Statistical analysis Results Effects of BTOs on soil fauna Effects of BTOs on plants EC50-values for collembolas and earthworms Discussion Field tests Toxicity tests References