Short communication

Effects of freezing and thawing of some grassland topsoils on oxygen uptakes and dehydrogenase activities

The aim of this study was to assess initial effects of warming on the nutrient pools of carbon and nitrogen of two most widespread ecosystem types, swamp meadow and alpine meadow, in the Qinghai-Tibet Plateau, China. The temperature of the air and upper-soil layer was passively increased using open-top chambers (OTCs) with two different temperature elevations. We analyzed air and soil temperature, soil moisture, biomass, microbial biomass, and nutrient dynamics after 2 years of warming. The use of OTCs clearly raised temperature and decreased soil moisture. The aboveground plant and root biomass increased in all OTCs in two meadows. A small temperature increase in OTCs resulted in swamp meadow acting as a net carbon sink and alpine meadow as a net source, and further warming intensified this processes, at least in a short term. On balance, the alpine ecosystems in the Fenghuoshan region acted as a carbon source.

Seasonally snow-covered alpine soils may be subjected to freeze/thaw cycles, particularly during years having little snow and during the late winter and early spring periods. Freeze/thaw cycles can stimulate soil mineralization and could therefore be one factor regulating nitrogen (N) and phosphorus (P) availability and cycling. In this study laboratory incubation experiments using four soils having contrasting properties have been used to characterize the change in N and P forms (microbial and soluble inorganic/organic) that occur after simulated freeze/thaw cycles.

Soil samples were collected from locations representing extreme examples of either direct human management (grazed meadow (site M) and extensive grazing beneath larch (site L)) or those disturbed by more natural events (recent avalanche and colonisation by alder (site A)) and from beneath the expected forest climax vegetation beneath fir (site F). Topsoil from these sites, maintained at two different water contents (20 and 30%, w/w), were exposed to either a single (SF) or four sequential (4SF) freeze/thaw cycles. Each cycle consisted of 12 h at −9 °C and 12 h at +4 °C mimicking a diurnal pattern.

A SF cycle reduced microbial N for soils from sites F and A and was accompanied by a significant increase in dissolved organic nitrogen (DON) at both moisture contents. In contrast, the microbial N of soils from M and L was not affected by the freeze/thaw cycles, suggesting a particular adaptation of soil microbes to these extremes in temperature. Freeze/thaw cycles resulted in a significant increase in the net ammonification in all soils.

Extractable total dissolved N (TDN) and total dissolved P (TDP) increased in all soils after a SF cycle, however, the relative importance of the different N and P forms differed. At the lower soil moisture content, NO₃⁻ concentrations remained constant or slightly decreased in all soils, except that from site M. In all other soils DON appeared to replace NO₃⁻ as the potentially mobile N source after the freeze/thaw cycles. The relative contribution of dissolved organic P to TDP after freeze/thaw remained significant, and greater than 50% in all soils.

Freeze/thaw cycles, in seasonally snow covered soils, are likely to have a selective effect on the microbial biomass. Freezing and thawing resulted in a pulse of net ammonification and DON release, which represent an important influence upon N cycling in these alpine systems.

Mesocosms from two subarctic heaths from high and low altitudes, respectively, and differing in the composition of vegetation were experimentally subjected to 18 diurnal freeze–thaw cycles fluctuating between day temperatures of +2 °C and night temperatures of –4 °C. Mesocosm respiration was consistently lower and higher in cycled mesocosms compared to mesocosms kept unfrozen (+2 °C) and frozen (–4 °C), respectively. When extrapolated to the entire non-growing season, our data suggest a non-growing season carbon (C) loss in the order of 20–30 g C m⁻², which is substantial compared with a likely annual incorporation of plant biomass C of 150–200 g C m⁻² per year. The results support the hypothesis that winter carbon fluxes must be included in annual carbon budgets for arctic ecosystems.

Microbial biomass C decreased only in cycled mesocosms, indicating that while microbial activity is mainly controlled by temperature, microbial biomass is strongly affected by temperature fluctuations around the freezing point. In contrast to microbial C, microbial biomass nitrogen (N) mass did not change in cycled mesocosms, leading to decreased microbial C to N ratio. Second, soil inorganic N concentration declined in the soils subjected to the cycles, showing that microbes surviving the freeze–thaw cycles had a high potential for sequestering soil N and N that presumably was released from dead microbes. The two sets of mesocosms responded very similarly to the freeze–thaw cycles, indicating that the observed responses to multiple diurnal freeze–thaw cycles are common for both ecosystem types despite the differences in plant species composition and climate.

In Colorado alpine dry meadow soils, microbial biomass has been observed to increase during fall and winter and to rapidly decline after snowmelt in the spring. It has been shown that these microbial population dynamics are linked to N availability to alpine plants, but the underlying mechanisms have not been explained. We hypothesized that: (1) freeze-thaw events in the spring cause reduction of the microbial biomass, (2) the winter microbial community is sensitive to prolonged temperatures above 0°C, and (3) the increase of biomass in fall and its decline in spring are due to changes in C availability. We performed laboratory experiments to test the effect of temperature regime on soil microbial biomass, respiration and C availability, and made seasonal measurements of C pools. Soil microbial biomass was unaffected by freeze-thaw events in which realistic rates of freezing and thawing were used. Some significant effects were observed at faster freezing rates. Despite this tolerance to temperature fluctuations, the winter microbial community showed sensitivity to prolonged temperatures above 0°C. This effect may have been caused indirectly by an effect of temperature on substrate availability. Two week incubations at increased temperatures caused a reduction in the quantity of extractable organic C in the soil. The soil concentrations of cellulose and hot water-soluble organic C were the lowest in the summer and the highest in spring and autumn, mirroring previously measured patterns of microbial biomass. This suggests that C from litter inputs could be a strong control over microbial biomass. Respiration rates in soils collected before snowmelt were high at 0°C, and did not respond immediately to addition of glutamate. At 22°C, or after a two week incubation at 0°C, respiration in these soils became substrate-limited. Respiration rates in soils collected during the summer were very low at 0°C, but responded immediately to glutamate addition at both 0 and 22°C. These results show that the C balance of the soil microbial biomass undergoes a critical shift between winter and summer due to an increase in temperature and a corresponding decrease in C availability. This shift could explain the decline in microbial biomass after snowmelt.

Four tundra and taiga soils were experimentally subjected to three freeze-thaw cycles (5 days each at −5°C and +5°C). After each thaw, there was an initial pulse (<24 h) in microbial respiration. The total amount of C respired in each thaw period was largest during the first cycle and decreased in successive cycles. Three cycles caused a net increase in total respired C relative to the +5°C control in wet meadow tundra soil, a reduction in birch soil, and had no net effect in either alder-poplar or tussock tundra soil. These different patterns apparently resulted from differences in the quality of the soil organic matter and the relative activity of the microbial biomass. Net N mineralization was generally enhanced relative to the +5°C controls in the first cycle, but was inhibited in the third cycle, similar to what was observed with C. Over multiple freeze-thaw cycles, the initial response of C and N mineralization appear to be driven by release from the microbial biomass, while over the longer-term, the response is driven by the reduction in attack on soil organic matter resulting from a reduced microbial population.

The respiratory burst after freezing and thawing of soil has its maximum immediately after thawing. The reduction in bacterial plate counts after a freeze-thaw cycle was moderate, and so was also the subsequent increase. Experiments with isolated bacteria from the same soil showed that there was a similar burst when bacteria were frozen and thawed in the stationary phase. Between different strains, there was a positive correlation between the killing effect of the freeze-thaw treatment and the respiratory increase per surviving bacterium. The increased O₂ uptake during the first 11 h after thawing corresponded to a C consumption 5–15% of that in the killed bacteria. A brief respiratory increase was observed when the supernatant from a frozen-thawed culture was added to bacteria in the stationary state. The burst was more persistent when frozen-thawed bacteria were centrifuged and then resuspended in supernatants from stationary cultures of unfrozen and frozen bacteria, or even in a mineral salts solution. The optical difference spectrum between the supernatants from frozen and unfrozen cultures demonstrated leakage of cellular material after freezing and thawing. Monod kinetics for growth shows that if the substrate concentration is substantially lower than K_s, and the biomass concentration relatively high, the respiratory rate is at maximum at the time when the substrate is added. The respiratory burst following freezing and thawing can thus be explained on the basis of simple growth kinetics.