Deepened snow alters soil microbial nutrient limitations in arctic birch hummock tundra

Abstract

Microbial activity in the long arctic cold season is low but cumulatively important. In particular, the size of the microbial biomass and soil solution nutrient pool at the end of winter may control the quantity of nutrients available to plants in the following spring. Microbial starvation and lysis as a result of increasingly severe soluble carbon (C) shortages over winter has been hypothesized as a potential mechanism for microbial nutrient release at thaw. These C shortages may be exacerbated by the warmer temperatures and increased winter precipitation that are consistently predicted for a large part of the low Arctic. In particular, warmer soil temperatures due to deeper snow may increase wintertime microbial activity and organic matter decomposition over the winter, potentially resulting in enhanced nutrient availability to plants in the following growing season.

In this study, we investigated nutrient limitations to soil microbial growth and activity in late winter under ambient and experimentally deepened snow (∼0.3 and 1 m respectively) in birch hummock tundra within the Canadian low Arctic. We hypothesized that the build-up of moderately deeper snow over winter would exacerbate soluble C-limitation to microbial growth and activity and increase soluble N accumulation, and thus stimulate the growth of bacteria relative to fungi. We measured the in situ response of the soil microbial biomass and soil soluble pools in control and snow-fenced plots at the end of winter, and then incubated soils from these plots with added C, nitrogen (N) and phosphorus (P) (at 0–15 °C) to characterize nutrient limitations to microbial growth and activity.

In late winter, deepened snow increased the microbial pool of N, yet decreased soil pools of dissolved organic N and C, and decreased bacterial counts. Fungal mass and hyphal lengths did not change, but remained dominant under both ambient and deepened snow. Deepened snow exacerbated the soluble C-limitation to microbial growth and reduced the P-limitation for microbial respiration. Fungal mass and hyphal length responses to nutrient addition were larger than the bacterial mass or abundance responses and fungi from under deepened snow responded more than those from under ambient snow, indicating a different potential structural and physiological response to substrate availability for these two soil microbial communities. Our results indicate that deeper snow may increase microbial nutrient pools and can alter the physiological functioning of the soil microbial community in late winter, suggesting that microbial N release and its availability to plants during spring thaw may be enhanced.

Introduction

The influence of temporal variation in soil microbial activity and growth on annual N cycling is not well understood in seasonally snow-covered ecosystems (Lipson et al., 2000, Grogan and Jonasson, 2003, Weintraub and Schimel, 2005, Edwards et al., 2006, Buckeridge and Jefferies, 2007). Ongoing low levels of microbial decomposition of organic matter during the long, cold winters that typify tundra ecosystems may be an important contributor to the annual nitrogen (N) supply for plants (Lipson and Monson, 1998, Grogan and Jonasson, 2003). Several studies have documented a decline in the winter-adapted microbial biomass carbon (Brooks et al., 1998, Lipson et al., 2000, Grogan and Jonasson, 2003, Edwards et al., 2006) and microbial biomass nitrogen (Brooks et al., 1998, Edwards et al., 2006) across the spring thaw phase, suggesting that the nutrients released from the microbial cytoplasm may be released into the soil solution as a characteristic spring pulse. If release from the microbial pool is a principal mechanism providing the soluble nutrient flush in spring (Herrmann and Witter, 2002), then the composition, size and nutrient content (Schimel et al., 2007) of that winter microbial pool is clearly a critical control on the supply and fate of that flush. This nutrient flush, in turn, may be an important resource for plants and the surviving microbial community at thaw, as well as over the subsequent growing season (Schimel and Clein, 1996).

Several hypothetical mechanisms for microbial lysis and cytoplasmic release at the end of winter have been proposed. Freeze–thaw cycles in laboratory experiments lead to bacterial death and the release of cytoplasmic nutrients (Skogland et al., 1988). However, field and mesocosm studies suggest that soil freeze–thaw cycle regimes that are typical of many tundra environments may adversely affect only a relatively small proportion of the microbial biomass (Lipson and Monson, 1998, Lipson et al., 2000, Herrmann and Witter, 2002, Grogan et al., 2004). Nutrient pulses may also be the result of faunal predation of fungi and bacteria (Ingham et al., 1986), as soil macrofauna often proliferate during thaw (Sjursen et al., 2005). Alternatively, or additionally, the microbial community may be responding osmotically to sudden and enormous increases in soil water at spring thaw (Schimel et al., 2007). Finally, lab incubations with alpine tundra soils (Lipson et al., 2000, Brooks et al., 2004) have suggested that cytoplasmic release at thaw is primarily the result of microbial mortality due to increasingly extreme C-limitation toward the end of winter (Schmidt and Lipson, 2004). Microorganisms may remain active throughout most of the cold season in alpine systems that have deep snow accumulation (∼2 m) and/or moderate winter soil temperatures. Under these circumstances, microbial C-limitation increases as inputs from plant litter are depleted, and the winter-adapted microbial community ultimately succumbs to warmer temperatures and C starvation during spring thaw (Lipson et al., 2000). Wintertime conditions in the Arctic may differ significantly from many alpine sites in that soil temperatures are often substantially lower in the Arctic (Brooks et al., 1998, Schimel et al., 2004), and spring thaw drainage is relatively restricted. Here, we focused on investigating the potential for the C-limitation hypothesis to explain microbial turnover and a spring nutrient flush in an arctic ecosystem. In particular, we were looking for evidence that late winter microbial growth in the Arctic is limited by available labile organic substrate.

Strong C-limitation on microbial growth and activity may have conflicting impacts on winter N mineralization and therefore soil N pools. Microbial immobilization of N for growth occurs when microbes are utilizing organic substrates with a relatively high C:N ratio (Weintraub and Schimel, 2003). Tundra plant communities and plant litter typically have high C:N ratios, reflecting the low N availability in these ecosystems (Shaver and Chapin, 1980). A seasonal change between summer net N immobilization and winter net N mineralization is typically recorded in tundra soils (Grogan and Jonasson, 2003, Schimel et al., 2004). Recent evidence suggests winter substrate use may be largely confined to microbial recycling of dead microbial cells and hyphae (Schimel and Mikan, 2005) or endogenous metabolism (the breakdown of living cell constituents/storage compounds for maintenance). If microbial enzymes are primarily decomposing microbial products in the cold season, providing inputs of C and N with a relatively low C:N to the soil solution, then even though the soluble N pool size does not change, the result should be net N mineralization, and a larger inorganic N pool in the soil by late winter (Schimel et al., 2004). Alternatively, intra-seasonal variation in labile substrate availability, as root exudates and plant litter with low C:N inputs decline over the winter months, may lead to initial high rates of net N mineralization in early winter, but net N immobilization by the late winter. Therefore, mobilized winter N may be directly available to plants in spring, or dependent upon controls over microbial survival at spring thaw.

Large increases in snow depth (to 3 m) increased soil temperature minima from −25 to −7 °C, increased respiration (up to 4×) and elevated late winter soil N pools (∼3×) (Schimel et al., 2004), indicating that wintertime microbial decomposition of organic matter and mineralization is highly sensitive to soil temperature. However, the potential increases in winter precipitation predicted by climate change scenarios (ACIA, 2004), or snow build-up as a result of shrub-snow feedbacks (Sturm et al., 2005), suggest that relatively moderate increases in snow depth (∼1 m) are much more likely. Moderate increases in snow depth may lead to more moderate increases in winter soil temperature. Recent research suggests that there may be a snow depth threshold (∼1 m) below which there are strong insulating effects of increasing snow accumulation on soil temperature and above which effects of additional snow accumulation on soil temperature are relatively small (Grogan and Jonasson, 2006). Here, we investigate soil and microbial C and N pools under ambient and moderately deepened snow to determine if more realistic snow depth increases for arctic tundra in future winters will stimulate soil microbial activity and further increase late winter C-limitation.

Soil microbial community change between summer and winter may be a key control on annual patterns of nutrient cycling and plant N uptake in seasonally snow-covered ecosystems (Schmidt et al., 2007). A fungal community phylogenetic study of an alpine tundra soil revealed a diverse and novel soil fungal community in winter as compared to spring and summer, and a larger fungal: bacterial biomass ratio in winter (Schadt et al., 2003). In contrast, an investigation of the soil microbial community in arctic tundra found that vegetation type was the main control on microbial community structure, with seasonal shifts in species composition only at fine taxonomic scales (Wallenstein et al., 2007). Soil bacterial and fungal groups include species that are psychrophiles (specialists with optima <10 °C, active at subzero temperatures and prevalent in stable, very cold environments) and psychrotolerants (generalists with optima <20 °C, active at subzero temperatures and prevalent in unstable, very cold environments (Morita, 1975)). Little is known about the specific functions of either the specialists or the generalists. In an arctic winter, thin films of liquid water within frozen soils allow for microbial activity at subzero temperatures, at least until −12 °C (Rivkina et al., 2000, Robinson, 2001). Fungi may be favoured over bacteria in the decomposition of organic matter in cold, dry soil, since their mycelial growth habit may allow exploitation beyond individual microsites of liquid water, or across thin films of liquid water in frozen soils. Such differences may have profound effects on soil biochemical cycling during winter and the subsequent growing season, not least because fungi and bacteria often differ strongly in N concentrations, and probably in N storage capabilities (Klionsky et al., 1990, Pokarzhevskii et al., 2003, Schimel et al., 2007). However, to date we are not aware of any substantial studies documenting relative fungal and bacterial abundances in winter soils in the Arctic. For instance, if bacterial growth and activity are disproportionately constrained by winter conditions in the Arctic, warmer soils as a result of deeper snow may significantly increase the bacterial component of the active microbial community. Since cold-adapted bacteria may have different nutrient limitations than cold-adapted fungi, they may respond differently to changes in nutrient availability in late winter, and may differ in their susceptibility to the mechanisms that cause microbial turnover and nutrient release at thaw. Thus, changes in wintertime fungal to bacterial ratios could substantially alter annual patterns of tundra ecosystem nitrogen cycling.

In this study, we measured in situ late winter soil and microbial nutrient pools, and bacterial and fungal mass and abundances, in soils from ambient and experimentally deepened snow plots. Our objective was to investigate if moderate increases in snow depth, which might realistically be expected as part of interannual variation or climate change, would increase soil biochemical activity over winter, enhancing microbial growth and N accumulation. Secondly, we incubated these soils with factorial combinations of C, N and P additions to determine the effect of deepened winter snow cover on microbial nutrient limitations to activity and growth (net biomass production) during the early spring thaw phase. Finally, we quantified the bacteria and fungi in these soils and investigated the impact of deepened snow on their relative abundances. Specifically, we tested the following hypotheses for a birch hummock tundra system in the Canadian low Arctic:

• labile C availability limits the growth and activity of soil microorganisms at the end of winter, and deeper snow exacerbates this limitation;

• moderate increases in snow depth enhance soil extractable N pools in late winter;

• fungi dominate the soil microbial community mass in late winter; and

• bacterial abundances increase under deeper snow.