Environ. Sci. Technol. 2009, 43, 2450–2455
Altitudinal and Chiral Signature of Persistent Organochlorine Pesticides in Air, Soil, and Spruce Needles (Picea abies) of the Alps H E Q I N G S H E N , †,‡ BERNHARD HENKELMANN,† WALKIRIA LEVY,† ADAM ZSOLNAY,§ PETER WEISS,| GERT JAKOBI,† MANFRED KIRCHNER,† WOLFGANG MOCHE,| KATHARINA BRAUN,| AND K A R L - W E R N E R S C H R A M M * ,†,⊥ Helmholtz Zentrum Mu ¨ nchen - German Research Center for Environmental Health, Ingolsta¨dter Landstrasse 1, D-85764 Neuherberg, Germany, University Department of Growth & Reproduction, Rigshospitalet, Blegdamsvej 9, DK-2100, Copenhagen, Denmark Helmholtz Zentrum Mu ¨ nchen German Research Center for Environmental Health, Institute of Soil Ecology, Ingolsta¨dter Landstrasse 1, D-85764 Neuherberg, Germany, Austrian Federal Environment Agency, and Department fu ¨ r Biowissenschaftliche Grundlagen, TUM-Technische Universita¨t Mu ¨ nchen, Weihenstephaner Steig 23, D-85350 Freising, Germany
Received June 27, 2008. Revised manuscript received October 20, 2008. Accepted October 20, 2008.
The present study investigated the distribution, transportation, and biodegradation of the selected chiral persistent organochlorine pesticides (OCP) in the Alps. In the complex environment, we found the movement and fate of OCP could be defined by many factors. Taking HCE as an example, below the timberline its accumulation from air into SPMD increased with altitude and seasonally changed, but the trends reversed above the timberline. In soil, the tendency of HCE concentrations vs organic materials followed a sigmoid curve, and HCE concentration-altitude correlations are positive in central Alps but negative in southern Alps. The HCE enantiomeric ratios (ERs) in soil correlated to HCE isomers concentrations, the humus pH values, and the sampling site altitudes. HCE shift from humus to mineral soil can also be traced by ERs. The altitudinal and longitudinal trends in needles suggested that R-HCH has a more complex movement than HCE in Alps. In conclusion, altitude conducted condensation, plant canopies, organic material in soil, and geographic specific precipitations may affect OCP distributions and transportation, whereas altitude conducted temperature and soil pH could dictate their fate in the environment.
Introduction Due to their persistence, the organochlorine pesticides (OCP) still can be found in soil and therefore continue to cycle * Corresponding author phone: +498931873147; fax: +498931873371; e-mail:
[email protected]. † Institute of Ecological Chemistry. ‡ Rigshospitalet. § Institute of Soil Ecology. | Austrian Federal Environment Agency. ⊥ TUM-Technische Universita¨t Mu ¨ nchen. 2450
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 7, 2009
through the environment, as soil is a potential source for atmospheric input through volatilization. Many processes can affect OCP concentrations in the environment. The longrange atmospheric transport (LRAT) (1), air-soil exchange (2), forest filter effect (3-5), “grasshopper” effect and fractionation (condensation) in mountain areas (6-8), and microbial mediated processes (9), which are all correlated with climate conditions, could act together in real environment. The Alpine chain, which lies in a transitional region between the Atlantic Ocean, the Mediterranean Sea, and the European continent, is located on the southern side of the extratropical westerlies and constitutes a frontline between the Mediterranean and the North Atlantic climate zone (10). This area has a complex mountain environment. The object of this study was to investigate the transportation and fate of OCP in this environment by using the combination of information from different types of samples: semipermeable membrane devices (SPMD), Norway spruce needle and soil (humus and mineral soil), which were collected from sites along Alps chains, and some of these sites were altitude profiles. In order to trace the fate of OCP regarding the soil microbial activity, the selected OCP are chiral compounds: namely alpha-hexachlorocyclohexane (R-HCH), cis-heptachloroepoxide (HCE), and oxychlordane (OXC). Because of the enantioselective biodegradation of these chiral OCP in soil, their enantiomeric ratios contain their degradation information (9, 11). The present study measured both of the concentration and ERs for R-HCH, HCE, and OXC in samples humus and mineral soil. Therefore, geographic factors (i.e., altitude, longitude, and latitude), soil characteristics (i.e., organic material content, pH), and ERs in soil will be discussed with reference to weather conditions (i.e., yearly mean temperature and precipitation). Combining the sampling information and Alps climate feature, it is expected that our data can give an overview of the distribution of these OCP in Alps, including their altitude trends. Also it is expected to help us understand their transportation between air and soil and their accumulation and degradation in the complex Alpine environment.
Experimental Section Sampling Sites. 35 sampling sites were located along the Alpine mountainous chain where 7 of them were altitude profiles (Figure 1). The altitude profiles are monitoring points at different heights (4 to 5 points from 730 to 1779 m a.s.l. for soil and needle samples and up to 3100 m a. s. l. for some SPMD) at the same geographical site. In general, the sampling sites were at an average altitude of 1400 m a. s. l.; concerning the height of the singular sites above valley ground, there are certain differences between the central and peripheral Alpine regions. The sites are situated in forest stands with dominant spruce trees (Picea abies (L.) Karst), away from possible contaminant sources. Needle, humus, and mineral soil samples were collected at all sampling sites. SPMD were deployed at the altitude profiles, below timberline in forest clearings and above timberline on buildings (mountain research stations: Sonnblick, Schneefernerhaus/Zugspitze, Weissfluhjoch). Soil Sampling Procedure. Soil samples from the remote Norway spruce forest were sampled between late September and October 2004 of the main campaign. The humus samples were collected by taking the whole humus found within a 30 × 30 cm metal frame. Ten pits along a 5 × 30 m rectangular grid were randomly collected per sampling site and mixed as a pooled sample. Mineral soil was also sampled after humus collection from the cores of the pits from 0 to 10 cm 10.1021/es801782n CCC: $40.75
2009 American Chemical Society
Published on Web 11/14/2008
FIGURE 1. Sampling sites with a mean precipitation chart of June (mm per day) (modified from http://www.monarpop.at and http:// www.map.meteoswiss.ch/map-doc/rr_clim.htm). Yellow sphere markers show the standard sites and orange ones show the altitude profile sites. AT-41, 44, 49, 50, 52, 55, CH-08, DE-20, 21, 22, and 23 are grouped into the north Alps sites; AT-42, 43, 45, 46, 47, 48, 51, 53, 54, 56, 57, CH-01, 02, 03,04, 05, 07, and IT-12 are grouped into central Alps sites; and the rest of the sampling sites CH-06, IT-04, 05, 06, 10, 11, SL-31, 32, 33, and 34 were set as the southern Alps sites. layer depth. After sampling the jars were immediately sealed. sampling sites (Figure 1). SPMD accumulate gaseous-phase At the laboratory, the samples were stored at -20 °C. pollutants from the air, whereas OCP associated with Needle Sampling Procedure. Needles of Norway spruce particulate material are basically not absorbed. SPMD were were sampled between late September and October 2004 deployed 3 m above the ground, thus the OCP concentrations (Oct-04) of the main campaign. Sampling was repeated at mainly reflect the accumulated ambient air concentrations some subsample sites between September and October 2005 instead of the equilibrium concentrations between air and (Oct-05) and between May and June 2006 (May-06). Three the close soil surface (2, 12). Although Period-3 embraces to five branches with different orientations were cut from Period-1 and Period-2, the concentrations of Period-3 the top seventh whirl of two dominant, vital adult trees. Six samples cannot be expected to be the sum of the concentramonths old twigs of the current year were collected, pooled, tions of Period-1 and Period-2 (SI Figure S1). For R-HCH, and transferred to the laboratory in airtight precleaned glass there is no correlation between the concentration of Period-3 jars, which were kept in dry ice. At the laboratory the samples samples and the sum concentrations of Period-1 and Period-2 were stored at -20 °C until deneedling. samples. Linearly regressing OXC and HEC, Period-3 sample SPMD Sampling Procedure. Low Density Polyethlylene concentrations are significantly correlated with the sum (LDPE) membrane lay-flat tubes (length 23 cm, width 2.5 concentrations of Period-1 and Period-2 samples. Period-3 cm, thickness ≈ 65 ( 2 µm) were filled with 0.7 mL of triolein partly represents the sum of Periods-1 and Period-2 for HCE 99% (Sigma-Aldrich, Taufkirchen, Germany) and heat sealed and OXC only when the passive device is still operating in under inert gas conditions (N2) in a glove chamber. Then, the linear uptake stage (kinetic uptake). In the case of R-HCH, the SPMD were placed into clean glass vials under nitrogen the device may be operating in the curvilinear uptake stage atmosphere to avoid contamination and stored at -20 °C. (13, 14). These hermetically sealed vials were used to transport the Although deployed in the forest clearings, because the SPMD from and to the place of deployment. The membrane forest filter effects (FFEs) can enforce the OCP accumulation devices were placed in Stevenson screen boxes of untreated (4, 5) in the nearby forest soil and the mobility of OCP wood (50 cm × 50 cm × 40 cm) at 3 m above ground level evaporation in air, SPMD samples were grouped into two in forest clearings resp. on buildings. These deployment parts: sampling sites below timberline and sampling sites devices hinder the wet deposition, direct solar radiation, and above timberline. For the below timberline sites, only the airflow turbulences but allow adequate airflow through them. altitude increase tendency of HCE in Period-3 is significant The sampling of altitude profiles was performed such as (SI Figure S2-1-S4-3). Compare to the other two campaigns, indicated in Figure 1. Two sampling campaigns were carried the concentrations were lower for Period-2 campaign (SI out consecutively: May - November 2005 (Period-1) and Table S2-1). That shows the condensation effect (6) reduced December 2005 - May 2006 (Period-2). A third campaign at winter because of the lower temperature and the smaller embracing the Period-1 and Period-2, thus from May 2005 temperature lapse rates than at summer (15). For the above to May 2006 (Period-3), was also performed. After sampling, timberline samples, the altitudinal decrease tendency is clear the SPMD were placed in sealed flasks and kept in dry ice. for HCE and OXC, but these trends are only significant for The samples were stored at -20 °C until analysis. For sample HCE in Period-1 and Period-3 campaigns and for OXC in preparation, instrument measurement, and statistical analyPeriod-3 campaign (SI Figures S2-1-S4-3). It may be also sis see sample preparation and measurement in the Supdue to the too small sample size. The decreasing concentraporting Information (SI). tion trends of OXC and HCE in above timberline samples may be related to the strong drop of total organic material Result and Discussion in soil (8) and the loss of forest canopy (4, 5), which could Altitude Trends and Seasonal Change of OCP in SPMD. All SPMD was collected from the northern and central Alps result in the diminishing concentrations of HCE and OXC in VOL. 43, NO. 7, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2451
FIGURE 2. -1, -2, and -3: Correlations between OCP and the total organic material (TOM) of soil samples (humus ) red dots, mineral soils ) black dots), where TOM is the sum of the organic carbon and nitrogen content. soil above timberline. Different from OXC and HCE in Period-1 and Period-3 campaigns, the altitudinal increases are significant for R-HCH for all altitude profiles (SI Figures S2-1 and S2-3). It may suggest that, above the timberline, the condensation effect (6) is still an important mechanism for the transportation of R-HCH at summer but not at winter. Therefore, comparing OXC and HCE, the trends of R-HCH along the altitude were less affected by FFEs likely because the concentrations of R-HCH in atmosphere are much higher than the concentrations of HCE and OXC (SI Tables S2-1 and S2-2). In addition the possible isomerization from lindane (16) could contribute to the amount of R-HCH transport and its condensation along altitude even at higher altitudes at summer. In conclusion, the altitude differences of OCP accumulation in SPMD substantially reflect the temperature lapse along altitude and the plant canopies’ difference (3, 4); the campaign differences mainly reflect the temperature difference (including the variation of lapse rates) between different seasons. It is also possible that the temperatures below -4 °C influence the SPMD uptake due to freezing of triolein (14). This effect generates more resistance to the air-SPMD mass transfer hindering the OCP accumulation within the sampling device. However, the comparability of results obtained by SPMD measurements in forest clearings and on buildings may be reduced caused by different boundary layer characteristics in the two circumstances. Additionally, a regression through all sites of the altitude profiles or trough all summit sites may be influenced by regional variations of the concentrations and of the altitude ranges between the altitude profiles; central Alpine altitude profiles are located at a higher altitude range than those at the peripheral Alpine regions (Northern Alps). The low number for the regression of the above timberline sites (n ) 3) requests further studies. Altitudinal Trends and Seasonal Change of OCP in Needles. OXC was not detectable in most needle samples. No apparent campaign differences for the concentrations of R-HCH and HCE in these 1/2 year old needle samples were observed (SI Table S2-2). The log-transformed altitudes are significantly correlated with the needle wax content for Oct04 and May-06 needles. The trend for the Oct-05 campaign was similar to the Oct-04 campaign but not significant (SI Figure S5). It may be affected by the smaller sample size of the Oct-05 campaign. The altitude tendency of wax content in needles is related to the weather conditions especially the temperature. There was no apparent correlation between needle wax and OCP concentrations. Generally, OCP decrease along the altitude in needle samples (SI Figures S6-1 and S6-2). The OCP-altitude correlations are significant for R-HCH in Oct-04 and May-06 samples and for HCE in Oct-04 samples; the altitude trends are stronger for May-06 samples than Oct-04 and Oct-05 samples. These trends did not change significantly before and after wax adjustment of OCP. The concentrations in needles may reflect the equilibrium (17) between the uptake of OCP from air (4, 18) and the evaporation of OCP from needles. The dry and wet particle 2452
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 7, 2009
deposition (19) on the needles may increase OCP concentrations, but the deposited particles are also removed by rain or snow ‘wash-off’ (3, 5) from time to time. Thus, unlike SPMD samples, needle concentrations mainly reflect the temporary air concentrations of OCP, and the air concentrations depend on temperature, wind direction, and local soil concentration. As a consequence, at higher altitude sites with lower temperatures the air concentrations may be generally lower than at lower altitude sites at higher temperatures. The air mobility is the dominating factor, and only a small part of the variation can be explained by altitudinal change. Interestingly, R-HCH has an increasing tendency from west to east along Alps chains for Oct-04 samples (SI Figure S7); this may suggest an east to west LRAT. There is no detectable trend for HCE. In conclusion, needle data suggested that temporary OCP concentrations decrease along altitude in air, and the trends are seasonally different. The altitudinal and longitudinal trends of R-HCH suggest its multiple factorial transportation. Organic Materials Affect OCP Concentrations in Soil. At the equilibrium status, the soil-air partition coefficient can be expressed as Ksa ) Cs/Ca ) 0.411FØocKoa, where Cs is OCP concentration in soil and Ca is its concentration in air, Koa is the octanol-air partition coefficient, Øoc represents the fraction of organic carbon, and F represents the soil density (kg L-1) (2, 20-22). Thus, Øoc is one of the important factors that describe OCP behavior in soil. Probably because the soil sampling sites are all below the timberline with the same canopies (3, 8), no apparent organic material-altitude trend was found. The organic carbon-water correlation and the nitrogen-water correlation in the soil samples were nearly linear, and the two correlations are nearly parallel (SI Figure S11). The relationships between organic carbon and OCP and between nitrogen and OCP have similar tendencies (SI Table S3). Due to this similarity, the total organic material (TOM, here defined as the sum of the organic carbon and nitrogen content) was used to elaborate the correlations between OCP and TOM (Figure 2-1, -2, -3). TOM positively correlated with OCP concentrations for mineral soil samples and humus samples at lower range (
