The rewetting of dry soils, or thawing of frozen soils, represents an abrupt step change in soil biophysical conditions, with implications for biogeochemical cycling. From an organismal perspective, soil rewetting and thawing are analogous because both processes increase the availability of soil water, rehydrate cells, and mobilize nutrients. Both processes are also relatively transient, with the duration of individual rewetting and thawing events varying depending on local climatic conditions, topography, drainage, vegetation type, and soil thermal properties (Balser and Firestone, 2005; Vargas et al., 2010b). The sudden flush of water and nutrients that occurs after rewetting and thawing precipitates major changes in plant and microbial activity, with organisms shifting rapidly from dormant or senescent states to active ones (Kieft et al., 1987; Schimel and Clein, 1996).
It is important to understand the change in fluxes of biogenic gases (i.e., CO2, CH4, N2O, NO and NH3) following rewetting and thawing events, as these biogenic gases are either by-products or end-products of soil-related microbial processes involved in C and N dynamics in soils. These gases also play crucial roles in atmospheric chemistry and radiative forcing as greenhouse gases (GHG). Furthermore, global climate models predict that future climatic change is likely to alter the frequency and intensity of drying-rewetting events and thawing of frozen soils (Meehl et al., 2006; Sheffield and Wood, 2008; Sinha and Cherkauer, 2010). The frequency and intensity of soil frost (i.e., annual soil freezing days and freeze-thaw cycles) is also likely to be modified since warming could lead to a reduction in the thickness of the insulating snowpack and thus colder winter soil temperatures (Henry, 2008; Gu et al., 2008). Thus, it is important to understand how soil rewetting and thawing influences biogenic gas fluxes, given the potential for GHGs such as CO2, CH4 and N2O to serve as either positive or negative feedbacks to future climate change.
While abrupt increases in soil CO2, N2O, NH3 and NO fluxes following rewetting are commonly observed in various agricultural lands and natural lands (Priemé and Christensen, 2001; Saetre and Stark, 2005), rewetting can either increase (Moore, 1998; Knorr et al., 2008) or inhibit (Kessavalou et al., 1998; Teh et al., 2005) CH4 oxidation. Similarly, increases in CO2, CH4 and N2O fluxes following soil thawing have been shown to affect total annual gas budgets (Röver et al., 1998; Papen and Butterbach-Bahl, 1999). Despite this growing number of studies, uncertainties in our understanding of the mechanisms and impacts on annual gas budgets continue to exist. These uncertainties are exacerbated by the coarse temporal sampling resolution in most flux measurements that do not capture the dynamic of the pulse (Groffman et al., 2006; Muhr et al., 2009), and unrealistic simulation of dry-wet and freeze-thaw events (Henry, 2007; Jentsch et al., 2007). These limitations are important for our understanding of soil GHG fluxes because even single pulse events have shown to substantially contribute to annual fluxes (Lee et al., 2004; Xu et al., 2004).
The growing number of studies on the individual importance of rewetting or thawing specifically for CO2 and N2O fluxes have been the focus of previous reviews (Henry, 2007; Matzner and Borken, 2008; Borken and Matzner, 2009; Groffman et al., 2009). This review differs by taking a comprehensive approach dealing with the effect of both rewetting and thawing on multiple biogenic gas fluxes (CO2, CH4, N2O, NO and NH3) and indentifying knowledge gaps and potential future research. The objectives of this manuscript are: 1) to summarize the effects of rewetting and thawing on multiple biogenic gas fluxes (CO2, CH4, N2O, NO and NH3) and highlight common patterns across studies; 2) discuss the underlying mechanisms and drivers of responses; 3) identify knowledge gaps and propose future research questions.
Tuesday, March 29, 2011
2. Methodology
2. 1. Data collection
Data on changes in gas fluxes of CO2, CH4, N2O, NO and NH3 following rewetting and thawing were acquired by searching existing refereed literature published between 1950 and 2010 using Web of Science and Google Scholar with search terms such as “rewetting”, “thawing”, “peak flux”, “peak emission” and name of gases. Field observations of rewetting of dry soils include events caused by natural rainfall, simulated rainfall in natural ecosystems, and irrigation in agricultural lands. Similarly, thawing of frozen soils include field observations of natural thawing, simulated freezing-thawing events (i.e., thawing of simulated frozen by snow removal), and thawing of seasonal ice in temperate and high latitude regions. We included both field and laboratory studies, but did not include the longer-term effects of changing active layer depths in this review, as changes in gas fluxes in response to permafrost thaw is affected by both changing soil and plant successional processes (Turetsky et al., 2007). The resulting data set comprised 222 field and laboratory experiments focused on rewetting of dry soils, and 116 field studies and laboratory experiments focused on thawing of frozen soils.
Data on changes in gas fluxes of CO2, CH4, N2O, NO and NH3 following rewetting and thawing were acquired by searching existing refereed literature published between 1950 and 2010 using Web of Science and Google Scholar with search terms such as “rewetting”, “thawing”, “peak flux”, “peak emission” and name of gases. Field observations of rewetting of dry soils include events caused by natural rainfall, simulated rainfall in natural ecosystems, and irrigation in agricultural lands. Similarly, thawing of frozen soils include field observations of natural thawing, simulated freezing-thawing events (i.e., thawing of simulated frozen by snow removal), and thawing of seasonal ice in temperate and high latitude regions. We included both field and laboratory studies, but did not include the longer-term effects of changing active layer depths in this review, as changes in gas fluxes in response to permafrost thaw is affected by both changing soil and plant successional processes (Turetsky et al., 2007). The resulting data set comprised 222 field and laboratory experiments focused on rewetting of dry soils, and 116 field studies and laboratory experiments focused on thawing of frozen soils.
3. A review of the effect of rewetting of dry soils on biogenic gas fluxes
Here we discuss, for each biogenic gas: a) how rewetting events influence gas fluxes in multiple ecosystems and in experimental settings; and b) the likely mechanisms and environmental controls underlying the observed patterns.
3. 1. Carbon dioxide flux from rewetting
3. 2. Methane flux from rewetting
3. 3. Nitrous oxide flux from rewetting
3. 4. Nitric oxide flux from rewetting
3. 5. Ammonia flux from rewetting
3. 1. Carbon dioxide flux from rewetting
General patterns of responses
Soil surface CO2 flux (RS) provides an integrated result of biological CO2 production throughout the soil column, changes in soil CO2 diffusivity in the soil profile, and in some areas geological processes. Soil CO2 is produced primarily by heterotrophic activity (i.e., decomposer organisms) and by autotrophic activity (i.e., living roots and mycorrhizae), where most of the produced CO2 is released to the atmosphere (Raich and Schlesinger, 1992; Schlesinger and Andrews, 2000).
Increase in RS following rewetting of dry soils have been reported in multiple terrestrial ecosystems and various land-use types, including pulses observed in cropland (Kessavalou et al., 1998), grazing pasture (Xu and Baldocchi, 2004), forest (Kim et al., 2010b), grassland (Joos et al., 2010), savannas (Castaldi et al., 2010) and desert (Sponseller and Fisher, 2008). Incubation experiments have yielded similar patterns, showing RS increases in soils from cropland (Beare et al., 2009), grazing pasture (Wu et al., 2010b), forest (Fierer and Schimel, 2003), grassland (Xiang et al., 2008), peatland (Goldhammer and Blodau, 2008) and desert (Sponseller and Fisher, 2008) ecosystems. For example, in an upper Sonoran Desert ecosystem, RS increased up to 30-fold immediately following experimental rewetting, and returned to background levels within 48 h (Sponseller, 2007). In soil moisture manipulations in a Norway spruce plantation, drought and rewetting treatments increased the annual CO2 flux by 51% (Borken et al., 1999). Lee et al. (2004) estimated that the peak RS flux in a single intensive storm amounted to a loss of 0.18 t C ha-1 to the atmosphere, or 5 − 10% of the annual net ecosystem production in a mid-latitude forest. These studies have reported responses ranging from short-term (ca. 6 − 24 h) up to 30 d responses (Table 1, Fig. 4), and relative RS changes ranging from 40% to >10,000% increases (Table 1, Fig. 2). Together, these studies support that hypothesis that rewetting a variety of soil types can have substantial affects on the C balance of terrestrial ecosystems (Borken et al., 1999; Lee et al., 2004; Xu et al., 2004). However, we caution that most of these studies lack the high temporal sampling resolution necessary to capture the full dynamic of the pulse (Groffman et al., 2006; Muhr et al., 2009, Vargas et al 2011).
It is important to recognize that RS could be suppressed during or after rainfall as previous studies have reported: 1) large (10-fold) decreases during light rainfall in arable soils (Rochette et al., 1991); 2) sharp RS decreases in no-tillage agricultural fields (Ball et al., 1999); and 3) reduced CO2 grassland fluxes with artificially increased variability of rainfall in mesic grasslands (Knapp et al., 2002). Possible explanations for these reduced RS rates are: 1) increased water in the soil pore space reduce soil CO2 diffusivity rates (Rochette et al. 1991; Šimůnek and Suarez, 1993); 2) soil CO2 may dissolve into the infiltrating water (Johnson et al., 2008); and 3) the restriction of the soil macro-porosity by the rainfall reduces soil air-filled pore space enhances anaerobiosis and reduces aerobic respiration (Linn and Doran 1984; Ball et al. 1999; Davidson et al., 2000). In the following sections we focus on the positive impact of rewetting on RS.
Mechanisms and drivers
Two broad mechanisms responsible for increased RS following rewetting have been commonly hypothesised: 1) enhanced microbial metabolism by substance supply, and 2) changes in physical protection of organic matter. First, microbial metabolism can be enhanced by the availability of accumulated substrates during soil drying periods. A large proportion of microorganisms, fine roots and mycorrhizae die during drought (Clein and Schimel, 1994; Teepe et al., 2001); these dead cells tend to have low C:N ratios and could rapidly decompose during rewetting (Kieft et al., 1987; Van Gestel et al., 1993). Microorganisms accumulate high concentration of solutes (osmolytes) to retain water inside the cell during drought conditions (Harris, 1981), which rapidly decompose on rewetting (Fiere and Schimel, 2003; Schimel et al., 2007). Root exudates from reviving plants could thus significantly affect soil surface flux (Crow and Wieder, 2005; Curiel Yuste et al., 2007). Second, rewetting could disrupt soil aggregates, exposing physically protected organic matter and increase the accessibility of substrate that can be rapidly mineralized (Groffman and Tiedje, 1988; Appel, 1998). Other physical mechanisms that can influence gas flux include infiltration, reduced diffusivity and gas displacement in the soil. For example, the infiltration of rainwater may displace CO2 that accumulates in soil pore spaces during dry periods (Huxman et al., 2004).
The magnitude of RS increases following rewetting depends on: 1) the size of soil organic pool; 2) the quality of organic matter, determined by its age, origin, and extent to which these substrates are protected from microbial attack by adsorption to clay surfaces and inclusion in micro-aggregates; and 3) the properties of soil biota (Van Gestel et al., 1993). Soil moisture conditions before rewetting also influence the response (Orchard and Cook, 1983; Cable et al., 2008), as can the length and severity of drought periods (Unger et al., 2010), and rain pulse size (Sponseller, 2007; Chen et al., 2009). Based on our literature review, we identified the existence of a threshold in soil moisture at 12−20% gravimetric moisture content, below which a substantial increases in RS after rewetting typically is observed (Davidson et al., 1998; Xu and Qi, 2001; Rey et al., 2002; Yuste et al., 2003; Dilustro et al., 2005; Cable et al., 2008; Chou et al., 2008; Kim et al., 2010b; Misson et al., 2010).
The effects of rewetting may decline with successive drying and rewetting cycles, possibly as a result of a limited pool of labile substrates that have built up over time or during the dry season (Schimel and Mikan, 2005; Goldberg et al., 2008). Importantly, Fernández et al. (2006) suggested that substrate availability, rather than soil moisture, influenced the duration of the CO2 pulse in response to rain events, while Vargas et al (2010b) noted that RS pulses may be driven not only by labile substrate availability, but also plant photosynthesis rates following the rain event. In addition, management practice (mowing or tillage) (Steenwerth et al., 2010), vegetation type (Shi et al., 2011) and high soil temperatures (Jager and Bruins, 1975; Borken et al., 1999) could influence the magnitude of the response of soil CO2 flux following rewetting of dry soils.
Soil surface CO2 flux (RS) provides an integrated result of biological CO2 production throughout the soil column, changes in soil CO2 diffusivity in the soil profile, and in some areas geological processes. Soil CO2 is produced primarily by heterotrophic activity (i.e., decomposer organisms) and by autotrophic activity (i.e., living roots and mycorrhizae), where most of the produced CO2 is released to the atmosphere (Raich and Schlesinger, 1992; Schlesinger and Andrews, 2000).
Increase in RS following rewetting of dry soils have been reported in multiple terrestrial ecosystems and various land-use types, including pulses observed in cropland (Kessavalou et al., 1998), grazing pasture (Xu and Baldocchi, 2004), forest (Kim et al., 2010b), grassland (Joos et al., 2010), savannas (Castaldi et al., 2010) and desert (Sponseller and Fisher, 2008). Incubation experiments have yielded similar patterns, showing RS increases in soils from cropland (Beare et al., 2009), grazing pasture (Wu et al., 2010b), forest (Fierer and Schimel, 2003), grassland (Xiang et al., 2008), peatland (Goldhammer and Blodau, 2008) and desert (Sponseller and Fisher, 2008) ecosystems. For example, in an upper Sonoran Desert ecosystem, RS increased up to 30-fold immediately following experimental rewetting, and returned to background levels within 48 h (Sponseller, 2007). In soil moisture manipulations in a Norway spruce plantation, drought and rewetting treatments increased the annual CO2 flux by 51% (Borken et al., 1999). Lee et al. (2004) estimated that the peak RS flux in a single intensive storm amounted to a loss of 0.18 t C ha-1 to the atmosphere, or 5 − 10% of the annual net ecosystem production in a mid-latitude forest. These studies have reported responses ranging from short-term (ca. 6 − 24 h) up to 30 d responses (Table 1, Fig. 4), and relative RS changes ranging from 40% to >10,000% increases (Table 1, Fig. 2). Together, these studies support that hypothesis that rewetting a variety of soil types can have substantial affects on the C balance of terrestrial ecosystems (Borken et al., 1999; Lee et al., 2004; Xu et al., 2004). However, we caution that most of these studies lack the high temporal sampling resolution necessary to capture the full dynamic of the pulse (Groffman et al., 2006; Muhr et al., 2009, Vargas et al 2011).
It is important to recognize that RS could be suppressed during or after rainfall as previous studies have reported: 1) large (10-fold) decreases during light rainfall in arable soils (Rochette et al., 1991); 2) sharp RS decreases in no-tillage agricultural fields (Ball et al., 1999); and 3) reduced CO2 grassland fluxes with artificially increased variability of rainfall in mesic grasslands (Knapp et al., 2002). Possible explanations for these reduced RS rates are: 1) increased water in the soil pore space reduce soil CO2 diffusivity rates (Rochette et al. 1991; Šimůnek and Suarez, 1993); 2) soil CO2 may dissolve into the infiltrating water (Johnson et al., 2008); and 3) the restriction of the soil macro-porosity by the rainfall reduces soil air-filled pore space enhances anaerobiosis and reduces aerobic respiration (Linn and Doran 1984; Ball et al. 1999; Davidson et al., 2000). In the following sections we focus on the positive impact of rewetting on RS.
Mechanisms and drivers
Two broad mechanisms responsible for increased RS following rewetting have been commonly hypothesised: 1) enhanced microbial metabolism by substance supply, and 2) changes in physical protection of organic matter. First, microbial metabolism can be enhanced by the availability of accumulated substrates during soil drying periods. A large proportion of microorganisms, fine roots and mycorrhizae die during drought (Clein and Schimel, 1994; Teepe et al., 2001); these dead cells tend to have low C:N ratios and could rapidly decompose during rewetting (Kieft et al., 1987; Van Gestel et al., 1993). Microorganisms accumulate high concentration of solutes (osmolytes) to retain water inside the cell during drought conditions (Harris, 1981), which rapidly decompose on rewetting (Fiere and Schimel, 2003; Schimel et al., 2007). Root exudates from reviving plants could thus significantly affect soil surface flux (Crow and Wieder, 2005; Curiel Yuste et al., 2007). Second, rewetting could disrupt soil aggregates, exposing physically protected organic matter and increase the accessibility of substrate that can be rapidly mineralized (Groffman and Tiedje, 1988; Appel, 1998). Other physical mechanisms that can influence gas flux include infiltration, reduced diffusivity and gas displacement in the soil. For example, the infiltration of rainwater may displace CO2 that accumulates in soil pore spaces during dry periods (Huxman et al., 2004).
The magnitude of RS increases following rewetting depends on: 1) the size of soil organic pool; 2) the quality of organic matter, determined by its age, origin, and extent to which these substrates are protected from microbial attack by adsorption to clay surfaces and inclusion in micro-aggregates; and 3) the properties of soil biota (Van Gestel et al., 1993). Soil moisture conditions before rewetting also influence the response (Orchard and Cook, 1983; Cable et al., 2008), as can the length and severity of drought periods (Unger et al., 2010), and rain pulse size (Sponseller, 2007; Chen et al., 2009). Based on our literature review, we identified the existence of a threshold in soil moisture at 12−20% gravimetric moisture content, below which a substantial increases in RS after rewetting typically is observed (Davidson et al., 1998; Xu and Qi, 2001; Rey et al., 2002; Yuste et al., 2003; Dilustro et al., 2005; Cable et al., 2008; Chou et al., 2008; Kim et al., 2010b; Misson et al., 2010).
The effects of rewetting may decline with successive drying and rewetting cycles, possibly as a result of a limited pool of labile substrates that have built up over time or during the dry season (Schimel and Mikan, 2005; Goldberg et al., 2008). Importantly, Fernández et al. (2006) suggested that substrate availability, rather than soil moisture, influenced the duration of the CO2 pulse in response to rain events, while Vargas et al (2010b) noted that RS pulses may be driven not only by labile substrate availability, but also plant photosynthesis rates following the rain event. In addition, management practice (mowing or tillage) (Steenwerth et al., 2010), vegetation type (Shi et al., 2011) and high soil temperatures (Jager and Bruins, 1975; Borken et al., 1999) could influence the magnitude of the response of soil CO2 flux following rewetting of dry soils.
3. 2. Methane flux from rewetting
General patterns of responses
Net CH4 flux is the result of the balance between the two off setting processes of methanogenesis (microbial production under anaerobic conditions) and methanotrophy (microbial consumption) (Dutaur and Verchot, 2007). Methanogenesis occurs via the anaerobic degradation of organic matter by methanogenic archaea within the archaeal phylum Euryarchaeota (Thauer, 1988). Methanotrophy occurs by methanotrophs metabolizing CH4 as their source of carbon and energy (Hanson and Hanson, 1996). In anoxic soils, emergent vegetation also influences CH4 flux to the atmosphere, as plants enable oxygen transport to the rhizosphere, transport through aerenchymateous tissue, and the production of labile substrates via root exudation (Joabsson et al. 1999).
The reported effects of rewetting on CH4 fluxes are variable. Rewetting reduced CH4 consumption or increased CH4 production in arable land (Syamsul Arif et al., 1996; Kessavalou et al., 1998; Hergoualc’h et al., 2008), peatland (Kettunen et al., 1996; Blodau and Moore, 2003; Dinsmore et al., 2009) and tropical forest (Silver et al., 1999). In a wheat-fallow cropping system, CH4 consumption declined by about 60% for 3 to 14 d after rewetting (Kessavalou et al., 1998). In peatland, a pulse of CH4 was observed after water table drawdown (Moore and Knowles, 1990; Shurpali et al., 1993) and significant pulses of CH4 fluxes were produced with both drainage (700 μg m−2 h−1 above the pre-change mean) and rewetting (over 160 μg m−2 h−1 above the value of prior to rewetting) within 1 or 2 days in a mesocosm study (Dinsmore et al., 2009). In contrast, other studies have reported that rewetting increased CH4 consumption, or reduced CH4 production, both in the field (Davidson et al., 2004; Borken et al., 2006; Davidson et al., 2008; Fiedler et al., 2008) and laboratory (Czepiel et al., 1995; West and Schmidt, 1998). In incubation experiments with alpine soil, CH4 oxidation increased significantly from 11 pmol CH4 (g dry weight)−1 h−1 to -29.5 − -67.0 pmol CH4 (g dry weight)−1 h−1 9 days after rewetting (West and Schmidt, 1998). Enhanced CH4 oxidation was promoted after rewetting for days to weeks in peatland (Öquist and Sundh, 1998; Kettunen et al., 1999; Goldhammer and Blodau, 2008) and rice field (Ratering and Conrad, 1998). However, in an in situ water table drawdown experiment, CH4 production declined in hummocks but stayed constant in hollows relative to control plots, suggesting a strong role of plant-mediated release of CH4 in some peatland microforms (Strack and Waddington, 2007). In summary, studies report a large uncertainty in CH4 responses after rewetting and there are much smaller responses in magnitude but fewer observations compared to other gases (Table 1, Fig. 2).
Mechanisms and drivers
In general, CH4 production rates are controlled by the availability of suitable substrates, alternative electron acceptors for competing redox reactions (i.e., sulfate reduction), the nutritional status of the ecosystem (i.e., bog versus fen), water table position or soil moisture content, temperature, and soil salinity (Thauer, 1988; Hanson and Hanson, 1996; Dutaur and Verchot, 2007).
The mechanisms and drivers underlying changes in CH4 flux following rewetting are complex because they involve the response of both methanogenesis and methanotrophy to changes in soil environment, particularly soil moisture, and availability of electron donors and acceptor that determine the redox status of soil. Rewetting can increase the availability of water-soluble C substrates (Zsolnay and Görlitz, 1994; Stark and Firestone, 1995), which soil methanotrophs utilize as an electron source (Whittenbury et al., 1970). In unfrozen soils, there was no correlation between soil temperature and CH4 consumption, suggesting strong substrate limitation on methanotrophs (Borken et al., 2006). Borken et al. (2006) also found that methanotrophs were stressed when water contents were below 0.15 g cm−3 (in the A horizon), thus rewetting can alleviate osmotic stress and promote the growth and activity of soil methanotrophs (Schnell and King, 1996; West and Schmidt, 1998). While several studies have shown that experimental drought increased CH4 consumption rates (cf. Borken et al., 2006, Davidson et al., 2008), Fiedler et al. (2008) found no evidence of increased methanotrophy in response to natural drought in forest soils. Methanotrophs responded quickly to water table manipulations in peat soil (Blodau and Moore, 2003). Rewetting also can inhibit methanotrophic activity in more poorly drained soils, for example if oxygen diffusion becomes limiting (Striegl, 1993). Because methanogenesis requires anaerobic soil conditions, drought typically suppresses CH4 production, while rewetting increases it. Methanogenic populations require some time to re-establish after rewetting (Fetzer et al., 1993).
In addition to environmental controls, both methanotrophy and methanogenesis are sensitive to interactions and competition with other microbial redox processes. Drying and rewetting of soils can increase SO4 pools through remineralization of organic sulfate and/or reoxidation of iron sulfides. This can stimulate sulfate reduction and effectively suppress methanogenesis (c.f., Blodau and Moore, 2003). In thick organic soils, this is more likely to occur in surface layers that experience fluctuating water tables rather than more saturated deeper peat layers (Goldhammer and Blodau, 2008). Overall, the mechanisms and drivers responsible for the various response of CH4 to thawing have not been clearly explored to our knowledge and further research is needed to identify the mechanisms controlling the response after rewetting at multiple ecosystems.
Net CH4 flux is the result of the balance between the two off setting processes of methanogenesis (microbial production under anaerobic conditions) and methanotrophy (microbial consumption) (Dutaur and Verchot, 2007). Methanogenesis occurs via the anaerobic degradation of organic matter by methanogenic archaea within the archaeal phylum Euryarchaeota (Thauer, 1988). Methanotrophy occurs by methanotrophs metabolizing CH4 as their source of carbon and energy (Hanson and Hanson, 1996). In anoxic soils, emergent vegetation also influences CH4 flux to the atmosphere, as plants enable oxygen transport to the rhizosphere, transport through aerenchymateous tissue, and the production of labile substrates via root exudation (Joabsson et al. 1999).
The reported effects of rewetting on CH4 fluxes are variable. Rewetting reduced CH4 consumption or increased CH4 production in arable land (Syamsul Arif et al., 1996; Kessavalou et al., 1998; Hergoualc’h et al., 2008), peatland (Kettunen et al., 1996; Blodau and Moore, 2003; Dinsmore et al., 2009) and tropical forest (Silver et al., 1999). In a wheat-fallow cropping system, CH4 consumption declined by about 60% for 3 to 14 d after rewetting (Kessavalou et al., 1998). In peatland, a pulse of CH4 was observed after water table drawdown (Moore and Knowles, 1990; Shurpali et al., 1993) and significant pulses of CH4 fluxes were produced with both drainage (700 μg m−2 h−1 above the pre-change mean) and rewetting (over 160 μg m−2 h−1 above the value of prior to rewetting) within 1 or 2 days in a mesocosm study (Dinsmore et al., 2009). In contrast, other studies have reported that rewetting increased CH4 consumption, or reduced CH4 production, both in the field (Davidson et al., 2004; Borken et al., 2006; Davidson et al., 2008; Fiedler et al., 2008) and laboratory (Czepiel et al., 1995; West and Schmidt, 1998). In incubation experiments with alpine soil, CH4 oxidation increased significantly from 11 pmol CH4 (g dry weight)−1 h−1 to -29.5 − -67.0 pmol CH4 (g dry weight)−1 h−1 9 days after rewetting (West and Schmidt, 1998). Enhanced CH4 oxidation was promoted after rewetting for days to weeks in peatland (Öquist and Sundh, 1998; Kettunen et al., 1999; Goldhammer and Blodau, 2008) and rice field (Ratering and Conrad, 1998). However, in an in situ water table drawdown experiment, CH4 production declined in hummocks but stayed constant in hollows relative to control plots, suggesting a strong role of plant-mediated release of CH4 in some peatland microforms (Strack and Waddington, 2007). In summary, studies report a large uncertainty in CH4 responses after rewetting and there are much smaller responses in magnitude but fewer observations compared to other gases (Table 1, Fig. 2).
Mechanisms and drivers
In general, CH4 production rates are controlled by the availability of suitable substrates, alternative electron acceptors for competing redox reactions (i.e., sulfate reduction), the nutritional status of the ecosystem (i.e., bog versus fen), water table position or soil moisture content, temperature, and soil salinity (Thauer, 1988; Hanson and Hanson, 1996; Dutaur and Verchot, 2007).
The mechanisms and drivers underlying changes in CH4 flux following rewetting are complex because they involve the response of both methanogenesis and methanotrophy to changes in soil environment, particularly soil moisture, and availability of electron donors and acceptor that determine the redox status of soil. Rewetting can increase the availability of water-soluble C substrates (Zsolnay and Görlitz, 1994; Stark and Firestone, 1995), which soil methanotrophs utilize as an electron source (Whittenbury et al., 1970). In unfrozen soils, there was no correlation between soil temperature and CH4 consumption, suggesting strong substrate limitation on methanotrophs (Borken et al., 2006). Borken et al. (2006) also found that methanotrophs were stressed when water contents were below 0.15 g cm−3 (in the A horizon), thus rewetting can alleviate osmotic stress and promote the growth and activity of soil methanotrophs (Schnell and King, 1996; West and Schmidt, 1998). While several studies have shown that experimental drought increased CH4 consumption rates (cf. Borken et al., 2006, Davidson et al., 2008), Fiedler et al. (2008) found no evidence of increased methanotrophy in response to natural drought in forest soils. Methanotrophs responded quickly to water table manipulations in peat soil (Blodau and Moore, 2003). Rewetting also can inhibit methanotrophic activity in more poorly drained soils, for example if oxygen diffusion becomes limiting (Striegl, 1993). Because methanogenesis requires anaerobic soil conditions, drought typically suppresses CH4 production, while rewetting increases it. Methanogenic populations require some time to re-establish after rewetting (Fetzer et al., 1993).
In addition to environmental controls, both methanotrophy and methanogenesis are sensitive to interactions and competition with other microbial redox processes. Drying and rewetting of soils can increase SO4 pools through remineralization of organic sulfate and/or reoxidation of iron sulfides. This can stimulate sulfate reduction and effectively suppress methanogenesis (c.f., Blodau and Moore, 2003). In thick organic soils, this is more likely to occur in surface layers that experience fluctuating water tables rather than more saturated deeper peat layers (Goldhammer and Blodau, 2008). Overall, the mechanisms and drivers responsible for the various response of CH4 to thawing have not been clearly explored to our knowledge and further research is needed to identify the mechanisms controlling the response after rewetting at multiple ecosystems.
3. 3. Nitrous oxide flux from rewetting
General patterns of responses
Three main processes produce nitrous oxide in soils: 1) nitrification, the stepwise oxidation of NH3 to nitrite (NO2−) and to nitrate (NO3−) (Kowalchuk and Stephen, 2001); 2) denitrification, the stepwise reduction of NO3− to NO2−, NO, N2O and ultimately N2, where facultative anaerobe bacteria use NO3− as an electron acceptor in the respiration of organic material under low oxygen (O2) conditions (Knowles, 1982); and 3) nitrifier denitrification, which is carried out by autotrophic NH3-oxidizing bacteria and the pathway whereby NH3 is oxidized to nitrite NO2−, followed by the reduction of NO2− to nitric oxide NO, N2O and molecular nitrogen (N2) (Wrage et al., 2001).
Field studies have observed increased soil N2O flux following wetting in cropland (Barton et al., 2008), grazing pasture (Kim et al., 2010a), forest (Butterbach-Bahl et al., 2004), grassland (Hao et al., 1988), savannah (Martin et al., 2003) and fen (Goldberg et al., 2010a). Laboratory incubation experiments with cropland soil (Beare et al., 2009), forest soil (Dick et al., 2001), grassland soil (Yao et al., 2010) and peatland soil (Dinsmore et al., 2009) have yielded similar results of increased N2O flux after rewetting. In tropical soils in Costa Rica, N2O flux pulses began within 30 min, peaking no later than 8 h after rewetting and 25 g N2O−N ha-1 was emitted for three simulated rain events over a 22-day period (control emitted 14 g N2O−N ha-1) and one episodic N2O production event driven by one moderate rain accounted for less than 15% to more than 90% of the total weekly production (Nobre et al., 2001). These studies have observed a short-term (ca., 12 h) up to 15 d N2O response following rewetting (Table 2), and an increase of N2O flux up to 80,000% with respect to the background conditions (Table 2, Fig. 2). Importantly, even a single wetting event can affect annual N2O flux (from 2% up to 50%) (Nobre et al., 2001; Barton et al., 2008; Goldberg et al., 2010a).
Mechanisms and drivers
The mechanisms responsible for increased N2O flux following rewetting have been commonly hypothesised as belonging to two categories: 1) enhanced microbial metabolism by substance supply, and 2) the physical mechanisms described above (§3.1). The relatively importance of processes responsible for N2O fluxes changes (i.e., nitrification, denitrification and nitrifier denitrification) is poorly understood, however, although several studies have found denitrification to be the most important (Groffman and Tiedje, 1988; Priemé and Christensen, 2001).
Magnitudes of increased N2O flux caused by wetting of dry soils vary depending on the labile N soil pool (Van Gestel et al., 1993; Schaeffer et al., 2003), soil texture (Appel, 1998; Austin et al., 2004), soil water content (Appel, 1998), size of the rewetting pulse (Ruser et al., 2006; Yanai et al., 2007), length of drought (van Haren et al., 2005), and soil compaction (Uchida et al., 2008; Beare et al., 2009). A significant relationship between the organic nitrogen extracted from dried soil samples and the magnitude of N2O flushes following soil drying-rewetting has been observed (Appel, 1998). Field and laboratory studies with arid and semiarid soils, fine-textured soils having higher water-holding capacity and labile C and N pools compared to coarse-textured soils showed greater flush of N2O flux following rewetting (Austin et al., 2004). In an incubation experiment with soils from potato field, the amount of increase in N2O flux following rewetting enhanced with the amount of water added (Ruser et al., 2006). Furthermore, in another experiment with soils from a field compaction trail, the production of N2O during the first 24 h following rewetting of dry soil was nearly 20 times higher in compacted than in uncompacted soil (Beare et al., 2009).
Three main processes produce nitrous oxide in soils: 1) nitrification, the stepwise oxidation of NH3 to nitrite (NO2−) and to nitrate (NO3−) (Kowalchuk and Stephen, 2001); 2) denitrification, the stepwise reduction of NO3− to NO2−, NO, N2O and ultimately N2, where facultative anaerobe bacteria use NO3− as an electron acceptor in the respiration of organic material under low oxygen (O2) conditions (Knowles, 1982); and 3) nitrifier denitrification, which is carried out by autotrophic NH3-oxidizing bacteria and the pathway whereby NH3 is oxidized to nitrite NO2−, followed by the reduction of NO2− to nitric oxide NO, N2O and molecular nitrogen (N2) (Wrage et al., 2001).
Field studies have observed increased soil N2O flux following wetting in cropland (Barton et al., 2008), grazing pasture (Kim et al., 2010a), forest (Butterbach-Bahl et al., 2004), grassland (Hao et al., 1988), savannah (Martin et al., 2003) and fen (Goldberg et al., 2010a). Laboratory incubation experiments with cropland soil (Beare et al., 2009), forest soil (Dick et al., 2001), grassland soil (Yao et al., 2010) and peatland soil (Dinsmore et al., 2009) have yielded similar results of increased N2O flux after rewetting. In tropical soils in Costa Rica, N2O flux pulses began within 30 min, peaking no later than 8 h after rewetting and 25 g N2O−N ha-1 was emitted for three simulated rain events over a 22-day period (control emitted 14 g N2O−N ha-1) and one episodic N2O production event driven by one moderate rain accounted for less than 15% to more than 90% of the total weekly production (Nobre et al., 2001). These studies have observed a short-term (ca., 12 h) up to 15 d N2O response following rewetting (Table 2), and an increase of N2O flux up to 80,000% with respect to the background conditions (Table 2, Fig. 2). Importantly, even a single wetting event can affect annual N2O flux (from 2% up to 50%) (Nobre et al., 2001; Barton et al., 2008; Goldberg et al., 2010a).
Mechanisms and drivers
The mechanisms responsible for increased N2O flux following rewetting have been commonly hypothesised as belonging to two categories: 1) enhanced microbial metabolism by substance supply, and 2) the physical mechanisms described above (§3.1). The relatively importance of processes responsible for N2O fluxes changes (i.e., nitrification, denitrification and nitrifier denitrification) is poorly understood, however, although several studies have found denitrification to be the most important (Groffman and Tiedje, 1988; Priemé and Christensen, 2001).
Magnitudes of increased N2O flux caused by wetting of dry soils vary depending on the labile N soil pool (Van Gestel et al., 1993; Schaeffer et al., 2003), soil texture (Appel, 1998; Austin et al., 2004), soil water content (Appel, 1998), size of the rewetting pulse (Ruser et al., 2006; Yanai et al., 2007), length of drought (van Haren et al., 2005), and soil compaction (Uchida et al., 2008; Beare et al., 2009). A significant relationship between the organic nitrogen extracted from dried soil samples and the magnitude of N2O flushes following soil drying-rewetting has been observed (Appel, 1998). Field and laboratory studies with arid and semiarid soils, fine-textured soils having higher water-holding capacity and labile C and N pools compared to coarse-textured soils showed greater flush of N2O flux following rewetting (Austin et al., 2004). In an incubation experiment with soils from potato field, the amount of increase in N2O flux following rewetting enhanced with the amount of water added (Ruser et al., 2006). Furthermore, in another experiment with soils from a field compaction trail, the production of N2O during the first 24 h following rewetting of dry soil was nearly 20 times higher in compacted than in uncompacted soil (Beare et al., 2009).
3. 4. Nitric oxide flux from rewetting
General patterns of responses
Nitric oxide can be produced from: 1) nitrification (Kowalchuk and Stephen, 2001); 2) denitrification (Knowles, 1982); and 3) nitrifier denitrification (Wrage et al., 2001) as described in §3.3. Increases in soil NO flux following rewetting have been reported in various terrestrial ecosystems including cropland (Guenzi et al., 1994), grazing pasture (Hutchinson and Brams, 1992), forest (Wu et al., 2010a), grassland (Hartley and Schlesinger, 2000), savanna (Martin et al., 2003), and desert (McCalley and Sparks, 2008). Laboratory incubations with grassland soil (Yao et al., 2010), grazing pasture soil (Hutchinson et al., 1993), forest soil (Dick et al., 2006) and desert soil (McCalley and Sparks, 2008) have reported similar results of increased NO flux after rewetting. NO rewetting studies have commonly reported short-term (ca., 1−3 d) response following rewetting (Table 2), and the rate of NO flux increase ranged from 40% to more than 800,000% (Table 2, Fig. 2). Some studies indicate that even a single rewetting event could substantially affect annual flux rates of NO (Davidson et al., 1991; Yienger and Levy, 1995; Kitzler et al., 2006), and rewetting events could be important for regional fluxes (Harris et al., 1996; Ghude et al., 2010).
Mechanisms and drivers
The mechanisms responsible for increased NO fluxes following rewetting have been commonly hypothesised as belonging to the two categories: 1) enhanced microbial metabolism by substance supply and 2) physical mechanisms described above (§3.1). Several studies found that nitrification is the dominant source of increased NO flux following wetting of dry soils (Davidson, 1992a; Davidson et al., 1993; Hutchinson et al., 1993). The magnitude of increased NO flux can be influenced by the duration and severity of antecedent dry periods (Butterbach-Bahl et al., 2004; McCalley and Sparks, 2008), change in soil moisture (Yienger and Levy, 1995) and temperature (Smart et al., 1999; McCalley and Sparks, 2008), vegetation type (Barger et al., 2005; McCalley and Sparks, 2008), soil type (Martin et al., 2003), microbial demand for N (Stark et al., 2002), frequency of wetting events (Davidson et al., 1991; Hartley and Schlesinger, 2000), previous disturbances (Levine et al., 1988; Poth et al., 1995), and agricultural management (Hutchinson and Brams, 1992). Interestingly, there are conflicting results concerning the magnitude of increased NO flux after rewetting, which were independent of both the size of rewetting pulse (Davidson, 1992b; Martin et al., 1998) and the periods of antecedent dry days (Martin et al., 1998). Also, other reports have suggested that lower amount of water addition result in higher NO pulses (Hutchinson et al., 1997; Dick et al., 2001). These conflicting results emphasize the uncertainty, and limitations, of predicting the magnitude of NO flux responses based on so few data.
Nitric oxide can be produced from: 1) nitrification (Kowalchuk and Stephen, 2001); 2) denitrification (Knowles, 1982); and 3) nitrifier denitrification (Wrage et al., 2001) as described in §3.3. Increases in soil NO flux following rewetting have been reported in various terrestrial ecosystems including cropland (Guenzi et al., 1994), grazing pasture (Hutchinson and Brams, 1992), forest (Wu et al., 2010a), grassland (Hartley and Schlesinger, 2000), savanna (Martin et al., 2003), and desert (McCalley and Sparks, 2008). Laboratory incubations with grassland soil (Yao et al., 2010), grazing pasture soil (Hutchinson et al., 1993), forest soil (Dick et al., 2006) and desert soil (McCalley and Sparks, 2008) have reported similar results of increased NO flux after rewetting. NO rewetting studies have commonly reported short-term (ca., 1−3 d) response following rewetting (Table 2), and the rate of NO flux increase ranged from 40% to more than 800,000% (Table 2, Fig. 2). Some studies indicate that even a single rewetting event could substantially affect annual flux rates of NO (Davidson et al., 1991; Yienger and Levy, 1995; Kitzler et al., 2006), and rewetting events could be important for regional fluxes (Harris et al., 1996; Ghude et al., 2010).
Mechanisms and drivers
The mechanisms responsible for increased NO fluxes following rewetting have been commonly hypothesised as belonging to the two categories: 1) enhanced microbial metabolism by substance supply and 2) physical mechanisms described above (§3.1). Several studies found that nitrification is the dominant source of increased NO flux following wetting of dry soils (Davidson, 1992a; Davidson et al., 1993; Hutchinson et al., 1993). The magnitude of increased NO flux can be influenced by the duration and severity of antecedent dry periods (Butterbach-Bahl et al., 2004; McCalley and Sparks, 2008), change in soil moisture (Yienger and Levy, 1995) and temperature (Smart et al., 1999; McCalley and Sparks, 2008), vegetation type (Barger et al., 2005; McCalley and Sparks, 2008), soil type (Martin et al., 2003), microbial demand for N (Stark et al., 2002), frequency of wetting events (Davidson et al., 1991; Hartley and Schlesinger, 2000), previous disturbances (Levine et al., 1988; Poth et al., 1995), and agricultural management (Hutchinson and Brams, 1992). Interestingly, there are conflicting results concerning the magnitude of increased NO flux after rewetting, which were independent of both the size of rewetting pulse (Davidson, 1992b; Martin et al., 1998) and the periods of antecedent dry days (Martin et al., 1998). Also, other reports have suggested that lower amount of water addition result in higher NO pulses (Hutchinson et al., 1997; Dick et al., 2001). These conflicting results emphasize the uncertainty, and limitations, of predicting the magnitude of NO flux responses based on so few data.
Subscribe to:
Posts (Atom)