Photosynthetic characteristics and biomass distribution of the dominant vascular plant species in a high Arctic tundra ecosystem, Ny-Ålesund, Svalbard: implications for their role in ecosystem carbon gain
- First Online:
- Cite this article as:
- Muraoka, H., Noda, H., Uchida, M. et al. J Plant Res (2008) 121: 137. doi:10.1007/s10265-007-0134-8
Studies on terrestrial ecosystems in the high Arctic region have focused on the response of these ecosystems to global environmental change and their carbon sequestration capacity in relation to ecosystem function. We report here our study of the photosynthetic characteristics and biomass distribution of the dominant vascular plant species, Salix polaris, Dryas octopetala and Saxifraga oppositifolia, in the high Arctic tundra ecosystem at Ny-Ålesund, Svalbard (78.5°N, 11.5°E). We also estimated net primary production (NPP) along both the successional gradient created by the proglacial chronosequence and the topographical gradient. The light-saturated photosynthesis rate (Amax) differed among the species, with approximately 124.1 nmol CO2 g−1leaf s−1 for Sal. polaris, 57.8 for D. octopetala and 24.4 for Sax. oppositifolia, and was highly correlated with the leaf nitrogen (N) content for all three species. The photosynthetic N use efficiency was the highest in Sal. polaris and lowest in Sax. oppositifolia. Distributions of Sal. polaris and D. octopetala were restricted to the area where soil nutrient availability was high, while Sax. oppositifolia was able to establish at the front of a glacier, where nutrient availability is low, but tended to be dominated by other vascular plants in high nutrient areas. The NPP reflected the photosynthetic capacity and biomass distribution in that it increased with the successional status; the contribution of Sal. polaris reached as high as 12-fold that of Sax. oppositifolia.
KeywordsHigh Arctic tundra ecosystemNet primary productionPhotosynthesisSalix polarisSaxifraga oppositifoliaSvalbard
The terrestrial ecosystems in the high Arctic region have been the focus of study due to their ecological characteristics of primary succession at the deglaciated area (Bliss et al. 1973). In recent years, the interest of researchers has intensified due to the potential effect of global climate change on the ecosystem dynamics and carbon sequestration of this region (Billings et al. 1982; IPCC 2001). Since 1994, we have been exploring the mechanisms of carbon cycle and sequestration in a deglaciated area of Ny-Ålesund, Svalbard, by measuring soil CO2 flux (Bekku et al. 2004a, b), the photosynthetic characteristics of bryophytes (Uchida et al. 2002) and dominant vascular plant species Salix polaris (Muraoka et al. 2002), the root respiratory characteristics of vascular plant species (Nakatsubo et al. 1998), net ecosystem CO2 exchange rates (Uchida et al. in preparation) and the decomposition process of soil organic matter (Yoshitake et al. 2007). The results of our studies have enabled us to construct a compartment model of carbon inputs and outputs of the dominant ecosystems (Nakatsubo et al. 2005).
In addition to the above mechanisms being responsible for the carbon dynamics of the ecosystems at Ny-Ålesund, the high Arctic tundra ecosystems have been shown to develop along the proglacial chronosequence (see Hodkins et al. 2003; Ohtsuka et al. 2006) with the topographical gradient having a major effect (Ohtsuka et al. 2006); vegetation is well developed in areas with high water and nutrient availability. The spatial distribution pattern of the different vegetation as well as the soil layer structure and microclimate would therefore affect the spatial pattern of carbon dynamics and sequestration (Williams and Rastetter 1999). For example, nutrient and water availability determines the photosynthetic capacity and, hence, biomass of the plants (Chapin and Shaver 1996), and the biomass would influence the carbon sequestration and decomposition processes. The distribution pattern of the plant species along the nutrient and water availability gradients (Hodkinson et al. 2003; Ohtsuka et al. 2006) would then also affect the capacity of carbon inputs into the ecosystems (Williams and Rastetter 1999; Johnson et al. 2000). The consequences of the interactions between the soil and vegetation development and carbon sequestration can be very dynamic along the proglacial chronosequence and topographical gradient from the edge of the glacier toward the shoreline of the Arctic Ocean at Ny-Ålesund.
In a previous study (Muraoka et al. 2002), we found that one of the dominant vascular plant species Sal. polaris, which constitutes much of the vegetation, especially in well-developed ecosystems, plays a crucial role in carbon sequestration of the deglaciated area due to its high leaf photosynthetic capacity and biomass. Recent investigations on the distribution of plant species have revealed that the number of species and the biomass of vascular plants increase along the chronosequence, partly due to the influence of topography (Ohtsuka et al. 2006). In the study reported here, our aim was to determine the leaf photosynthetic characteristics in terms of CO2 gas exchange and chlorophyll fluorescence response to light incidence for three dominant vascular plant species, Saxifraga oppositifolia, which is distributed throughout the area from near the edge of the glacier to the shoreline where vegetation has developed, Sal. polaris and Dryas octopetala, which are the dominant plant species in well-developed ecosystems. To gain insights into the relationship between the primary succession and ecosystem carbon gain, we attempted to estimate the spatial distribution of net primary production (NPP) by combining the photosynthetic characteristics and biomass distributions of the two representative species Sax. oppositifolia and Sal. polaris.
Materials and methods
The study site was located at the glacier foreland of the East Brøgger Glacier near Ny-Ålesund in the northwestern part of Spitsbergen, Svalbard (78.5°N, 11.5°E). Annual mean air temperature and precipitation in this area between 1995 and 1998 were −5.5°C and 362 mm, respectively. Snow melt occurs in early July, and snow starts to accumulate in late August; thus, the growth season of the plants is about 2 months. The study site is a polar semi-desert (Bliss and Svoboda 1984), and a mix community of bryophytes, such as Sanionia uncinata (Hedw.) Loeske and Aulacomnium turgidum (Wahlenb.) Schwaegr., and vascular plants, such as Saxifraga oppositifolia L., Salix polaris Wahlenb., Dryas octopetala L. and Luzula confusa Lindeb., constitute the well-developed plant community.
Vegetation and species distribution
Three 2.1-km-long line transects were drawn from the edge of the East Brøgger Glacier toward the shoreline of the Arctic Ocean for the vegetation survey in the summer of 2003. The lines were set apart at intervals of about 400 m, and 2 × 2-m quadrats were situated on each transect at approximately 120-m interval. Floristic composition and community structure were surveyed in each plot, including total vegetation coverage (%; vascular plants, lichen, algal crust, mosses), coverage by vascular plants (%) and the species composition of these vascular plants. Detailed information on the species composition is given by Ohtsuka et al. (2006).
Measurements of aboveground biomass (mainly leaves) were taken for Sax. oppositifolia, Sal. polaris and D. octopetalla in the summer of 2004. We selected three plots (20 × 20 cm) dominated by each species (i.e., apparent coverage of 100%) and clipped the aboveground parts. The collected leaves were dried at 60°C for 48 h to obtain leaf biomass per ground area.
Measurements of leaf photosynthetic characteristics
Leaf photosynthetic characteristics were measured for plants collected at the sites near the laboratory where the vegetation is well-developed. The plants were dug up carefully with root systems still attached and transferred by a plastic tray. The plants were watered and kept outside the laboratory so as to receive the ambient irradiance and air temperature. Prior to the measurements shoots were separated, taking care not to cut the roots.
Leaf photosynthetic response to light incidence was measured in the mid-summer of 2004 using an originally developed chamber and a portable photosynthesis measuring system (LI-6400; Li-Cor, Lincoln, NB). The chamber (diameter 10.0 cm; height 6.8 cm) was made of clear acryl in which the air was circulated by a small fan. A quantum sensor (LI-190SA, Li-Cor) and two thermocouples were placed in the chamber to measure photosynthetically active photon flux density (PPFD) and the temperature of the sample plants, respectively. Light was supplied by a metal halide light source equipped with a cooling filter (LA-180Me; Hayashi Tokei Kogyo, Japan) to avoid heating of the chamber and sample plants. Air temperature and relative humidity were manipulated by the LI-6400 to be about 9.0°C and 40%, respectively. Due to the light source and cover by the chamber, leaf temperature during the gas exchange measurements was 14–20°C; however, this temperature range does not reduce the photosynthetic activity of the plants (see Muraoka et al. 2002). Leaf photosynthetic rate and dark respiration rate at a constant ambient CO2 concentration of 370 μmol mol−1 were obtained as follows. First, one or two shoots (aboveground parts) with the root system attached were dug up and washed carefully and the light-response of CO2 gas exchange rates measured. Second, the leaves were carefully removed from the shoots and the CO2 exchange rates of the stem and roots measured. By adding the CO2 release rates (respiration) of the stem and roots to the CO2 absorption rates of the whole sample plant, we obtained the light-response of the leaf photosynthetic rate and leaf dark respiration rate. After the measurements, the samples were dried at 60°C for 48 h and weighed to calculate the photosynthetic and respiratory rates on a dry weight basis.
Leaf chlorophyll fluorescence was measured in the mid-summer of 2006 using a portable pulse-amplitude modulated fluorometer (PAM-2100; Heinz-Walz, Germany). Following the measurements of maximum photochemical efficiency for leaves that kept in the dark for 30 min [Fv/Fm = (Fm – Fo)/Fm, where Fm is the maximum fluorescence and Fo is the yield of fluorescence in the absence of actinic light; Schreiber et al. 1995], the photochemical efficiency of the light-response of photosystem (PS) II [ΔF/Fm′ = (Fm′ – Ft)/Fm′, where Fm′ is the maximum fluorescence in the light and Ft is the steady-state fluorescence in the light; Genty et al. 1989; Bilger et al. 1995] was measured by changing the incident light (5 min for each light intensity) using a metal halide light source equipped with a cooling filter (LA-180Me). For the fluorescence measurements, we dug up a small block of vegetation in the field, placed it on plastic tray and watered it well. We selected some shoots that were relatively isolated from the other shoots in the block and placed the fiber probe near the target shoots, taking special care not to shade the leaves from the light source. During the measurements, air temperature and ambient CO2 concentration were not controlled, we attempted to maintain ambient outdoor conditions by keeping the doors and windows opened. Following the measurements, the leaves within the area of flash light (diameter 22 mm) were harvested, dried at 60°C for 48 h and weighed. Since the fluorescence measurements were made for “clumped” small and not horizontal leaves, relative electron transport rate (JT) was calculated by correcting the measured leaf area from the leaf dry weight (Mleaf) and specific leaf area (SLA). JT was calculated as [ΔF/Fm′ × PPFD × 0.5/Mleaf × SLA], in which SLA was taken to be 0.0135 m2 g−1 for Sal. polaris and 0.00847 m2 g−1 for D. octopetala (obtained from leaf mass per area values reported by Kudo et al. 2001), and 0.5 assumes an equal distribution of photons between PS II and PS I. JT was not calculated for Sax. oppositifolia since we were not able to obtain its SLA.
All dried leaf samples were then used to measure nitrogen (N) content using an NC analyzer (Sumigraph NC-900; Sumika Chemical Analysis Service, Osaka, Japan).
Estimation of net primary production
Net primary production (NPP) of the leaves was estimated by combining the light-photosynthetic response curves, temperature dependency of the maximum photosynthetic rate and dark respiration rate, data on PPFD and leaf temperature obtained in the field from July 5 to August 19 in 2000 (Muraoka et al. 2002).
Thus, temperature-corrected values of Amax and R were used to obtain Agmax for Eq. (1), and then daily NPP based on leaf biomass was calculated by combining Eq. (1) and 10-min average PPFD data obtained at the surface of well-developed vegetation using a set of quantum sensors (LI-190SA; Li-Cor) and a datalogger. Because of the midnight sun, leaf photosynthetic rate was summed for 24 h per day.
The NPP in leaves of two representative vascular plant species, Sax. oppositifolia and Sal. polaris, were calculated by combining the data of vegetation coverage, vascular plant coverage, contribution of each species to the vascular plant coverage and leaf biomass per ground area of a plot that each species dominated (42.6 g leaf m−2 ground for Sax.oppositifolia and 26.5 g m−2 for Sal. polaris).
Leaf photosynthetic characteristics
Photosynthetic capacity (Amax), respiratory rate (R), nitrogen content (N) and N use efficiency (NUE) of the leaves of Saxifraga oppositifolia, Salix polaris and Dryas octopetala
Amax (nmolCO2 g−1 s−1)
R (nmolCO2 g−1 s−1)
NUE (μmolCO2 g−1N s−1)
24.4 ± 4.4 a
13.0 ± 2.3 a
1.82 ± 0.13 a
1.34 ± 0.24 a
124.1 ± 12.0 b
21.2 ± 6.8 b
3.05 ± 0.17 b
4.07 ± 0.29 b
58.0 ± 9.0 c
12.3 ± 2.4 a
2.20 ± 0.22 c
2.65 ± 0.41 c
Vegetation distribution and net primary production
The leaf-level studies on photosynthetic gas exchange and N content clearly show that the leaf N content largely determines the photosynthetic CO2 fixation capacity of the three species studied, which differ in successional status (Figs. 1, 2, 3). The present data on Amax of Sax. oppositifolia, Sal. polaris and D. octopetala were comparable to those reported by Crawford et al. (1993), Wookey et al. (1995) and Muraoka et al. (2002), respectively. Moreover, the leaf N contents of the three species were also comparable to those reported in the studies conducted in the high Arctic (compare Kudo et al. 2001; Tolvanen and Henry 2001; Van der Wal et al. 2000). The critical effect of leaf N content on photosynthetic capacity has been widely recognized for various plant species, ranging from herbaceous plants to tree species, to the degree that photosynthetic rate at ambient CO2 concentration is largely dependent on soluble protein content dominated by the enzyme ribulose 1,5-bisphosphate carboxylase (Field and Mooney 1986; Evans 1989).
Apart from the remarkable difference in photosynthetic capacity among the species, we did not find any significant difference in PS II photochemical efficiency (ΔF/Fm′), although there was a slight difference in electron transport rate (JT) between Sal. polaris and D. octopetala (Fig. 2). The similarity of ΔF/Fm′ among the three species could be partly due to the clumping effects of the plants; their small leaves were not able to be fixed horizontally so that the leaves were somewhat overlapped and/or erect. This result suggests the need for special care to be taken in the assessment of photosynthetic activity in these high-Arctic plants when using such an indirect method and that portable high-resolution spectral radiation meters may be more useful for in situ measurements than remote sensing (Gamon and Qiu 1999). In calculating JT, we attempted to correct for the influence of leaf clumping by using the published values of SLA. The higher JT in Sal. polaris is consistent with the latter having a higher Amax than D. octopetala.
It was interesting to find that the leaf N content was species-specific (Table 1) when the soil nutrient availability for the sample plants should have been similar since the study area was chosen at locations where vegetation was well developed and soil N content may well have been high (ranging from 1.3 to 2.1 km from the glacier’s edge; Fig. 5 and Ohtsuka et al. 2006). The species-specific leaf N content and resulting Amax would be constrained by the nutrient absorption capacity that is highly associated with the cost–benefit relationship between the acquisition and utilization of resources, partly as an adaptive strategy for successful growth and survival (compare Grime 1979; Field and Mooney 1986; Chapin 1989; Chapin et al. 1993). The remarkably low leaf N content and Amax in Sax. oppositifolia may be attributable to its successional status as a pioneer species in the high Arctic (compare Kume et al. 2003). Unfortunately we have not examined the effects of soil nutrient availability on leaf N content and, consequently, photosynthetic capacity over the research area. Further research on these direct consequences over the deglaciated area would provide further insights into the soil and vegetation developments along the proglacial chronosequence and topographical gradient.
The simple relationship between Amax and leaf N content in all three species suggests that photosynthetic CO2 fixation, the primary process of ecosystem carbon sequestration, can be estimated based on leaf N content and biomass per ground area, although we may still need to confirm whether the relationship can be extrapolated to other vascular plant species. Williams and Rastetter (1999) and Van Wijk et al. (2005) reported that the vegetation leaf area index (LAI) for the Arctic ecosystems is highly correlated with the total leaf N per ground area. The changes in coverages of vegetation (vascular plants, bryophytes and lichens) and of vascular plants along the 2.1-km line transects in our study were well correlated with the soil N content (Ohtsuka et al. 2006). The development of the soil layer and the availability of nutrients could be the major determinants of the development of vegetation and species number, thus directly affecting primary succession in the high Arctic ecosystems (Bliss et al. 1973; Chapin et al. 1993; Hodkinson et al. 2003; Ohtsuka et al. 2006). In addition, with increasing nutrient availability and, consequently, with increasing species number along the successional gradient of ecosystems, species composition can change, and the vegetation may become dominated by species having both a high nutrient absorption capacity and a hight photosynthetic N utilization efficiency (Table 1), both of which support a high growth rate (e.g. Grime 1994). As a result, these consequences lead to a higher photosynthetic carbon fixation per ground area along the successional gradient, as suggested by our simulation of NPP per ground area in which we focused on the representative species Sax. oppositifolia and Sal. polaris along the gradient of vegetation coverage and biomass (Fig. 5).
As discussed by Van Wijk et al. (2005) if we can quantify the spatial distribution of vegetation and the foliar biomass by indirect measurement techniques, such as remote sensing (portable devices, air-borne and satellite), and incorporate our knowledge on the relationship between Amax and leaf N content, we may then be able to estimate the spatial and temporal distribution of net primary production of Arctic ecosystems (e.g. Street et al. 2007). In fact, combined analysis of the vegetation survey, satellite remote sensing and incident radiation depending on the relationship between topography and sun angle has been providing information on the spatial distribution of the vegetation structure around Ny-Ålesund (Spjelkavik 1995; Nilsen et al. 1999a, b). Using these multi-scale approaches to measure the components determining the ecosystem carbon budget, we would then be able to make some progress in evaluating the consequences of primary succession and ecosystem function in the glacier foreland in the high Arctic ecosystem. This task is essential to any determination of the possible impacts of global climate change on the ecosystem carbon cycling from regional to global scales.
We thank S. Yoshitake of Hiroshima University and M. Adachi of Gifu University for their assistance in the field survey. Mr. H. Koike of Meiwafosis Co., Ltd (Tokyo, Japan) helped us to develop an original chamber for the LI-6400. Colleagues at the National Institute for Agro-Environmental Sciences kindly allowed us to use the NC analyzer. Thanks are also due to Prof. H. Kanda and Dr. S. Morimoto of National Institute of Polar Research and to the Norwegian Polar Institute for their support during our stay at Ny-Ålesund. We also thank the editor and two anonymous reviewers for their critical comments on the manuscript. This study was supported by a Grant-in-Aid for Scientific Research from Japan Society for the Promotion of Science, and partly by JSPS 21st century COE program “Satellite Ecology” at Gifu University.