Seasonal and diel changes in photosynthetic activity of the snow alga Chlamydomonas nivalis (Chlorophyceae) from Svalbard determined by pulse amplitude modulation fluorometry

Authors

  • Marek Stibal,

    1. Faculty of Biological Sciences, University of South Bohemia, České Budějovice, Czechia
    2. Institute of Botany, Czech Academy of Sciences, Třeboň, Czechia
    Search for more papers by this author
  • Josef Elster,

    1. Faculty of Biological Sciences, University of South Bohemia, České Budějovice, Czechia
    2. Institute of Botany, Czech Academy of Sciences, Třeboň, Czechia
    Search for more papers by this author
  • Marie Šabacká,

    1. Faculty of Biological Sciences, University of South Bohemia, České Budějovice, Czechia
    2. Institute of Botany, Czech Academy of Sciences, Třeboň, Czechia
    Search for more papers by this author
  • Klára Kaštovská

    1. Institute of Botany, Czech Academy of Sciences, Třeboň, Czechia
    2. Centre of Biology, Institute of Hydrobiology, Czech Academy of Sciences, České Budějovice, Czechia
    Search for more papers by this author

  • Editor: Rosa Margesin

Correspondence: Marek Stibal, Bristol Glaciology Centre, School of Geographical Sciences, University of Bristol, University Road, Bristol, BS8 1SS, UK. Tel.: +44 0 117 331 7313; fax: +44 0 117 928 7878; e-mail: marek.stibal@bristol.ac.uk

Abstract

The seasonal and diel dynamics of the physiological state and photosynthetic activity of the snow alga Chlamydomonas nivalis were investigated in a snowfield in Svalbard. The snow surface represents an environment with very high irradiation intensities along with stable low temperatures close to freezing point. Photosynthetic activity was measured using pulse amplitude modulation fluorometry. Three types of cell (green biflagellate vegetative cells, orange spores clustered by means of mucilaginous sheaths, and purple spores with thick cell walls) were found, all of them photosynthetically active. The pH of snow ranged between 5.0 and 7.5, and the conductivity ranged between 5 and 75 μS cm−1. The temperature of snow was stable (−0.1 to +0.1°C), and the incident radiation values ranged from 11 to 1500 μmol photons m−2 s−1. The photosynthetic activity had seasonal and diel dynamics. The Fv/Fm values ranged between 0.4 and 0.7, and generally declined over the course of the season. A dynamic response of Fv/Fm to the irradiance was recorded. According to the saturating photon fluence values Ek, the algae may have obtained saturating light as deep as 3 cm in the snow when there were higher-light conditions, whereas they were undersaturated at prevalent low light even if on the surface.

Introduction

Large areas in polar and alpine regions are seasonally or perennially covered in snow, and may therefore appear abiotic at first glance. However, well-developed microbial assemblages have been found in various snow environments around the world (Hoham & Duval, 2001, and references therein). Snow algal populations living within liquid water retained among snow crystals are the most prominent microorganisms living in such cold habitats.

The snow surface is a typical low-temperature and high-irradiation environment: the temperature is rather stable around the freezing point, and the intensities of light may reach very high values (Gorton et al., 2001). Snow algae possess some specific ecological and physiological adaptations to the harsh environment of the snow. Most of the ‘true’ snow algae, defined as those that grow and reproduce entirely within the water during snowmelt, are green algae of the order Chlamydomonadales (Chlorophyceae). Chlamydomonas nivalis (Bauer) Wille is the most common inhabitant of snowfields in Svalbard (Kol & Eurola, 1974; Newton, 1982; Müller et al., 1998, 2001), causing the well-known red colouration of snow. Its complex life cycle comprises vegetative motile cells with flagella, and resting stages that drift passively as the snow is melting, and that may be clustered together with mucilaginous sheaths (Kawecka & Drake, 1978; Kawecka, 1981). Screening pigments can be produced in the cells in order to provide protection against excessive light (Bidigare et al., 1993; Gorton et al., 2001). Snow algae may play an important role as primary producers, and so comprise the base of the food web in snow (Hoham & Duval, 2001). They can also be a considerable source of organic matter for downstream ecosystems.

Little attention has been paid to photosynthesis in snow to date. The method of labelled carbon uptake was used for the measurements of CO2 fixed during photosynthesis in snow algal populations in North American mountains (Thomas, 1972; Mosser et al., 1977) and Slovakian mountains (Komárek et al., 1973). Williams et al. (2003) measured the gas exchange between a snowpack containing Chlamydomonas nivalis and the atmosphere, and CO2 fixation or release from snow, in the Rocky Mountains. Remias et al. (2005) used oxygen evolution measurements for photosynthesis determinations of Chlamydomonas nivalis in the Austrian Alps. However, there is still a lack of information from the polar environment, where snow and its inhabitants play a much greater role in the whole ecosystem. Also, the seasonal dynamics of snow algal populations in terms of their life cycle progression, growth and physiology are largely unknown.

Pulse amplitude modulation (PAM) fluorometry has been successfully used for measurements of the physiological state of cells and their photosynthetic efficiency in other cold environments, such as Antarctic fast ice (Kühl et al., 2001; McMinn et al., 2003) or marine benthic communities (Schwarz et al., 2003; McMinn et al., 2004). It has recently proven useful for snow algal photosynthesis assessments in a pilot study in temperate mountains (Kvíderováet al., 2005). It gives information about the state of photosystem II (PSII). The flow of electrons through PSII indicates the overall rate of photosynthesis, and so allows estimation of the performance of photosynthesis. Also, damage to PSII can be the first sign of stress in the cell. However, the fluorescence can only be used for measurements of photosynthesis efficiency, and the actual rate of CO2 fixation is very difficult to measure in the field, due to changing environmental and physiological factors such as CO2 concentrations, photorespiration, and nitrogen metabolism (Maxwell & Johnson, 2000). Nevertheless, it is a very useful and simple method for measurements of the daily courses of photosynthetic potentials, or for comparing the differences in these in one population in changing conditions throughout the entire growing season.

The main goal of this study was to monitor a snow algal population dominated by Chlamydomonas nivalis, and the physiological state and photosynthetic activity of this organism in an Arctic snowfield in Svalbard. The field site has been described in detail previously (Stibal & Elster, 2005), as have the surrounding soil and freshwater habitats (Kaštovskáet al., 2007; Stibal et al., 2006). This aim of this study was to gain a deeper insight into the functioning of cold-active microbial populations.

Materials and methods

Study area and sampling

The fieldwork was undertaken at a snowfield situated by the snout of Werenskioldbreen, a valley glacier in southwest Svalbard (77°04.428′N; 15°14.943′E) at an altitude of c. 100 m above sea level on a steep slope with a west–southwest ascent. The studied snowfield had been known to last for the entire summer season and to harbour snow algal populations (Stibal & Elster, 2005). Samples were collected throughout the summer season 2004, when liquid water percolated through the snow and promoted the growth of the snow algal population (July through August).

Snow samples were collected at daily intervals if possible. The snow was allowed to melt at c. 5°C. This temperature was a compromise between a higher temperature, which could have caused more intensive collapse of algal cells, and a naturally low temperature around 0°C, which would have meant very long melting times. The cell densities and the proportions of the life-forms (green flagellate cells, orange spores, and purple spores) were determined using a Cyrus-I counting chamber and a Leitz HM-Lux field microscope. The size of the cells was measured and the biovolume calculated, assuming the cells to be spherical. There was little debris in the snow, so the cell densities could be determined immediately without previous filtration. Owing to rather high cell densities, there was no need for centrifugation. The measurements were reasonably accurate despite some cell collapses, as the collapsed cells can also be easily determined and counted. Electrical conductivity and pH were measured with a Testo 252 multimeter with pH and conductivity probes (Testo, Lenzkirch, Germany). Incident radiation was measured with a custom-built quantum meter PU-550 (Institute of Botany CAS, Třeboň, Czechia) with a flat sensor, which was especially designed for field measurements in severe conditions, being capable of enduring strong mechanical stress. This sensor was precalibrated and postcalibrated to a quantum meter QSL-101 (Biospherical Instruments, San Diego, CA) with a spherical sensor to simulate the all-sides absorption of the irradiance, resulting in the following relationship:

image

where Ei is incident radiation measured by our flat sensor, and E0 measured by the QSL-101 spherical sensor is considered to be roughly equal to photon fluence reaching algae on the snow surface. The light penetration through the snow to a depth of 20 cm was determined for different solar radiation conditions. To conduct these measurements, the sensor of the quantum meter was immersed from the side in a dug snow profile, so that the amount of light coming from the side was negligible. This was repeated at depths of 1, 3, 5, 10 and 20 cm, and also on the snow surface. On the basis of the acquired curves, the light attenuation coefficients were obtained, according to the following equation:

image

where E0 is the photon fluence at the snow surface, ED is the photon fluence reaching a snow algal cell at depth D, and a is the attenuation coefficient.

Chlorophyll fluorescence measurements

Chlorophyll fluorescence was measured with a portable double-modulated PAM fluorometer FL-3300 coupled to fluorwin software (Photon System Instruments, Brno, Czechia). PAM fluorometry is based on the supply of weak, modulated light pulses (the measuring light), which allow chlorophyll fluorescence to be monitored without inducing photosynthesis. In the dark-adapted state, a minimum fluorescence (F0) is determined when the measuring light is turned on. Maximum fluorescence Fm is achieved by exposing the dark-adapted sample to a pulse of very intense light. If the sample is experiencing additional actinic light, a similar, saturating pulse will lead to a lower maximum fluorescence Fm, which is the maximal fluorescence yield of an illuminated sample. This fluorescence then declines until an equilibrium level F is attained. This method is capable of measuring the quantum yield ΦPSII, defined as

image

The maximum effective quantum yield or photosynthetic efficiency Fv/Fm is defined as:

image

The relative electron transport rate (rETR) of PSII during photosynthesis is defined as:

image

where E is the photosynthetic photon flux density in μmol photons m−2 s−1 (Maxwell & Johnson, 2000). With the use of rapid light curves (RLCs), the photoadaptive state of the algal population can be estimated, and maximum relative electron transport rate rETRmax and saturating photon fluence Ek can be measured.

The fluorometric measurements were carried out from 22 July to 12 August 2004 at daily intervals if possible. Sometimes, measurements could not be performed, due to severe weather conditions. Moreover, two 24-h measurements were conducted on 31 July and 12 August. These days had different solar radiation conditions. The measurements were made every 2 h from midnight to midnight. Aliquots of snow samples (5 mL) were allowed to melt in the dark for at least 30 min, which also provided the dark adaptation, and then placed in a cuvette and measured at c. 5°C. During the measurements, the samples were well mixed in order to avoid sedimentation of algal cells. The measurements were made in duplicate for each sample, and a sample of melted snow without any algal cells was measured as a blank every time. The RLCs were performed with 12 photon fluence values (0–900 μmol photons m−2 s−1) with a 30-s exposure of the sample to each photon fluence. The data obtained were fitted using the function from Eilers & Peeters (1988):

image

where rETR is the relative electron transport rate, and E is the photosynthetic photon flux density. The parameters of the P–E curves, α (light-limited photosynthetic efficiency), rETRmax (maximum relative electron transport rate) and Ek (saturating photon fluence) were derived from the constants a, b and c using the following equations (Eilers & Peeters, 1988):

image

On the basis of the values of Ek and the light attenuation coefficients, the snow depth at which the algae were light-saturated was determined according to the following equation:

image

where Dk is the saturation depth, and a the light attenuation coefficient. Different attenuation coefficients were used for different weather conditions. sigmaplot (SPSS, Chicago, IL) was used for performing the fittings and calculations of all the photosynthetic parameters.

Results

Physicochemical properties of the snow

In the summer season of 2004, the Werenskiold snowfield showed a patchy orange–red or pink–purple colouration, which marked the presence of snow algal spores. No apparent stratification was observed in the snow structure. The pH and conductivity values are given in Table 1, in which the samples are sorted according to macroscopic appearance (colour), and thus dominant life-stage. The temperature of the snow ranged between −0.1 and +0.1°C for the entire sampling period; this is not shown in Table 1. The mean air temperature was 5°C, with minimum and maximum values of 2 and 12°C, respectively. Ninety millimetres of precipitation fell in the form of rain and drizzle during the sampling period. The values of incident radiation ranged between 11 and 1500 μmol photons m−2 s−1, and were dependent on the weather conditions. In the late summer, they also changed with the time of day, because the sun descended close to the horizon. The penetration of light into the snow column in different weather conditions is shown in Fig. 1. It is evident that although the snow surface may be exposed to high intensities of incident radiation, 1 cm below the surface the irradiation can drop to less than 50% of the surface value on days with overcast or foggy weather, or even to c. 25% on cloudless days. Over the experimental period, clear skies with a photon flux density exceeding 1000 μmol photons m−2 s−1 occurred on only 3 days (7, 11 and 12 August); otherwise, overcast with frequent dense fogs prevailed.

Table 1.   Ecological parameters in snow samples from Werenskiold snowfield containing different stages of Chlamydomonas nivalis (mean ± SD)
 nIncident radiation,
Ei (μmol photons
m−2 s−1)
pHElectrical
conductivity
(μS cm−1)
Cell density of
flagellated
vegetative cells, cV
(103 cells mL−1)
Cell density of
orange spores,
cO (103 cells mL−1)
Cell density of
purple spores,
cP (103 cells mL−1)
Total cell
density, cTOTAL
(103 cells mL−1)
Vegetative cells5246 ± 1027.55 ± 0.0537 ± 3224 ± 2070 ± 00 ± 0224 ± 207
Orange spores85223 ± 1666.68 ± 0.8029 ± 224 ± 30591 ± 4445 ± 17600 ± 455
Purple spores5303 ± 447.56 ± 0.1311 ± 30 ± 098 ± 57102 ± 36200 ± 36
Figure 1.

 Penetration of light into snow column of the Werenskiold snowfield in different weather conditions; a denotes the light attenuation coefficient.

Chlamydomonas nivalis stages in the snow

The period of the measurements covered almost the entire growing season with liquid water present in the snow. Different stages of the life cycle of Chlamydomonas nivalis were found in the studied snowfield, mostly in the upper layer of snow (0–5 cm) including the surface. Figure 2 shows the abundances of all life-cycle stages of Chlamydomonas nivalis over the course of the sampling period. At the beginning (22–28 July), green oval-shaped biflagellate vegetative cells were often found. They were capable of active motion. Sometimes, these cells caused a slight green colouration of the snow. Copulations of these cells and the formation of quadriflagellate planozygotes were observed, and sometimes these zygotes dropped their flagella and turned into round-shaped zygotes or spores. However, most flagellated cells seemed to be very susceptible to the changed environment, and possibly the combination of a higher temperature and bright light on a microscopic slide caused their rapid collapse. The highest concentration of these vegetative cells was more than 5 × 105 cells mL−1 of melted snow. The size of the cells was 5–10 × 7–15 μm. The combination of their active motion and rapid collapse made it very difficult to determine their size, and so the number should be taken as indicative only.

Figure 2.

 Abundances of green vegetative cells, orange spores and purple spores in the Werenskiold snowfield over the course of summer 2004.

By far the most abundant stage throughout the entire summer season comprised the orange–red round-shaped spores, most likely formed from the biflagellate vegetative cells or quadriflagellate planozygotes by casting off of the flagella and thickening of the cell walls, hereafter referred to as ‘orange spores’. They were often clustered by means of mucilaginous sheaths surrounding the cells. The diameter of the cells was 10.17±2.28 μm (mean±SD), and their mean biovolume was 550 μm3.

Other spores found in the snow, probably formed from the orange spores, were large, spherical cells, with an intense purple–red colour; these had thick cell walls, and were almost always solitary, with a diameter of 28.25±4.88 μm, and a mean biovolume of 11 804 μm3. They will be referred to hereafter as ‘purple spores’. The biovolume ratio of the orange spores and purple spores was c. 21. The orange spores and purple spores were clearly distinguishable, both microscopically and macroscopically. Orange–red snow patches contained orange spores, sometimes with a low number of purple spores, whereas the pink–purple colour of snow occurring mostly in the lower parts of the snowfield indicated almost pure purple spore ‘populations’. No purple spores were found after 31 July, when the last remnants of the lower parts of the snowfield melted away.

As well as Chlamydomonas nivalis, other species were found in the snow samples: the cell density of Chloromonas nivalis reached more than 4 × 104 cells mL−1 on some occasions, and some cells of Raphidonema spp. were also found. The mean concentrations of different stages of Chlamydomonas nivalis in the samples sorted according to the dominant life stage from summer 2004 are shown in Table 1. It can be seen that the orange spores were present in all the samples, and the purple spores represented a minor component in the samples of ‘orange spores’.

Photosynthetic parameters

The photosynthetic parameters of all the cell types measured during summer 2004 are shown in Table 2. Whereas in the orange spores the photosynthetic efficiency and the relative electron transport were somewhat lower than in the vegetative cells, the purple spores showed extremely high values of all photosynthetic parameters. Their saturating photon fluence was an order of magnitude higher than in the former two types of cell, and reached over 3700 μmol photons m−2 s−1. Figure 3 shows how the photosynthetic parameters changed over the course of the season in a single spot dominated by orange spores.

Table 2.   Photosynthetic parameters in snow samples from Werenskiold snowfield containing different stages of Chlamydomonas nivalis (mean ± SD)
 nFv/Fm (relative units)Saturating photon fluence,
Ek (μmol photons m−2 s−1)
α (relative units)rETRmax (relative units)
  1. Values for purple spores are given in parentheses because of their possible lack of validity.

Vegetative cells50.67 ± 0.19826 ± 6220.66 ± 0.21560 ± 497
Orange spores850.54 ± 0.07523 ± 6440.47 ± 0.08274 ± 389
Purple spores5(0.90 ± 0.04)(4038 ± 2746)(0.91 ± 0.05)(3736 ± 2653)
Figure 3.

 Dynamics of the photosynthetic parameters of Chlamydomonas nivalis over the course of summer 2004.

A similar pattern can be seen in the course of all the measured parameters. There was an initial peak in all the parameters on 24 July, 2 days after the commencement of the fluorometric measurements, when Fv/Fm values reached over 0.65, and rETRmax was over 2000. This was followed by a relatively steep decline until 31 July, when Fv/Fm was as low as 0.5, and rETRmax was below 60. As there was a 3-day gap in the measurements, the exact turning point date is unclear. However, from 3 August, a slight rise can be seen for 5 or 6 days, followed by a final small drop until 12 August, when the last measurements were taken. The values of the photosynthetic parameters were more scattered in the second part of the season, and the second increase was not very pronounced, especially in the case of Fv/Fm, for which some values lower than 0.45 were determined. Figure 4 shows the depths at which the snow algal cells were saturated with respect to light according to their Ek values, and the irradiances measured at the moment of sampling. The orange spores were undersaturated even though they were on the surface during the first half of the season, when the light was generally low, whereas from 1 August on, they were mostly saturated, with the saturating depth reaching more than 2 cm on some days. The purple spores seemed to have never reached saturation while in the snow.

Figure 4.

 Light-saturation depths and irradiance over the course of summer 2004. The white area (positive snow depth values) represents the snow column. Points in the grey area (negative depth values) represent samples of algae that would have not been saturated with respect to light, even when on the snow surface. Irradiance was measured at the moment of sampling.

The daily courses of photon flux density and photosynthetic parameters measured on 31 July and 12 August 2004 are shown in Figs 5 and 6, respectively. The samples contained only orange spores. On 31 July, it was overcast, with a rather stable irradiance over the course of the day not exceeding 250 μmol photons m−2 s−1. The maximum effective quantum yield (Fv/Fm) ranged between 0.48 and 0.57, the light-limited photosynthesis efficiency (α) between 0.41 and 0.54, and the saturating photon fluence (Ek) between 79 and 334 μmol photons m−2 s−1. The maximum relative electron transport rate (rETRmax) was also relatively stable and ranged within the interval 38–140, with an outlier at 250 (Fig. 5). On 12 August, there was fog lying above the snowfield until noon, and then clear skies, with intense direct radiation reaching 1460 μmol photons m−2 s−1 at 2 p.m. Around 10 p.m., a very dense fog began to form again, and reduced the irradiance to values below 50 μmol photons m−2 s−1. The photosynthetic parameters were more variable: the maximum effective quantum yield (Fv/Fm) was between 0.41 and 0.64, the light-limited photosynthesis efficiency (α) was between 0.37 and 0.59, the saturating photon fluence (Ek) was between 56 and 935 μmol photons m−2 s−1, and the maximum relative electron transport rate (rETRmax) was between 14 and 585 (Fig. 6).

Figure 5.

 Daily courses of photon flux density and photosynthetic parameters on 31 July 2004.

Figure 6.

 Daily courses of photon flux density and photosynthetic parameters on 12 August 2004 (note the different scales from Fig. 5).

Discussion

Composition and abundance of the snow algal community

The studied snowfield was dominated by Chlamydomonas nivalis, and small numbers of other algae, such as Chlamydomonas nivalis or Raphidonema spp., were found. This is in accordance with the previous observations of snow algae in Svalbard (Kol & Eurola, 1974; Newton, 1982; Müller et al., 1998, 2001; L. Benning, unpublished data). The most probable explanation of the dominance of Chlamydomonas nivalis is its complex life cycle, which enables effective exploitation of the snow niche as well as survival through the harsh periods when there is no liquid water available, plus effective photoprotection machinery (Bidigare et al., 1993). Species of Raphidonema, previously considered to be the ‘true’ snow algae, have been recently shown to be temporary residents in snow, and to have indigenous soil habitats (Stibal & Elster, 2005). The abundances of Chlamydomonas nivalis found in this study were very high, and may have partially been caused by the physical concentration of cells. Passive drifting of spores clustered by means of mucilage, and concentration of algae on the surface in areas where liquid water runs into or over the snow or where soil runoff flows onto the snow surface, have been observed in snow fields (Williams et al., 2003; Stibal & Elster, 2005).

Seasonal dynamics

The photosynthetic parameters in the green vegetative cells and the orange spores resembled those documented in snow algae from the Czech mountains (Kvíderováet al., 2005), and were generally higher than those recorded in algae from similarly cold but low-light habitats (Kühl et al., 2001; McMinn et al., 2003, 2004). The Fv/Fm values were similar to those determined in an Antarctic lichen growing in a cold, high-light environment (Barták et al., 2005).

The purple spores showed extremely high values for all the parameters. Although the samples containing purple spores were treated in exactly the same way as the other samples, the obtained values are probably exaggerated, if not completely artificial. The measurements may have been affected by the rapid sedimentation of the rather heavy cells, and some other unknown factors may have interfered as well. Moreover, a very low amount of purple spores was present at the locality and, therefore, an insufficient number of samples was measured. It is also possible that these spores belong to a different Chlamydomonas-like species with indistinguishable vegetative cells. However, the high values of Ek and resulting undersaturation with respect to light in the snow column are consistent with the intense colour, and therefore with the expected high content of red screening pigments. A more detailed study of these cells would be needed in order to deepen our knowledge of their biology and to eventually correct the values of the photosynthetic parameters.

The physiological state and photosynthetic activity of Chlamydomonas nivalis in the studied snowfield clearly showed seasonal dynamics, and these appeared to be dependent on some environmental factors. The observed course of the photosynthetic activity during the summer season included a first peak at the beginning of the experimental period, a subsequent decline, and another less distinctive peak. The first peak could represent the maximum activity of a fresh population of spores. The spores generally seem to be better adapted to the high irradiation and low temperatures than the green motile cells, and thus they were in a good physiological state. Over the course of the season, the spores may have begun to accumulate more screening pigments and therefore may have lost some of their activity, due to the reduced amount of photosynthetic active radiation available for the photosystems. Robinson et al. (1998) suggested that nutrient limitation could have been the cause of a decline in the Fv/Fm of algae living on platelet ice in Antarctica; nutrient limitation could also have occurred in our case. Also, some cells could be devoured by invertebrates living in the snow (Hoham & Duval, 2001), or damaged by various other factors. The damage could have been manifested as a decline of the photosynthetic activity. The second mild peak discernible for all the measured parameters may have been related to a change in the weather towards higher light conditions (7–12 August), and thus to an increase in the photosynthetic active radiation available to the cells. Another possible explanation for the decline and subsequent rise of photosynthetic activity could be a variable rate of production of reserve compounds such as starch or lipids: starch is mainly found in newly formed zygotes (Hoham, 1980). This can cause consequent changes in the energy allocation to intracellular processes. There may also be other, unclear, factors that contribute to the seasonal changes in the physiological state and photosynthetic activity of Chlamydomonas nivalis.

Diel dynamics

Whereas regular diel variations in the photosynthetic activity of snow algae in the polar regions may possibly occur at the end of the growing season with more regular light oscillations, a dynamic response to varying irradiance is more likely to take place in the peak season. The daily courses of the photosynthetic activity of Chlamydomonas nivalis in the studied Svalbard snowfield were found to be affected by the actual incident radiation. Similar dynamics have been described, for example, in Antarctic sea-ice algae (McMinn et al., 2003). Generally, there was greater variability in all the measured parameters on 12 August, when the values of the incident radiation also had a wider range. The values of the maximum effective quantum yield (Fv/Fm) showed a decrease when radiation increased on both days, which may have been a consequence of inhibitory irradiance. This has been shown in some organisms from similarly low-temperature and high-irradiation environments, such as lichens' photobionts in Antarctica (Barták et al., 2003), and in low-temperature and low-light Antarctic ice algae (McMinn et al., 2003).

Horizontal distribution of the algae in the snow

Regarding the light intensities in the snow and the saturating photon fluence values Ek, it is obvious that the vegetative cells and the common orange spores may be oversaturated with respect to light if they are on the surface. As the green motile cells are photoresponding (Vladimirov et al., 2004), they can actively escape to lower layers with a more favourable light intensity, where they give rise to the nonmotile spores. The spores can later be located on the snow surface as the snow melts, and they have to cope with the high irradiance, probably using the red screening pigments (Bidigare et al., 1993). Interestingly, there was no sign of acclimation of Ek to the changing irradiance, as has been described elsewhere (e.g. McMinn et al., 2004). This may be explained by the fact that the irradiance levels in the snow field were unstable and the changes occurred irregularly. Photoinhibition in the orange spores was likely to occur in some samples at the two highest photon fluences used for the RLC measurements, 620 and 900 μmol photons m−2 s−1 (data not shown), which is in contrast to the findings of studies performed in temperate mountains (Mosser et al., 1977; Williams et al., 2003). Whereas orange spores were light-saturated at values lower than 600 μmol photons m−2 s−1 at our locality, the purple spores showed an exceptionally high-saturation photon fluence over 3000 μmol photons m−2 s−1. This could probably be attributed to a very high content of screening pigments. However, an insufficient number of samples was measured, and these values might be artificial. They should, therefore, be taken with great caution.

Incident radiation thus seems to be a key factor that exerts control on the photosynthetic rate and efficiency in snow algae, and their vertical distribution in the snow column. In the high alpine environments of North America, values well above 2000 μmol photons m−2 s−1, corresponding to a photon fluence of 4500 μmol photons m−2 s−1, have been measured, and the photosynthesis of the snow algae, dominated by Chlamydomonas nivalis, was saturated at 1000 μmol photons m−2 s−1, with photoinhibition occurring at 2400 μmol photons m−2 s−1 (Mosser et al., 1977; Gorton et al., 2001). Such irradiances also allow the algae to photosynthesize at depths as great as 50 cm below the snow surface (Thomas, 1972). In Svalbard, lower irradiances are usual, due to a lower solar angle and frequent clouds, seldom exceeding 1500 μmol photons m−2 s−1 (Hanelt, 1998; this study). Thus, the algal cells are able, or even forced, to live in the upper layers of the snow in these latitudes, and they can colonize even rather shallow snow fields on very steep slopes.

Effects of other environmental factors

Although temperature has been shown to have an effect on the net photosynthesis of Chlamydomonas nivalis (Mosser et al., 1977; Remias et al., 2005), it is not likely to affect the dynamics of activity in situ, as the temperature is very stable in the melting snow over the summer season. The efficiency and rate of photosynthesis can be affected by other environmental factors, such as the chemistry of the snowmelt. No significant correlation was found between photosynthetic activity and electrical conductivity, which is often used as a proxy for the solute content. This could be attributed to the fact that the conductivity was constantly low, and so did not cause any variability. It is likely that nutrients, mainly nitrogen and phosphorus, affect photosynthetic activity. Subtle changes in their contents not affecting the overall conductivity may possibly have a great impact on the algal populations in the snow, and so bear the responsibility for a great part of the unexplained variability. Higher concentrations of cells may also cause self-shading and thus reduce the effective photon flux and subsequently the photosynthetic activity. Inorganic particles attached to cell surfaces or dark vacuoles inside the cells, probably formed as a result of pinocytosis of some elements (Remias et al., 2005; Lütz-Meindl & Lütz, 2006; L. Benning, unpublished results), could possibly provide supplementary shading for the cells of Chlamydomonas nivalis.

Acknowledgements

This study was supported by grant GA ASCR B6005409. The Delegation Fund of the Faculty of Biological Sciences provided financial support for MS and MŠ. Thanks are due to Anna Kowalska and Josef Řehák for help in the field, and to the University of Wrocław for the comfort and hospitality of the Baranowski station. Three anonymous reviewers are thanked for their comments on the manuscript.

Ancillary