Effect of freezing-thawing on nitrogen mineralization in vegetation soils of four landscape zones of Changbai Mountain

Abstract

• Introduction

We studied the effect of freezing-thawing on nitrogen (N) mineralization of four vegetation soils from typical vegetation zones of Changbai Mountain with a laboratory incubation experiment. The soils were treated with two levels of soil water content, representing the low and high soil water contents found during late autumn and early spring in Changbai Mountain, respectively, and underwent cycling of freezing at −5 or −25°C and thawing at 5°C up to 15 times.

• Objectives

The main purpose of this study was to examine the effects of freezing temperature, frequency of freezing-thawing cycles, and soil water content on N mineralization of four soils to reveal the different effects of spring and autumn freezing-thawing on soil N mineralization in Changbai Mountain.

• Results

The results showed that inorganic N in the soils increased 1.67–26.77 times after 15 cycles of freezing-thawing, but N mineralization rate decreased with increased cycling of freezing-thawing. The lower freeze temperature and higher soil water content generally enhanced soil N mineralization. The results implied that freezing-thawing of vegetation soils to increase soil N mineralization to favor the growth of plants, and also increase the possibility of runoff loss of soil nutrients, is more effective in the spring than in the autumn.

1 Introduction

Nitrogen (N) is one of the key nutrients limiting plant growth in terrestrial ecosystems. Thus, there is a significant relationship between the productivity of forest ecosystems and soil N availability (Dalias et al. 2002). Soil organic N is the predominant form of soil N. Since plants take up mainly inorganic N such as ammonium (NH ⁺₄ ) and nitrate (NO ⁻₃ ), soil organic N first needs to be transformed into inorganic N by microbial mineralization before it can be used by plants. The mineralization of organic N in soils is an essential part of N cycling in ecosystems (Chapin et al. 2002). This process depends primarily on the activity of soil microbes, which are affected by many environmental factors including temperature, soil moisture and soil porosity (De Neve et al. 2003).

Soil temperature is one of the most important factors affecting soil N mineralization (Dalias et al. 2002). For terrestrial ecosystems in mid- and high-latitude areas, the cycle of freezing-thawing is controlled normally by changes in air temperature (Henry 2007). The intensity and frequency of soil frost depend mainly on regional climate and the thickness of snow cover on the soil. Along with global climate change, soil frost may be strengthened in some regions due to less snow in cold seasons, while the general warming trend may decrease the intensity, frequency and duration of soil frost in other regions (Matzner and Borken 2008). The cycle of soil freezing-thawing results in a number of changes to soils, including the death of soil microbes (Larsen et al. 2002), destruction of soil aggregates (Edwards 1991; Kvarno and Oygarden 2006), redistribution of soil water (Hardy et al. 2001) and the accelerated death of fine roots due to decreased nutrient adsorption (Tierney et al. 2001). These changes may further affect the mineralization and nitrification of soil N, subsequently increasing the content of soluble organic matter, decreasing the amount of nutrients absorbed by plants, enhancing the loss of nutrients (Deluca et al. 1992; Neilsen et al. 2001; Fitzhugh et al. 2001; Vestgarden and Austnes 2009).

Two methods have been used to study the effect of freezing-thawing on soil N mineralization. The first is to control the thickness of snow in the field (Tierney et al. 2001; Schimel et al. 2004; Groffman et al. 2006). The second involves laboratory incubation experiments controlling incubation temperature (Deluca et al. 1992; Neilsen et al. 2001; Freppaz et al. 2007). Previous laboratory incubation studies have focused mainly on examining the effect of temperature and frequency of freezing-thawing cycling on soil N mineralization. The soil water content in the spring, likely reaching over-saturation due to snowmelt, is generally higher than soil water content in the autumn (Henry 2007). Nevertheless, the effect of such differences in soil water content between the spring and autumn and, further, the effect of freezing-thawing on soil N mineralization are not well studied.

Changbai Mountain is the highest mountain in northeastern China. The temperate forest ecosystems in Changbai Mountain represent typical mountain forest ecosystems in East Asia. Due to the topography and the varying hydrothermal conditions, there is a distinct vertical distribution of vegetation zones on the mountain, presenting a condensed picture of the array of temperate and boreal forests found across northeastern China (Hao et al. 2007). Previous studies on N cycling in the temperate forest ecosystems of Changbai Mountain have all been conducted in the growing season (Zhou and Ou 2001). Little attention has been focused on N dynamics of the forests in the non-growing season. Soils of different elevation-controlled vegetation zones in Changbai Mountain are subjected to repeat cycling of freezing-thawing during the late autumn and early spring.

Thus, we present here a laboratory incubation study using four vegetation soils from Changbai Mountain. The primary objective of this study was to examine the effects of freezing temperature, the frequency of freezing-thawing cyles, and soil water content on N mineralization of the four soils in order to reveal the different effects of spring and autumn freezing-thawing on soil N mineralization in Changbai Mountain.

2 Materials and methods

2.1 Field site

The study site is located in the Changbai Mountain National Natural Reserve, northeastern China (41°43′–42°26′N, 127°42′–128°17′E). This natural reserve was established in 1960 and joined the Biosphere Reserve Network in 1980 as a part of the United Nation Man and Biosphere (MAB) Programme. This natural reserve is characterized by the most representative mountain forest ecosystems, with distinct altitudinal distribution of soils, vegetation, and climate. The reserve contains 1,337 vascular plant species, including 1,250 seed plant species. There are four vegetation zones from low to high altitude: broad-leaved Korean pine forest (BKPF), spruce-fir forest (SFF), Betula ermanii forest (EBF) and subalpine tundra (ST). The climate of the area is characterized by a mountain climate with a dry and windy spring, a short and rainy summer, a cool and foggy autumn, and a cold and long winter (Figs. 1, 2). The mean annual temperature varies between 3.3 and 7.3°C, with mean temperature ranging from 8.7 to 19.3°C in July, and from −23.3 to −16.1°C in January. There are many freezing-thawing cycles in late autumn and early spring (shaded parts of Fig. 1). Annual mean precipitation is about 600–1,400 mm (Shao et al. 1996).

Fig. 1

Daily variations in maximum (gray solid line) and minimum (black solid line) temperature in broad-leaved Korean pine forest (BKPF), spruce-fir forest (SFF), Erman’s (Betula ermanii) birch forest (EBF) and subalpine tundra (ST) from October 2008 to April 2009

Fig. 2

Mean monthly precipitation (1990–2009)

2.2 Soil collection

Soil samples were collected from BKPF, SFF, EBF, and ST using a soil auger on 12 October 2008 before the first occurrence of soil freezing under field conditions (Table 1). The average soil water content of these four vegetation types was 38.4%, 55.7%, 59.5%, and 75.6%, respectively. To collect soil samples, 5 1 m × 1 m plots were chosen in each vegetation zone. Plots were at least 20 m apart. The diameter of the soil auger was 5 cm and the sampling depth was 0–15 cm. The soil samples from each vegetation zone were mixed thoroughly, sieved through a 4-mm sieve, and then used for the incubation experiment and analysis of soil properties after being air-dried.

Table 1 Major soil properties (0–15 cm depth) of four vegetation zones of Changbai Mountain. Values shown are mean±SD (n = 3)

2.3 Incubation experiment

Soil freezing-thawing cycles occur both in late autumn and early spring on Changbai Mountain. However, the soil water content of the vegetation is generally higher in the early spring due to snowmelt. A laboratory incubation experiment was conducted to examine the effects of freezing temperature, freezing-thawing frequency, and soil water content on soil organic N mineralization. A sample of 50 g dry weight of soil was filled into a 200 ml plastic bottle. De-ionized water was added to the bottle to adjust soil moisture content. The low soil water content treatment (WL) was set according to the soil water content at the sampling date to study the effect of freezing-thawing cycling on soil N mineralization in late autumn, and the high soil water treatment (WH) was set near to the saturation of soil water to study the effect freezing-thawing cycling on the soil N mineralization in early spring. The plastic bottles filled with soils were sealed with plastic film to avoid water evaporation from the soil before the freezing-thawing treatment.

The bottles were frozen at either −5°C or −25°C for 1 day, and then thawed at 5°C for 1 day. That process defined one cycle of freezing-thawing. Three bottles of soil (n = 3) from each treatment were taken for analysis after 1, 3, 7 and 15 freeze-thaw cycles. The inorganic N content (NO ⁻₃ -N and NH ⁺₄ -N) in the soil was analyzed to see the effect of freeze-thaw cycling on soil N mineralization.

2.4 Soil analysis

After a defined incubation time of freezing-thawing cycling, the contents of NH ⁺₄ -N and NO ⁻₃ -N in the soil were extracted immediately with 2 M KCl using a 1:5 soil:extractant (w/v) ratio on a reciprocal shaker for 30 min. The extracts were filtered through filter paper and stored in a freezer until analysis of NH ⁺₄ -N and NO ⁻₃ -N concentrations (Raison et al. 1987). The soil water content was determined by drying the soil at 105°C for 48 h. The soil N mineralization rate was calculated as the difference in the inorganic N content in the soil before and after incubation divided by incubation time. A similar formula was used to calculate soil net N ammonification rate and soil net N nitrification rate.

$$ {\text{Soil net N mineralization rate }} = {\text{ }}\left( {{\text{inorganic N content after incubation }} - {\text{ inorganic N content before incubation}}} \right)/{\text{ incubation days}} $$

$$ {\text{Soil net N ammonification rate}} = \left( {{\text{NH}}_{4}^{ + } - {\text{N content after incubation}} - {\text{NH}}_{4}^{ + } - {\text{N content before incubation}}} \right)/{\text{incubation day}} $$

$$ {\text{Soil net N nitrifications rate}} = \left( {{\text{NO}}_{3}^{ - } - {\text{N content after incubation }} - {\text{ NO}}_{3}^{ - } - {\text{N content before incubation}}} \right)/{\text{incubation day}} $$

2.5 Statistical analysis

All soil inorganic N results were expressed as mg N /kg oven-dried soil. Treatment effects on soil N mineralization were also tested using ANOVAs. All statistical analysis was performed using SPSS version 16.0 (SPSS, Chicago, IL). Statistical analysis was judged significant if P < 0.05 ,and highly significant if P < 0.001.

3 Results

3.1 Mineral N flux of soils during incubation

The initial content of inorganic N in the soils was influenced significantly by vegetation type (P < 0.001). The initial content of inorganic N of BKPF, SFF, EBF and ST was 19.44 ± 0.65, 23.93 ± 1.06, 12.38 ± 0.27 and 14.71 ± 0.48 mg/kg, respectively. Ammonium concentration was the dominant inorganic N form except in soil from BKPF.

ANOVA results indicated that total inorganic N (TIN) of four vegetation soils was influenced significantly by freezing-thawing times (FT), soil water content, and freezing temperature (P < 0.05) (Table 2). The TIN contents of all soils increased with increased frequency of freezing-thawing cycles (Fig. 3). After freezing at −5°C and thawing at 5°C, the TIN content of BKPF soil was 35.95 ± 0.55 mg/kg at the high moisture treatment after 15 times freezing-thawing cycles but not significantly different (P = 0.072) from 48.96 ± 0.98 mg/kg, i.e., the TIN in the low moisture treatment. Nitrate was found to be the main form of TIN in both treatments (Fig. 3a). After receiving the same freezing-thawing treatment, the TIN content in the soils of SFF, EBF and ST was 60.54 ± 1.22, 56.39 ± 1.47 and 44.66 ± 1.00 mg/kg at high soil moisture, and 52.87 ± 0.96, 46.35 ± 1.13 and 37.83 ± 0.77 mg/kg at low soil moisture, respectively. ANOVA results revealed a significant difference between the low and high soil moisture treatments on soil mineralization (P < 0.05), with ammonium being the predominant form of inorganic N (Table 2, Fig. 3b–d).

Table 2 Results of three-way ANOVAs (F-values) testing the effects of freezing-thawing times (FT), soil water content (SW) and freeze temperature (T) on total inorganic N in the four vegetation zones of Changbai Mountain

Fig. 3

Changes in soil inorganic N (mg/kg) in the process of freezing-thawing. FT0 No freezing-thawing, FTn n freezing-thawing cycles (n = 1, 3, 7, and 15). Letters above the bars indicate significant differences between times after one-way ANOVA with Tukey’s test (n = 3, P < 0.05)

After soil samples were treated by freezing at −25°C and thawing at 5°C for 15 times, soil TIN content of BKPF, SFF, EBF and ST increased to 104.47 ± 2.83, 103.16 ± 2.64, 103.38 ± 3.03 and 86.92 ± 2.41 mg/kg at high soil moisture, and 81.65 ± 2.07, 86.02 ± 2.07, 80.06 ± 2.26 and 59.17 ± 1.48 mg/kg at low soil moisture, respectively. Soil TIN contents in the low and high moisture treatments were significantly different (P < 0.05). At the end of incubation, ammonium was the dominant form of soil inorganic N in all four soils.

After 15 freezing-thawing cycles, soil TIN content was significantly higher when freezing at −25°C than freezing at −5°C. High soil moisture appeared to enhance soil organic mineralization except for treatment of BKPF soil frozen at −5°C and thawed at 5°C (Fig. 3, Table 2).

3.2 Soil net N mineralization rate

With the increase in freeze frequency, soil net N mineralization rate decreased gradually (Fig. 4). After 15 times of freezing at −5°C and thawing at 5°C, soil net N mineralization rates of BKPF, SFF, EBF and SA decreased from 2.53, 1.94, 8.43 and 3.24 mg/(kg.d) to 0.55, 1.22, 1.47 and 1.00 mg/(kg.d) at high soil moisture, respectively. While at low soil moisture, those values decreased from 1.22, 2.13, 1.76 and 2.10 mg/(kg.d) to 0.98, 0.96, 1.13 and 0.77 mg/(kg.d), respectively.

Fig. 4

Net N mineralization rate [mg/(kg day⁻¹)] following the process of freezing-thawing of forest soils. FT0 No freezing-thawing, FTn n freezing-thawing cycles (n = 1, 3, 7, and 15). WL Low soil water content treatment, WH high soil water content treatment

After 15 cycles of freezing at −25°C and thawing at 5°C, soil net N mineralization rates of BKPF, SFF, EBF and SA decreased from 10.51, 4.50, 9.76 and 5.64 mg/(kg.d) to 2.83, 2.64, 3.03 and 2.41 mg/(kg.d) at high soil moisture, respectively. While at low soil moisture, those values decreased from 5.10, 5.32, 6.96 and 5.99 mg/(kg.d) to 2.07, 2.07, 2.26 and 1.48 mg/(kg.d), respectively.

At the end of incubation, the rates of net ammonification, net nitrification and net mineralization were all positive, except for the net nitrification rate of SFF soil freezing at −5°C and thawing at 5°C at low soil water content. In the process of freezing and thawing, net N mineralization rates of all four soils freezing at −25°C was significantly higher than those freezing at −5°C, regardless of low or high soil water content (P < 0.05). Soils with higher water content usually had high net N mineralization rate, regardless of the freeze temperature. The only exception was BKPF soil. The net N mineralization rate of BKPF soil undergoing cycles of freezing at −5°C and thawing at 5°C at low water content was higher than that at high water content, due mainly to the higher rate of nitrification at low water content (Table 3).

Table 3 Ammonification, nitrification and mineralization rates of vegetation soils over 15 FT cycles. WL Low soil water content treatment, WH high soil water content treatment

4 Discussion

Microbial activity and soil N cycling decrease when the temperature is low or below zero (Dalias et al. 2002). A decrease in microbial biomass carbon is found after several cycles of freezing-thawing but there is no obvious change in the content of microbial biomass N (Dalias et al. 2002). The activity of extracellular enzymes remaining at −20°C also indicates that soil microbes can survive several cycles of freezing-thawing of soil (Larsen et al. 2002). Nevertheless, previous studies on the effect of freezing-thawing cycling on soil organic N mineralization focused mostly on the effects of the duration and rate of freezing-thawing cycles and freeze temperature. These latter experiments were generally conducted by setting the soil below 0°C, and then setting the soil above 0°C for a period of time typically below 15 times of cycling (Henry 2007); in addition, the influence of soil water content on the mineralization of soil organic N in the process of freezing-thawing has not been well studied.

In the present study, we investigated the effects of freeze temperature, and freezing-thawing frequency as well as soil water content on mineralization of soil organic N in four vegetation soils of Changbai Mountain (Table 2). After 15 cycles of freezing-thawing, soil TIN content increased 1.67–26.77 times (Fig. 2), suggesting that soil N mineralization proceeds from late autumn to early spring when freezing-thawing cycling occurs in the four types of vegetation soils of Changbai Mountain. Freezing-thawing increased soil TIN content, similar to the effect of alternation of wetting and drying, and fumigating with chloroform. These treatments could potentially kill soil microbes and cause the release of nutrients from microbial cells (Deluca et al. 1992). When soils are thawed, soil microbes surviving the freeze period would be even more active, with rich substrates from dead microbes promoting the release of soil inorganic N (Neilsen et al. 2001; Tierney et al. 2001).

The results from our study show that soil TIN content increases while the net mineralization rate decreases with increasing freezing-thawing frequency. The highest net mineralization rate observed after the first cycle of freezing-thawing indicated that freezing-thawing could enhance the transformation of soil N in a very short period, but the effect became weaker with increasing freezing-thawing frequency. Since the laboratory incubation did not consider the roles of plant absorption and leaching in nutrient accumulation, the products of soil N mineralization accumulated likely inhibit mineralization (Amador et al. 2005). Moreover, soil frost kills soil microbes and freezing-thawing cycling likely would kill even more soil microbes (Sulkava and Huhta 2003). The number and activity of microbes decreases after several cycles of freezing-thawing (Herrmann and Witter 2002). So freezing-thawing cycling causes a flush of microbial C and N during the first cycle but, after repeated cycles, the ability of microbial communities to decompose soil organic matter decreases (Koponen and Martikainen 2004).

The results of this study showed that soils had a higher content of extracted inorganic N after freezing at the lower temperature. Our results are consistent with the results of Elliott and Henry (2009). Lower freeze temperature had a more intensive effect on soil microbes, including the death of more soil microbes (Sulkava and Huhta 2003). Organic matter from non-microbial origin also contributed to the increase of soil mineral N under the lower temperature freeze condition (Freppaz et al. 2007). Freezing-thawing cycles might cause the release of NH ⁺₄ -N from the soil inorganic and organic colloids that would have been previously unavailable, and the lower freeze temperature would lead to worse damage to soil aggregates (Kvarno and Oygarden 2006).

The soil water content affected mineralization of soil organic N significantly during the process of freezing-thawing of these four forest soils (Table 2). For each of the four forest soils, soil TIN content was generally higher at high water content than that at low water content (Fig. 2). After 15 cycles of freezing-thawing, the mineralization rates in soil at low and high water contents became similar (Fig. 3). The smaller soil pore space in the soil at higher water content would cause a more destructive effect on soil aggregates and soil microbes of ice crystals formed upon freezing (Sulkava and Huhta 2003; Kvarno and Oygarden 2006), leading to the enhanced release of inorganic N at thawing. Nevertheless, with increasing freezing-thawing cycling, the residual microbes adapted gradually to the low temperature conditions (Larsen et al. 2002) and soil aggregates became increasingly stable (Edwards 1991), resulting in minimized differences in soil mineralization rates between the low and high soil water content treatments after 15 times of freezing-thawing.

The four vegetation soils in Chanbai Mountain region undergo freezing-thawing cycles in spring and autumn under similar conditions of freeze temperature, frequency of freezing-thawing cycle (Fig. 1) and precipitation (Fig. 4). However, the effect of freezing-thawing cycling on soil N mineralization in the two seasons will differ, due primarily to the higher soil water content in the spring due to snowmelt. Based on the results from our study, the spring freezing-thawing cycling is expected to be more effective at increasing soil N mineralization to favor the growth of plants in Changbai Mountain ecosystems, but also increase the risk of N loss through surface runoff from spring snowmelt.