1. Introduction
The rapid growth of industry in East Asia is associated with high anthropogenic emissions of pollutants. Pollutants from the Asian continent may be actively transported to the atmosphere over the Japan Sea (Reference Hatakeyama, Murano, Sakamaki, Mukai, Bandow and KomakaziHatakeyama and others, 2001; Reference Watanabe, Ishizaka and TakenakaWatanabe and others, 2001a). The air pollutants promote cloud and precipitation water acidification and affect the microphysical properties of the clouds (Reference Watanabe, Ishizaka and TakenakaWatanabe and others, 2001a). Recently, a large amount of acidic species (e.g. non-sea-salt sulfate) has been deposited in the Hokuriku District, along the Japan Sea coast in central Japan (Reference Honoki, Tsushima and HayakawaHonoki and others, 2001, Reference Honoki, Watanabe, Iida, Kawada and Hayakawa2007). In addition to pollutants, mineral Asian dust particles (Kosa in Japanese), rich in CaCO3, are also transported from the Asian continent (Reference TsurutaTsuruta, 1991; Reference Watanabe, Kasuga, Yamada and KawakamiWatanabe and others, 2006c). Kosa particles not only affect the radiation intensity in the atmosphere, but also contribute to neutralization of acid in water droplets (Reference Watanabe, Ishizaka and TakenakaWatanabe and others, 1999; Reference Watanabe and HonokiWatanabe and Honoki, 2003). The oxidative capacity in the atmosphere may also become high over East Asian countries, including Japan. Ozone (O3) concentrations in the background troposphere have significantly increased (Reference Akimoto, Nakane, Matsumoto and CalvertAkimoto and others, 1994). According to Reference TanimotoTanimoto (2009), an increase of about 1 ppb a–1 in the average concentration of O3 was observed from 1998 to 2006 over Japan, especially in the spring, when background O3 shows maximum concentrations in the middle to high latitudes (Reference Watanabe, Nojiri and KariyaWatanabe and others, 2005b). The increase in O3 may accelerate the formation of peroxides (hydrogen peroxide (H2O2) and organic hydroperoxides (ROOH)) which act as important oxidants of SO2 in the liquid phase (Reference WatanabeWatanabe and others, 2009).
Mount Tateyama (3015ma.s.l.), a part of the mountain range in the Hokuriku District (Japan Sea side in central Japan), is near the Japan Sea coast in central Japan, where a large amount of air pollution may be transported from Asia as well as from industrial regions in Japan. Recently, a serious decline of the forest in the vicinity of Mount Tateyama has been observed (Reference Kume, Numata, Watanabe, Honoki, Nakajima and IshidaKume and others, 2009). Observation of the atmospheric environment at Mount Tateyama has been typically performed in summer and autumn (Reference Watanabe, Honoki, Yoshihisa, Nishino and YanaseWatanabe and others, 2006b, Reference Watanabe2010a). Due to severe conditions, it is difficult to perform direct observation from late autumn to spring. The large amount of snow cover formed in spring at Mount Tateyama not only acts as a water resource in the Hokuriku District but also records the environmental signals at high altitude during the cold months (6 months, November–April). More than 5m of snow cover is formed every April near the summit of Mount Tateyama. Snowmelt runoff is caused after May, and snow cover usually becomes extinct in July.
Snow-pit observation at a high elevation is useful to evaluate air quality in the free troposphere during cold months. In a previous study at Mount Tateyama in the 1990s (Reference OsadaOsada and others, 2000, Reference Osada, Iida, Kido, Matsunaga and Iwasaka2004), the chemical composition in snow was measured. Reference Osada, Iida, Kido, Matsunaga and IwasakaOsada and others (2004) elucidated the formation processes of mineral dust layers in snow at Mount Tateyama. A snow-pit study was also conducted in the northern Japan Alps (Reference Shah, Tanaka, Kuramoto and SuzukiShah and others, 2008). However, there is a shortage of chemistry data in snow pits at high-elevation sites (>2000 m) in Japan where trans-boundary air pollution and Kosa particles are actively transported. Moreover, measurements of peroxides in snow in an alpine region have scarcely been performed, especially in Japan, while peroxides in polar snow have been investigated (Reference Bales, Losleben, McConnell, Fuhrer and NeftelBales and others, 1995; Reference Neftel, Bales, Jacob and DelmasNeftel and others, 1995; Reference Kamiyama, Motoyama, Fujii and WatanabeKamiyama and others, 1996; Reference McConnell, Winterle, Bales, Thompson and StewartMcConnell and others, 1997, 1998; Reference JacobiJacobi and others, 2002; Reference Hutterli, McConnell, Bales and StewartHutterli and others, 2003). Measurements of peroxides in an ice core may give information about the past atmospheric oxidation conditions (Reference Sigg and NeftelSigg and Neftel, 1988, Reference Sigg and Neftel1991; Reference Watanabe, Kamiyama, Watanabe and SatowWatanabe and others, 1998; Reference Gillett, van Ommen, Jackson and AyersGillett and others, 2000; Reference Frey, Bales and McConnellFrey and others, 2007). The peroxide concentrations in snow are highly affected by post-depositional modification (Reference Neftel, Bales, Jacob and DelmasNeftel and others, 1995; Reference McConnell, Winterle, Bales, Thompson and StewartMcConnell and others, 1997, 1998; Reference Hutterli, McConnell, Bales and StewartHutterli and others, 2003). At Mount Tateyama, environmental factors (e.g. temperature and impurities concentrations) are significantly different from those in polar regions; however, snow-pit observations may also be useful for the interpretation of ice-core analysis, especially for unstable species such as peroxides.
The aim of this study is to understand the behavior of snow chemistry at Mount Tateyama, especially the concentrations of acidic species, chemical stratigraphy and the relationship between the concentrations of peroxides and snow type or quality and ionic species. In this paper, the characteristics of the chemical composition, including peroxides, in the snow pits at Murododaira, near the summit of Mount Tateyama, are examined.
2. Methods
Figure 1 is a map of Japan showing the location of Mount Tateyama. The snow-pit observation was performed at Murododaira (36.6˚ N, 137.6˚ E; 2450ma.s.l.), near the summit of Mount Tateyama (Fig. 1), Toyama Prefecture, Japan, on 22–24 April 2005, 20–22 April 2006, 19–21 April 2007 and 18–20 April 2008. The snow cover at the sampling site from 2005 to 2008 was 6.4m (3.0mw.e.), 8.5m (4.0mw.e.), 5.6 m (2.5mw.e.) and 6.6m (3.1m w.e.) deep, respectively. Snow temperature in the pits was <0˚C; therefore, leaching of chemical constituents might have been suppressed, and ionic constituents might have been well preserved during the 6 months (November–April).
Fig. 1. Map of Japan showing the location of Mount Tateyama.
Snow samples were collected from the pit wall after observation of stratigraphy and measurement of density, and the sampling interval for chemical analysis was usually 10 cm. Unfortunately, sampling below depths of 4 and 5 m could not be done in 2005 and 2006, respectively. Snow samples could be taken from the top to the bottom in 2007 and 2008. Snow samples that had not melted were transported to Toyama Prefectural University and preserved in a freezer (–10˚C). Measurements of peroxides were performed within 1 week. The snow samples were melted just before the analysis of peroxides. Unfortunately, the peroxide concentrations could not be measured below 4 m depth in the 2006 pit. The pH of the melted solution was measured by an electrode using a pH meter. The dissolved ionic species were analyzed using an ion chromatograph (Yokogawa, IC7000).
Peroxide concentrations were measured by the fluorometric method (Reference Lazrus, Kok, Lind, Gitlin and McLarenLazrus and others, 1985) using p-hydroxyphenyl acetic acid and peroxidase reagents as described in detail by Reference Watanabe, Iwai, Takeda and TakebeWatanabe and others (2005c). The fluorescence reagents, 1 mL p-hydroxyphenyl acetic acid (1.5×10–2 M) in a phosphate buffer (adjusted to pH 7) and 1 mL peroxidase (50 unit mL–1) in a phosphate buffer (adjusted to pH 7), were added to the 1mL liquid sample. Several minutes later, 1 mL NaOH (1 N) was added to stabilize the generated fluorescence. The amount of fluorescence in the solution was determined using a fluorescence spectrophotometer (Hitachi Corporation, Model F-2500). This method determines the concentration of total peroxides (H2O2 + ROOH). The detection limit was 0.1 μmol kg–1. To calibrate, a standard H2O2 solution was made from a stock solution titrated with a standard KMnO4 solution. We also determined individual peroxides using HPLC separation (Reference Kok, McClaren and StaffelbachKok and others, 1995). The concentrations of organic hydroperoxides (e.g. methylhydroperoxide (MHP)) were not detected. Therefore, peroxides in the snow pits can be regarded as H2O2.
3. Results and Discussion
3.1. Major ions
A summary of the chemistry in the snow pits at Murododaira (2450ma.s.l.) in April from 2005 to 2008 is presented in Table 1. The levels of non-sea-salt sulfate (nssSO4 2–) and
Table 1. Summary of the chemistry in the snow pits at Murododaira (2450ma.s.l.), measured each April from 2005 to 2008. The units of the concentrations of peroxides and ions are μmol kg–1 and μEq kg–1, respectively. N and BDL denote the number of samples and a value below the detection limit of peroxides, <0.1 μmol kg–1, respectively
non-sea-salt calcium (nssCa2+) were calculated using
where (SO4 2–/Na+)sea water and (Ca2+/Na+)sea water are the concentration ratios of SO4 2– to Na+ and Ca2+ to Na+ in sea water, which are 0.12 and 0.044 (equivalent ratio), respectively (Reference Keene, Pszenny, Galloway and HawleyKeene and others, 1986). The ionic species can be classified into three types by their origin: anthropogenic pollution (NO3 –, nssSO4 2– and NH4 +), sea-salt particles (Na+, Cl– and Mg2+) and Kosa particles (nssCa2+) (Reference OsadaOsada and others, 2000). The concentrations of major ions were much higher than those in polar regions (Reference MotoyamaMotoyama and others, 2001; Reference Frey, Bales and McConnellFrey and others, 2007; Reference VirkkunenVirkkunen and others, 2007). The range in pH was 4.0–6.0. Acidic snow was deposited at Murododaira. Sulfuric and nitric acids contribute to the acidification of precipitation.
As mentioned above, the ionic constituents seemed to be well preserved in the snow pits during the 6 months. The mean concentrations of nssSO4 2– and NO3 – from 2005 to 2008 are 21–27 and 7–12 μEq kg–1, respectively (Table 1); they are higher than those in no-melt snowpack at Murododaira in the 1990s reported by Reference OsadaOsada and others (2000). On the other hand, no significant difference is seen between the concentrations of the other ions in the 2000s and those in the 1990s. Anthropogenic emission (e.g. SO2) in China has increased since 2000 (Reference OharaOhara and others, 2007). According to Reference Osada, Ohhara, Uno, Kido and IidaOsada and others (2009), significant increasing trends of submicrometer aerosols, which might have been mainly contributed by Chinese-derived anthropogenic emission, at Mount Tateyama were found during winter to spring in the 2000s. The nssSO4 2– and NO3 – deposition also seemed to increase at Murododaira during the cold months. We estimated the sub-annual (6 months, November–April) deposition of nssSO4 2– and NO3 – at Murododaira to be 66– 84 mEqm–2 (6 months)–1 and 27–30 mEqm–2 (6 months)–1, respectively. The calculated deposition fluxes are significantly higher than those at low altitude on the Japan Sea side shown in Reference NoguchiNoguchi and others (2007).
In this study, the snow-pit observation was performed before the melt season (usually after May at Murododaira). Snow temperature in the pits was <0˚C. The pit walls were mainly composed of compacted snow (fine grain, non-melt rounded grains) or solid-type depth hoar (faceted crystals); however, granular snow (coarse grain, melt forms) layers were sporadically seen in the snow pits. The relationship between Na+ and Mg2+ concentrations in the snow pits at Murododaira in April from 2005 to 2008 is shown in Figure 2. Most of the samples are similar to the Mg2+/Na+ ratio in sea water (or higher than the ratio in sea water; the higher Mg2+ is due to dust particles). A low ratio of Mg2+/ Na+ (or a high ratio of log(Na+/Mg2+)) is evidence of deep percolation or runoff of ions (Reference Iizuka, Igarashi, Watanabe, Kamiyama and WatanabeIizuka and others, 2000, Reference Iizuka, Igarashi, Kamiyama, Motoyama and Watanabe2002; Reference VirkkunenVirkkunen and others, 2007). High values of the melt indicator log(Na+/Mg2+) have been observed in Svalbard (Reference VirkkunenVirkkunen and others, 2007). A significantly low Mg2+/Na+ ratio was also detected in the snowpack at Murododaira during melting (personal communication from K. Watanabe, 2010); however, a low ratio of Mg2+/Na+ was not seen in the pits from 2005 to 2008 (Fig. 2). Redistribution of ions by meltwater percolation is not significant in this study.
Fig. 2. Relationship between Na+ and Mg2+ concentrations in the snow pits at Murododaira from 2005 to 2008. SW indicates the ratio in the case of sea water.
Figure 3 shows the stratigraphy and the vertical profiles of the pH, Na+, nssSO4 2–, NOnssCa2+ and peroxide concentrations in the snow pits at Murododaira in April 2005 (from the top to 4 m depth) and 2006 (from the top to 5 m depth). The concentrations of nssCa2+ were high in the layers higher than 2–3m depth (Fig. 3). The high concentrations of nssCa2+ are due to Kosa particles, rich in CaCO3, which are actively transported over Japan in spring (Reference Watanabe, Suzuki and DokiyaWatanabe and others, 2005a, Reference Watanabe, Takebe, Sode, Igarashi, Takahashi and Dokiya2006c), and correspond to the dust layers. There is heavy snow cover near the sampling site in winter and spring; therefore, local soil dust is not blown up. Snow layers higher than 2–3m depth may be deposited during the spring months. Layers lower than 2–3m depth may be regarded as sediment in the winter, and the bottom of the snow pits seems to be deposited during late autumn. Kosa particles contribute to the neutralization of acidic species; therefore, the pH in the snow was also high in the parts higher than 3m. No vertical trend of Na+, nssSO4 2– and NO3 – was seen. Sea-salt particles and air pollutants may be transported in both the winter and spring. High concentrations of Na+, which might have been due to the transport of abundant sea-salt particles, were detected at 1 and 3 m depth in the 2005 pit and 4.2 m depth in the 2006 pit (Fig. 3). During the strong monsoon wind, convective activity over the Japan Sea side is sometimes accelerated, and sea-salt particles are transported vertically. High concentrations of nssSO4 2– or NO3 – are due to the transport of air pollution. The high-nssSO4 2– layers deposited in the winter (e.g. at 3.7 m depth in the 2005 pit and ~4m depth in the 2006 pit) correspond to the low-pH layers. Long- or middle-range transport of pollutants may accelerate the acidification of precipitation in winter.
Fig. 3. Stratigraphy (/: lightly compacted snow (decomposing and fragmented precipitation particles); filled circles: compacted snow (fine grain, rounded grains); open circles: granular snow (coarse grain, melt forms); squares: solid-type depth hoar (faceted crystals)) and vertical profiles of pH, Na+, nssSO4 2–, NO3 – (dotted line), nssCa2+ and peroxide concentrations in the snow pits at Murododaira in April 2005 (upper panel) and 2006 (lower panel). Peroxide concentrations could not be measured below 4 m depth in the 2006 pit.
Figure 4 shows the stratigraphy and the vertical profiles of nssSO4 2–, nssCa2+ and peroxide concentrations in the snow pits from the top to the bottom at Murododaira in April 2007 and 2008. The pH, Na+ and NO3 – were not shown in Figure 4. High concentrations of nssCa2+ were detected above 1m in the 2007 pit; however, the relatively high peaks of nssCa2+ were also observed below 3 m depth in the 2008 pit deposited in the winter. Judging from the results of the pit observation, transport of Kosa particles during winter might have been more active in 2008 than in the other years. No vertical trend of nssSO4 2– was seen in the 2007 and 2008 pits.
Fig. 4. Stratigraphy (+: new snow (precipitation particles); /: lightly compacted snow (decomposing and fragmented precipitation particles); filled circles: compacted snow (fine grain, rounded grains); open circles: granular snow (coarse grain, melt forms); squares: solid-type depth hoar (faceted crystals)) and vertical profiles of nssSO4 2–, nssCa2+ and peroxide concentrations in the snow pits from the top to the bottom at Murododaira in April 2007 (upper panel) and 2008 (lower panel).
The peaks of nssCa2+ in the snow pits usually correspond to the layers containing high concentrations of nssSO4 2– (Fig. 3 and 4). The results show that not only Kosa particles but also air pollutants might have been transported long-range from the Asian continent. A typical example (21 April 2005) of 5 day backward trajectories by the Hybrid Single-Particle Lagrangian Integrated Trajectory 4 model (US National Oceanic and Atmospheric Administration (NOAA) Air Resources Laboratory (ARL), Silver Spring, MD, 1997, available at http://www.arl.noaa.gov/ready/hysplit4.html) is illustrated in Figure 5. The starting heights for the calculations were 2000, 2500 and 3000 m. High concentrations of nssSO4 2– and nssCa2+ were recorded in the surface layer deposited on 21 April 2005 (snowfall was observed) before the snow-pit observation of 2005 (Fig. 3a). The air mass originating in the arid regions of the Asian continent was transported to Mount Tateyama through the polluted areas of the continent, especially from the coasts of the Yellow Sea, where SO2 emission is large (Reference OharaOhara and others, 2007). To elucidate the transport processes of Kosa particles and air pollution requires more analysis; however, precise dating of the snow layers, which is needed for trajectory analysis, is unavailable. In this study, the typical case on 21 April 2005 (dating is reliable) is shown.
3.2. Peroxides
The concentrations of peroxides in the snow pits ranged from below the level of detection to 3.1 μmol kg–1. There are comparable peroxide data in the snow pits in polar regions. The peroxide concentrations at Murododaira were lower than those in snow samples from the pit wall at NorthGRIP, Greenland (Reference MotoyamaMotoyama and others, 2001). Reference Sigg, Staffelbach and NeftelSigg and others (1992) reported that a high concentration of H2O2 in the air was observed during summer at Summit, Greenland, and the H2O2 concentration in fresh snow samples was ~20 μmol kg–1. High concentrations of H2O2 were detected in the snow layers deposited during summer and autumn in Greenland (Reference MotoyamaMotoyama and others, 2001). The peroxide concentrations at Murododaira were similar to those in the snow pits in Svalbard (Reference MotoyamaMotoyama and others, 2001). The H2O2 in Svalbard might have been flushed out by the percolation of snowmelt water after accumulation.
Fig. 5. Results of 5 day backward trajectory analysis; the starting time is 12 h Japan Standard Time (JST) (3 h UTC) on 21 April 2005. AGL: above ground level.
The peroxide concentrations in new snow (precipitation particles) were higher than those in compacted snow (fine grain, rounded grains) or lightly compacted snow (decomposing and fragmented precipitation particles). During the 2007 pit study, the highest concentration was detected in the surface new snow layer (Fig. 4a). New snow layers were not seen in the pits during the 2005, 2006 and 2008 observations (Fig. 3 and 4); fortunately, snowfall samples (new snow samples) could be taken at Murododaira just after the pit observations in 2006 and 2008. Table 2 shows the range of the concentrations of peroxides, major ions and pH in the new snow (precipitation particles) samples at Murododaira in 2006 and 2008. The new snow samples contained high concentrations of peroxides. The concentrations of peroxides were 4.4–5.5 μmol kg–1 in April 2006 and 2.0–7.1 μmol kg–1 in April 2008 (Table 2).
Table 2. Summary of the chemistry in the new snow (precipitation particles) at Murododaira (2450ma.s.l.) in April 2006 and 2008. The units of the concentrations of peroxides and ions are μmol kg–1 and μEq kg–1, respectively. N denotes the number of samples
According to Reference Sigg, Staffelbach and NeftelSigg and others (1992), the diffusional growth of snow is an important process of H2O2 scavenging. There is no fractionation between H2O2 and H2O during diffusional growth (co-condensation). Assuming that co-condensation could be applied, we roughly estimated the concentration of H2O2 in the atmosphere before snowfall events by the mean peroxide concentration in the new snow samples (C = 4.5 μmol kg–1), air pressure (P = 7.45×104 Pa), ambient air temperature (T = 268 K), and water vapor content (W= 3.0 gm–3) at Murododaira as follows:
where (H2O2)air is the H2O2 concentration in the ambient air (ppbv) and R is the gas constant (8.31 J mol–1 K–1). The calculated H2O2 concentration is ~0.4 ppbv and corresponded well to the H2O2 concentration over the Hokuriku District during the cold months (Reference Watanabe, Eda and AokiWatanabe and others, 2010b). The concentration is also comparable with the H2O2 mixing ratio at Summit, Greenland, in spring (Reference FreyFrey and others, 2009).
The concentrations of peroxides in the upper compacted and lightly compacted snow layers (<1.0 μmol kg–1) were significantly lower than those in new snow samples (Fig. 3 and 4). Moreover, the peroxide concentrations were very low in the compacted snow below 4m depth, deposited mainly in winter; for example, the mean ratio of the peroxide concentration above and below 4m depth to that in the new snow layer in the 2007 pit was ~0.1 and ~0.04, respectively. The results indicate that peroxides are decomposed and degassed in the process of metamorphism of new snow to compacted snow and reconstruction of atmospheric peroxide concentrations is difficult. Air-to-snow transfer processes of H2O2 in polar snow have been discussed (Reference McConnell, Winterle, Bales, Thompson and StewartMcConnell and others, 1997, Reference McConnell, Bales, Stewart, Thompson, Albert and Ramos1998; Reference Hutterli, McConnell, Bales and StewartHutterli and others, 2003; Frey and others, 2007). H2O2 uptake in near-surface snow and firn has been observed, particularly at low temperature (Frey and others, 2006). H2O2 adsorption may be suppressed (H2O2 desorption may be predominant) at Murododaira because of the relatively high temperature. Moreover, photolysis of peroxides in surface snow may be caused by intensive ultraviolet radiation, whereas the photolysis loss of H2O2 seems to be small (Reference GrannasGrannas and others, 2007). According to Reference France, King and Lee-TaylorFrance and others (2007), H2O2 photolysis in snowpack is an important source of hydroxyl radicals in the atmosphere.
The peroxide concentrations in the snow pits were much lower than those in rainwater during warm months (Reference Watanabe, Ishizaka, Minami and YoshidaWatanabe and others, 2001b, Reference Watanabe, Nojiri and Kariya2005b, Reference Watanabe, Takebe, Sode, Igarashi, Takahashi and Dokiya2006a). It is natural for the photochemical production of peroxides to accelerate in warm months (Reference Sakugawa and KaplanSakugawa and Kaplan, 1989; Reference Watanabe and TanakaWatanabe and Tanaka, 1995). Peroxide concentrations are higher in rainwater than in snowfall in cold months (Reference WatanabeWatanabe and others, 2009). We simultaneously collected rainwater at Toyama Prefectural University (at low altitude in Toyama Prefecture) and snowfall (new snow) at Murododaira on 20 April 2003. The concentrations of peroxides were 1.0– 2.0 μmol kg–1 in the snow but 22 μmol kg–1 in the rainwater (Reference Watanabe, Iwai, Takeda and TakebeWatanabe and others, 2005c). The scavenging processes of H2O2, the main component of peroxides, from the atmosphere are different in rainwater and snowfall. While the diffusional growth of snow is an important process of H2O2 scavenging (c-condensation), H2O2 may be preferentially absorbed by rain droplets because of the high solubility of H2O2, for which Henry’s law constant is high. Henry’s law constant is given by Reference Lind, Kok, Lind and KokLind and Kok (1994) as follows:
where H(T) is Henry’s law constant (M atm–1), T is the ambient temperature (K), A = 6338 and B = 9.74. The concentration of H2O2 in the gas phase during the rain at low altitude on 20 April 2003 was roughly calculated using Henry’s law at ~0.1 ppbv. To estimate the H2O2 concentration in the atmosphere before the rain event, rainfall intensity data which could not be measured are required.
High concentrations of peroxides in the snow pits were also detected in the granular snow (coarse grain, melt forms) layers (e.g. at 0.9m depth in the 2005 pit, 2 and 3 m depth in the 2006 pit, 2 m depth in the 2007 pit, and 2.5 m depth in the 2008 pit (Fig. 3 and 4)). Granular snow (coarse grain, melt forms) is formed by percolation of rainwater or snowmelt water, which dissolves peroxides. Surface snow-melt water as well as rainwater seems to contain high concentrations of peroxides. As a result, relatively high concentrations of peroxides would be preserved in granular snow layers. The concentrations of peroxides were usually low below 4m depth in snow pits. However, the peroxide peak was detected at 4.4m depth in the 2008 pit, which corresponds to the granular snow layer (Fig. 4b). Recently, rain events were observed at Murododaira even in early spring and winter. Unfortunately, we cannot perform quantitative analysis because rainfall intensity data are not available at Murododaira. It is possible that warmer temperatures, which cause more rainfall events during cold months, may increase the peroxide concentrations recorded in snow cover in an alpine region, while the H2O2 adsorption by the air-to-snow transfer processes may be suppressed due to an increase in temperature.
A negative correlation was seen between the peroxides and the nssCa2+ (Fig. 3 and 4). High concentrations of peroxides were detected in the layers in which nssCa2+ concentrations were low. Peroxides were hardly detected, particularly in the dust layers, where high concentrations of nssCa2+ were seen. This anticorrelation has been reported in ice cores. According to Reference Fuhrer, Neftel, Anklin and MaggiFuhrer and others (1993), H2O2 was preserved in sections of ice from Greenland having low Ca2+. The high concentrations of nssCa2+ are due to the deposition of soil particles (Kosa particles), which contain abundant metal components. The metal components that act as catalysts might have decomposed the peroxides in snow. High peroxide concentrations may be preserved in granular snow layers having low concentrations of nssCa2+. The peroxide concentrations also anticorrelated with the nssSO4 2– concentrations (Fig. 3 and 4). It is possible that peroxides were also consumed by the liquid-phase oxidation of SO2. Detailed observation of stratigraphy and measurements of ionic species or trace metals may be important to interpret the past behavior of peroxides by ice-core analysis.
4. Summary
Snow cover at high elevations in Japan recorded atmospheric environmental signals during the winter and spring. Mount Tateyama is located near the coast of the Japan Sea, where air pollution and Asian dust (Kosa) particles are actively transported from the Asian continent. Snow-pit observation and sampling of snow in the pits at Murododaira (2450ma.s.l.) near the summit of Mount Tateyama were performed each April from 2005 to 2008. Chemical analysis, such as pH, major-ions and peroxide measurements of the snow samples, was conducted. The mean concentrations of nssSO4 2– and NO3 – in the snow pits are higher than those in snowpack at Murododaira in the 1990s reported by Reference OsadaOsada and others (2000).
High concentrations of non-sea-salt sulfate (nssSO4 2–), which might have been due to trans-boundary air pollution, were measured in both the winter snow layers and the spring layers. The pH of snow samples was usually higher in the layers deposited in the spring than in the winter layers. High concentrations of nssCa2+ corresponding to dust layers that might have been derived from Kosa particles were detected mainly in the spring layers. Kosa particles, rich in CaCO3, contributed to the neutralization of acid. The Kosa layers usually contained high concentrations of sulfate. The results indicate that Kosa particles might have been transported with air pollution from the continent.
The concentrations of peroxide were high in the new snow (precipitation particles) and granular snow (coarse grain, melt forms) layers in the snow pits. The peroxide concentrations in the snow layers were negatively correlated with the nssCa2+ concentrations. Peroxides were hardly detected, especially in the dust layers. Soil particles seemed to decompose the peroxides in snow. High peroxide concentrations may be preserved in granular snow layers having low concentrations of nssCa2+. Reconstruction of atmospheric peroxides in an alpine region during cold months from a snow-pit study is difficult. Detailed observation of stratigraphy and measurements of ionic species or trace metals may be important to interpret the past behavior of peroxides by ice-core analysis. More quantitative analysis is required.
Acknowledgements
We thank K. Satow, N. Wada, A. Kume, K. Aoki, W. Shimada, H. Honoki, I. Suzuki, H. Iida, N. Ikeda, A. Tomatsu, K. Noritake, K. Yamada, N. Miyashita, M. Ohata, A. Takahashi, H. Kawamura, T. Murakami and F. Endo, and members of Toyama University, Toyama Prefectural University, Kanazawa University, Nagoya University, Tateyama Caldera Sabo Museum, Toyama Science Museum, Murodo-sanso, Tateyama Kurobe Kanko, Inc. and Toyama Prefecture Road Public Corporation, for their support and coordination of the snow-pit study at Mount Tateyama. This work was supported in part by Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology (Nos. 16710011, 18310022, 20310113 and 22310022).
